status
stringclasses 1
value | repo_name
stringclasses 31
values | repo_url
stringclasses 31
values | issue_id
int64 1
104k
| title
stringlengths 4
233
| body
stringlengths 0
186k
⌀ | issue_url
stringlengths 38
56
| pull_url
stringlengths 37
54
| before_fix_sha
stringlengths 40
40
| after_fix_sha
stringlengths 40
40
| report_datetime
timestamp[us, tz=UTC] | language
stringclasses 5
values | commit_datetime
timestamp[us, tz=UTC] | updated_file
stringlengths 7
188
| chunk_content
stringlengths 1
1.03M
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,225 | [Feature][Alert] Add close alert method | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
Right now, we can set the alert when workflow executes failure.
In some cases, if the alert plugin is a third party service like [pagerduty](https://www.pagerduty.com), when we send an alert we will call the service to send an alert, and it may send 5 minutes unless we close the alert(We can close the alert by manual too).
But we may hope when the workflow success, the before alert can close automatically, so this issue is want to discuss add a new method closeAlert in alert plugin, we will provide a default method, the default method will do nothing, you can override this.
When the workflow execute success we will check if there exist a alert, and insert a close alert event, the AlertServer side will consume the event and execute the closeMethod.
### Use case
_No response_
### Related issues
_No response_
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11225 | https://github.com/apache/dolphinscheduler/pull/11221 | 01fc6c8a3a3aa68a45879ec5130250c99492a86b | 8e21c38c0041aa2689eb9d9bc703f67292c98e88 | 2022-08-01T03:59:56Z | java | 2022-08-01T05:04:27Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | f (!completeTaskMap.contansKey(despNodeTaskCode)) {
return DependResult.WAITING;
}
Integer depsTaskId = completeTaskMap.get(despNodeTaskCode);
ExecutonStatus depTaskState = taskInstanceMap.get(depsTaskId).getState();
f (depTaskState.typeIsPause() || depTaskState.typeIsCancel()) {
return DependResult.NON_EXEC;
}
f (taskNode.sBlockngTask()) {
contnue;
}
f (taskNode.sCondtonsTask()) {
contnue;
}
f (!dependTaskSuccess(depsNode, taskCode)) {
return DependResult.FAILED;
}
}
}
logger.nfo("taskCode: {} completeDependTaskLst: {}", taskCode, Arrays.toStrng(completeTaskMap.keySet().toArray()));
return DependResult.SUCCESS;
}
/**
* Ths functon s specally used to handle the dependency stuaton where the parent node s a prohbted node.
* When the parent node s a forbdden node, the dependency relatonshp should contnue to be traced
*
* @param taskCode taskCode
* @param ndrectDepCodeLst All ndrectly dependent nodes |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,225 | [Feature][Alert] Add close alert method | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
Right now, we can set the alert when workflow executes failure.
In some cases, if the alert plugin is a third party service like [pagerduty](https://www.pagerduty.com), when we send an alert we will call the service to send an alert, and it may send 5 minutes unless we close the alert(We can close the alert by manual too).
But we may hope when the workflow success, the before alert can close automatically, so this issue is want to discuss add a new method closeAlert in alert plugin, we will provide a default method, the default method will do nothing, you can override this.
When the workflow execute success we will check if there exist a alert, and insert a close alert event, the AlertServer side will consume the event and execute the closeMethod.
### Use case
_No response_
### Related issues
_No response_
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11225 | https://github.com/apache/dolphinscheduler/pull/11221 | 01fc6c8a3a3aa68a45879ec5130250c99492a86b | 8e21c38c0041aa2689eb9d9bc703f67292c98e88 | 2022-08-01T03:59:56Z | java | 2022-08-01T05:04:27Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | */
prvate vod setIndrectDepLst(Strng taskCode, Lst<Strng> ndrectDepCodeLst) {
TaskNode taskNode = dag.getNode(taskCode);
Lst<Strng> depCodeLst = taskNode.getDepLst();
for (Strng depsNode : depCodeLst) {
f (forbddenTaskMap.contansKey(Long.parseLong(depsNode))) {
setIndrectDepLst(depsNode, ndrectDepCodeLst);
} else {
ndrectDepCodeLst.add(depsNode);
}
}
}
/**
* depend node s completed, but here need check the condton task branch s the next node
*/
prvate boolean dependTaskSuccess(Strng dependNodeName, Strng nextNodeName) {
f (dag.getNode(dependNodeName).sCondtonsTask()) {
Lst<Strng> nextTaskLst = DagHelper.parseCondtonTask(dependNodeName, skpTaskNodeMap, dag, getCompleteTaskInstanceMap());
f (!nextTaskLst.contans(nextNodeName)) {
return false;
}
} else {
long taskCode = Long.parseLong(dependNodeName);
Integer taskInstanceId = completeTaskMap.get(taskCode);
ExecutonStatus depTaskState = taskInstanceMap.get(taskInstanceId).getState();
f (depTaskState.typeIsFalure()) {
return false;
}
} |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,225 | [Feature][Alert] Add close alert method | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
Right now, we can set the alert when workflow executes failure.
In some cases, if the alert plugin is a third party service like [pagerduty](https://www.pagerduty.com), when we send an alert we will call the service to send an alert, and it may send 5 minutes unless we close the alert(We can close the alert by manual too).
But we may hope when the workflow success, the before alert can close automatically, so this issue is want to discuss add a new method closeAlert in alert plugin, we will provide a default method, the default method will do nothing, you can override this.
When the workflow execute success we will check if there exist a alert, and insert a close alert event, the AlertServer side will consume the event and execute the closeMethod.
### Use case
_No response_
### Related issues
_No response_
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11225 | https://github.com/apache/dolphinscheduler/pull/11221 | 01fc6c8a3a3aa68a45879ec5130250c99492a86b | 8e21c38c0041aa2689eb9d9bc703f67292c98e88 | 2022-08-01T03:59:56Z | java | 2022-08-01T05:04:27Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | return true;
}
/**
* query task nstance by complete state
*
* @param state state
* @return task nstance lst
*/
prvate Lst<TaskInstance> getCompleteTaskByState(ExecutonStatus state) {
Lst<TaskInstance> resultLst = new ArrayLst<>();
for (Integer taskInstanceId : completeTaskMap.values()) {
TaskInstance taskInstance = taskInstanceMap.get(taskInstanceId);
f (taskInstance != null && taskInstance.getState() == state) {
resultLst.add(taskInstance);
}
}
return resultLst;
}
/**
* where there are ongong tasks
*
* @param state state
* @return ExecutonStatus
*/
prvate ExecutonStatus runnngState(ExecutonStatus state) {
f (state == ExecutonStatus.READY_STOP || state == ExecutonStatus.READY_PAUSE || state == ExecutonStatus.WAITING_THREAD || state == ExecutonStatus.READY_BLOCK ||
state == ExecutonStatus.DELAY_EXECUTION) {
return state;
} else { |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,225 | [Feature][Alert] Add close alert method | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
Right now, we can set the alert when workflow executes failure.
In some cases, if the alert plugin is a third party service like [pagerduty](https://www.pagerduty.com), when we send an alert we will call the service to send an alert, and it may send 5 minutes unless we close the alert(We can close the alert by manual too).
But we may hope when the workflow success, the before alert can close automatically, so this issue is want to discuss add a new method closeAlert in alert plugin, we will provide a default method, the default method will do nothing, you can override this.
When the workflow execute success we will check if there exist a alert, and insert a close alert event, the AlertServer side will consume the event and execute the closeMethod.
### Use case
_No response_
### Related issues
_No response_
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11225 | https://github.com/apache/dolphinscheduler/pull/11221 | 01fc6c8a3a3aa68a45879ec5130250c99492a86b | 8e21c38c0041aa2689eb9d9bc703f67292c98e88 | 2022-08-01T03:59:56Z | java | 2022-08-01T05:04:27Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | return ExecutonStatus.RUNNING_EXECUTION;
}
}
/**
* exsts falure task,contans submt falure、dependency falure,execute falure(retry after)
*
* @return Boolean whether has faled task
*/
prvate boolean hasFaledTask() {
f (ths.taskFaledSubmt) {
return true;
}
f (ths.errorTaskMap.sze() > 0) {
return true;
}
return ths.dependFaledTaskMap.sze() > 0;
}
/**
* process nstance falure
*
* @return Boolean whether process nstance faled
*/
prvate boolean processFaled() {
f (hasFaledTask()) {
logger.nfo("The current process has faled task, the current process faled");
f (processInstance.getFalureStrategy() == FalureStrategy.END) {
return true;
}
f (processInstance.getFalureStrategy() == FalureStrategy.CONTINUE) {
return readyToSubmtTaskQueue.sze() == 0 && actveTaskProcessorMaps.sze() == 0 && watToRetryTaskInstanceMap.sze() == 0; |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,225 | [Feature][Alert] Add close alert method | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
Right now, we can set the alert when workflow executes failure.
In some cases, if the alert plugin is a third party service like [pagerduty](https://www.pagerduty.com), when we send an alert we will call the service to send an alert, and it may send 5 minutes unless we close the alert(We can close the alert by manual too).
But we may hope when the workflow success, the before alert can close automatically, so this issue is want to discuss add a new method closeAlert in alert plugin, we will provide a default method, the default method will do nothing, you can override this.
When the workflow execute success we will check if there exist a alert, and insert a close alert event, the AlertServer side will consume the event and execute the closeMethod.
### Use case
_No response_
### Related issues
_No response_
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11225 | https://github.com/apache/dolphinscheduler/pull/11221 | 01fc6c8a3a3aa68a45879ec5130250c99492a86b | 8e21c38c0041aa2689eb9d9bc703f67292c98e88 | 2022-08-01T03:59:56Z | java | 2022-08-01T05:04:27Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | }
}
return false;
}
/**
* whether task for watng thread
*
* @return Boolean whether has watng thread task
*/
prvate boolean hasWatngThreadTask() {
Lst<TaskInstance> watngLst = getCompleteTaskByState(ExecutonStatus.WAITING_THREAD);
return CollectonUtls.sNotEmpty(watngLst);
}
/**
* prepare for pause
* 1，faled retry task n the preparaton queue , returns to falure drectly
* 2，exsts pause task，complement not completed, pendng submsson of tasks, return to suspenson
* 3，success
*
* @return ExecutonStatus
*/
prvate ExecutonStatus processReadyPause() {
f (hasRetryTaskInStandBy()) {
return ExecutonStatus.FAILURE;
}
Lst<TaskInstance> pauseLst = getCompleteTaskByState(ExecutonStatus.PAUSE);
f (CollectonUtls.sNotEmpty(pauseLst) || processInstance.sBlocked() || !sComplementEnd() || readyToSubmtTaskQueue.sze() > 0) {
return ExecutonStatus.PAUSE;
} else {
return ExecutonStatus.SUCCESS; |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,225 | [Feature][Alert] Add close alert method | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
Right now, we can set the alert when workflow executes failure.
In some cases, if the alert plugin is a third party service like [pagerduty](https://www.pagerduty.com), when we send an alert we will call the service to send an alert, and it may send 5 minutes unless we close the alert(We can close the alert by manual too).
But we may hope when the workflow success, the before alert can close automatically, so this issue is want to discuss add a new method closeAlert in alert plugin, we will provide a default method, the default method will do nothing, you can override this.
When the workflow execute success we will check if there exist a alert, and insert a close alert event, the AlertServer side will consume the event and execute the closeMethod.
### Use case
_No response_
### Related issues
_No response_
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11225 | https://github.com/apache/dolphinscheduler/pull/11221 | 01fc6c8a3a3aa68a45879ec5130250c99492a86b | 8e21c38c0041aa2689eb9d9bc703f67292c98e88 | 2022-08-01T03:59:56Z | java | 2022-08-01T05:04:27Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | }
}
/**
* prepare for block
* f process has tasks stll runnng, pause them
* f readyToSubmtTaskQueue s not empty, kll them
* else return block status drectly
*
* @return ExecutonStatus
*/
prvate ExecutonStatus processReadyBlock() {
f (actveTaskProcessorMaps.sze() > 0) {
for (ITaskProcessor taskProcessor : actveTaskProcessorMaps.values()) {
f (!TASK_TYPE_BLOCKING.equals(taskProcessor.getType())) {
taskProcessor.acton(TaskActon.PAUSE);
}
}
}
f (readyToSubmtTaskQueue.sze() > 0) {
for (Iterator<TaskInstance> ter = readyToSubmtTaskQueue.terator(); ter.hasNext(); ) {
ter.next().setState(ExecutonStatus.KILL);
}
}
return ExecutonStatus.BLOCK;
}
/**
* generate the latest process nstance status by the tasks state
*
* @return process nstance executon status
*/ |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,225 | [Feature][Alert] Add close alert method | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
Right now, we can set the alert when workflow executes failure.
In some cases, if the alert plugin is a third party service like [pagerduty](https://www.pagerduty.com), when we send an alert we will call the service to send an alert, and it may send 5 minutes unless we close the alert(We can close the alert by manual too).
But we may hope when the workflow success, the before alert can close automatically, so this issue is want to discuss add a new method closeAlert in alert plugin, we will provide a default method, the default method will do nothing, you can override this.
When the workflow execute success we will check if there exist a alert, and insert a close alert event, the AlertServer side will consume the event and execute the closeMethod.
### Use case
_No response_
### Related issues
_No response_
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11225 | https://github.com/apache/dolphinscheduler/pull/11221 | 01fc6c8a3a3aa68a45879ec5130250c99492a86b | 8e21c38c0041aa2689eb9d9bc703f67292c98e88 | 2022-08-01T03:59:56Z | java | 2022-08-01T05:04:27Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | prvate ExecutonStatus getProcessInstanceState(ProcessInstance nstance) {
ExecutonStatus state = nstance.getState();
f (actveTaskProcessorMaps.sze() > 0 || hasRetryTaskInStandBy()) {
// actve
ExecutonStatus executonStatus = runnngState(state);
logger.nfo("The workflowInstance has task runnng, the workflowInstance status s {}", executonStatus);
return executonStatus;
}
// block
f (state == ExecutonStatus.READY_BLOCK) {
ExecutonStatus executonStatus = processReadyBlock();
logger.nfo("The workflowInstance s ready to block, the workflowInstance status s {}", executonStatus);
return executonStatus;
}
// watng thread
f (hasWatngThreadTask()) {
logger.nfo("The workflowInstance has watng thread task, the workflow status s {}",
ExecutonStatus.WAITING_THREAD);
return ExecutonStatus.WAITING_THREAD;
}
// pause
f (state == ExecutonStatus.READY_PAUSE) {
ExecutonStatus executonStatus = processReadyPause();
logger.nfo("The workflowInstance s ready to pause, the workflow status s {}", executonStatus);
return executonStatus;
}
// stop
f (state == ExecutonStatus.READY_STOP) {
Lst<TaskInstance> stopLst = getCompleteTaskByState(ExecutonStatus.STOP);
Lst<TaskInstance> kllLst = getCompleteTaskByState(ExecutonStatus.KILL); |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,225 | [Feature][Alert] Add close alert method | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
Right now, we can set the alert when workflow executes failure.
In some cases, if the alert plugin is a third party service like [pagerduty](https://www.pagerduty.com), when we send an alert we will call the service to send an alert, and it may send 5 minutes unless we close the alert(We can close the alert by manual too).
But we may hope when the workflow success, the before alert can close automatically, so this issue is want to discuss add a new method closeAlert in alert plugin, we will provide a default method, the default method will do nothing, you can override this.
When the workflow execute success we will check if there exist a alert, and insert a close alert event, the AlertServer side will consume the event and execute the closeMethod.
### Use case
_No response_
### Related issues
_No response_
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11225 | https://github.com/apache/dolphinscheduler/pull/11221 | 01fc6c8a3a3aa68a45879ec5130250c99492a86b | 8e21c38c0041aa2689eb9d9bc703f67292c98e88 | 2022-08-01T03:59:56Z | java | 2022-08-01T05:04:27Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | Lst<TaskInstance> falLst = getCompleteTaskByState(ExecutonStatus.FAILURE);
ExecutonStatus executonStatus;
f (CollectonUtls.sNotEmpty(stopLst) || CollectonUtls.sNotEmpty(kllLst) || CollectonUtls.sNotEmpty(falLst) || !sComplementEnd()) {
executonStatus = ExecutonStatus.STOP;
} else {
executonStatus = ExecutonStatus.SUCCESS;
}
logger.nfo("The workflowInstance s ready to stop, the workflow status s {}", executonStatus);
return executonStatus;
}
// process falure
f (processFaled()) {
logger.nfo("The workflowInstance s faled, the workflow status s {}", ExecutonStatus.FAILURE);
return ExecutonStatus.FAILURE;
}
// success
f (state == ExecutonStatus.RUNNING_EXECUTION) {
Lst<TaskInstance> kllTasks = getCompleteTaskByState(ExecutonStatus.KILL);
f (readyToSubmtTaskQueue.sze() > 0 || watToRetryTaskInstanceMap.sze() > 0) {
//tasks cu
return ExecutonStatus.RUNNING_EXECUTION;
} else f (CollectonUtls.sNotEmpty(kllTasks)) {
// tasks m
return ExecutonStatus.FAILURE;
} else {
// f the
return ExecutonStatus.SUCCESS;
}
}
return state; |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,225 | [Feature][Alert] Add close alert method | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
Right now, we can set the alert when workflow executes failure.
In some cases, if the alert plugin is a third party service like [pagerduty](https://www.pagerduty.com), when we send an alert we will call the service to send an alert, and it may send 5 minutes unless we close the alert(We can close the alert by manual too).
But we may hope when the workflow success, the before alert can close automatically, so this issue is want to discuss add a new method closeAlert in alert plugin, we will provide a default method, the default method will do nothing, you can override this.
When the workflow execute success we will check if there exist a alert, and insert a close alert event, the AlertServer side will consume the event and execute the closeMethod.
### Use case
_No response_
### Related issues
_No response_
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11225 | https://github.com/apache/dolphinscheduler/pull/11221 | 01fc6c8a3a3aa68a45879ec5130250c99492a86b | 8e21c38c0041aa2689eb9d9bc703f67292c98e88 | 2022-08-01T03:59:56Z | java | 2022-08-01T05:04:27Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | }
/**
* whether complement end
*
* @return Boolean whether s complement end
*/
prvate boolean sComplementEnd() {
f (!processInstance.sComplementData()) {
return true;
}
Map<Strng, Strng> cmdParam = JSONUtls.toMap(processInstance.getCommandParam());
Date endTme = DateUtls.strngToDate(cmdParam.get(CMDPARAM_COMPLEMENT_DATA_END_DATE));
return processInstance.getScheduleTme().equals(endTme);
}
/**
* updateProcessInstance process nstance state
* after each batch of tasks s executed, of the process nstance s updated
*/
prvate vod updateProcessInstanceState() throws StateEventHandleExcepton {
ExecutonStatus state = getProcessInstanceState(processInstance);
f (processInstance.getState() != state) {
logger.nfo("Update workflowInstance states, orgn state: {}, target state: {}",
processInstance.getState(),
state);
updateWorkflowInstanceStatesToDB(state);
StateEvent stateEvent = new StateEvent();
stateEvent.setExecutonStatus(processInstance.getState());
stateEvent.setProcessInstanceId(ths.processInstance.getId());
stateEvent.setType(StateEventType.PROCESS_STATE_CHANGE);
// replace |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,225 | [Feature][Alert] Add close alert method | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
Right now, we can set the alert when workflow executes failure.
In some cases, if the alert plugin is a third party service like [pagerduty](https://www.pagerduty.com), when we send an alert we will call the service to send an alert, and it may send 5 minutes unless we close the alert(We can close the alert by manual too).
But we may hope when the workflow success, the before alert can close automatically, so this issue is want to discuss add a new method closeAlert in alert plugin, we will provide a default method, the default method will do nothing, you can override this.
When the workflow execute success we will check if there exist a alert, and insert a close alert event, the AlertServer side will consume the event and execute the closeMethod.
### Use case
_No response_
### Related issues
_No response_
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11225 | https://github.com/apache/dolphinscheduler/pull/11221 | 01fc6c8a3a3aa68a45879ec5130250c99492a86b | 8e21c38c0041aa2689eb9d9bc703f67292c98e88 | 2022-08-01T03:59:56Z | java | 2022-08-01T05:04:27Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | ths.stateEvents.add(stateEvent);
} else {
logger.nfo("There s no need to update the workflow nstance state, orgn state: {}, target state: {}",
processInstance.getState(),
state);
}
}
/**
* stateEvent's executon status as process nstance state
*/
publc vod updateProcessInstanceState(StateEvent stateEvent) throws StateEventHandleExcepton {
ExecutonStatus state = stateEvent.getExecutonStatus();
updateWorkflowInstanceStatesToDB(state);
}
prvate vod updateWorkflowInstanceStatesToDB(ExecutonStatus newStates) throws StateEventHandleExcepton {
ExecutonStatus orgnStates = processInstance.getState();
f (orgnStates != newStates) {
logger.nfo("Begn to update workflow nstance state , state wll change from {} to {}",
orgnStates,
newStates);
processInstance.setState(newStates);
f (newStates.typeIsFnshed()) {
processInstance.setEndTme(new Date());
}
try {
processServce.updateProcessInstance(processInstance);
} catch (Excepton ex) {
// recover
processInstance.setState(orgnStates);
processInstance.setEndTme(null); |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,225 | [Feature][Alert] Add close alert method | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
Right now, we can set the alert when workflow executes failure.
In some cases, if the alert plugin is a third party service like [pagerduty](https://www.pagerduty.com), when we send an alert we will call the service to send an alert, and it may send 5 minutes unless we close the alert(We can close the alert by manual too).
But we may hope when the workflow success, the before alert can close automatically, so this issue is want to discuss add a new method closeAlert in alert plugin, we will provide a default method, the default method will do nothing, you can override this.
When the workflow execute success we will check if there exist a alert, and insert a close alert event, the AlertServer side will consume the event and execute the closeMethod.
### Use case
_No response_
### Related issues
_No response_
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11225 | https://github.com/apache/dolphinscheduler/pull/11221 | 01fc6c8a3a3aa68a45879ec5130250c99492a86b | 8e21c38c0041aa2689eb9d9bc703f67292c98e88 | 2022-08-01T03:59:56Z | java | 2022-08-01T05:04:27Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | throw new StateEventHandleExcepton("Update process nstance status to DB error", ex);
}
}
}
/**
* get task dependency result
*
* @param taskInstance task nstance
* @return DependResult
*/
prvate DependResult getDependResultForTask(TaskInstance taskInstance) {
return sTaskDepsComplete(Long.toStrng(taskInstance.getTaskCode()));
}
/**
* add task to standby lst
*
* @param taskInstance task nstance
*/
publc vod addTaskToStandByLst(TaskInstance taskInstance) {
f (readyToSubmtTaskQueue.contans(taskInstance)) {
logger.warn("task was found n ready submt queue, task code:{}", taskInstance.getTaskCode());
return;
}
logger.nfo("add task to stand by lst, task name:{}, task d:{}, task code:{}",
taskInstance.getName(),
taskInstance.getId(),
taskInstance.getTaskCode());
TaskMetrcs.ncTaskInstanceByState("submt");
readyToSubmtTaskQueue.put(taskInstance);
} |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,225 | [Feature][Alert] Add close alert method | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
Right now, we can set the alert when workflow executes failure.
In some cases, if the alert plugin is a third party service like [pagerduty](https://www.pagerduty.com), when we send an alert we will call the service to send an alert, and it may send 5 minutes unless we close the alert(We can close the alert by manual too).
But we may hope when the workflow success, the before alert can close automatically, so this issue is want to discuss add a new method closeAlert in alert plugin, we will provide a default method, the default method will do nothing, you can override this.
When the workflow execute success we will check if there exist a alert, and insert a close alert event, the AlertServer side will consume the event and execute the closeMethod.
### Use case
_No response_
### Related issues
_No response_
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11225 | https://github.com/apache/dolphinscheduler/pull/11221 | 01fc6c8a3a3aa68a45879ec5130250c99492a86b | 8e21c38c0041aa2689eb9d9bc703f67292c98e88 | 2022-08-01T03:59:56Z | java | 2022-08-01T05:04:27Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | /**
* remove task from stand by lst
*
* @param taskInstance task nstance
*/
prvate boolean removeTaskFromStandbyLst(TaskInstance taskInstance) {
return readyToSubmtTaskQueue.remove(taskInstance);
}
/**
* has retry task n standby
*
* @return Boolean whether has retry task n standby
*/
prvate boolean hasRetryTaskInStandBy() {
for (Iterator<TaskInstance> ter = readyToSubmtTaskQueue.terator(); ter.hasNext(); ) {
f (ter.next().getState().typeIsFalure()) {
return true;
}
}
return false;
}
/**
* close the on gong tasks
*/
publc vod kllAllTasks() {
logger.nfo("kll called on process nstance d: {}, num: {}",
processInstance.getId(),
actveTaskProcessorMaps.sze());
f (readyToSubmtTaskQueue.sze() > 0) {
readyToSubmtTaskQueue.clear(); |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,225 | [Feature][Alert] Add close alert method | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
Right now, we can set the alert when workflow executes failure.
In some cases, if the alert plugin is a third party service like [pagerduty](https://www.pagerduty.com), when we send an alert we will call the service to send an alert, and it may send 5 minutes unless we close the alert(We can close the alert by manual too).
But we may hope when the workflow success, the before alert can close automatically, so this issue is want to discuss add a new method closeAlert in alert plugin, we will provide a default method, the default method will do nothing, you can override this.
When the workflow execute success we will check if there exist a alert, and insert a close alert event, the AlertServer side will consume the event and execute the closeMethod.
### Use case
_No response_
### Related issues
_No response_
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11225 | https://github.com/apache/dolphinscheduler/pull/11221 | 01fc6c8a3a3aa68a45879ec5130250c99492a86b | 8e21c38c0041aa2689eb9d9bc703f67292c98e88 | 2022-08-01T03:59:56Z | java | 2022-08-01T05:04:27Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | }
for (long taskCode : actveTaskProcessorMaps.keySet()) {
ITaskProcessor taskProcessor = actveTaskProcessorMaps.get(taskCode);
Integer taskInstanceId = valdTaskMap.get(taskCode);
f (taskInstanceId == null || taskInstanceId.equals(0)) {
contnue;
}
TaskInstance taskInstance = processServce.fndTaskInstanceById(taskInstanceId);
f (taskInstance == null || taskInstance.getState().typeIsFnshed()) {
contnue;
}
taskProcessor.acton(TaskActon.STOP);
f (taskProcessor.taskInstance().getState().typeIsFnshed()) {
StateEvent stateEvent = new StateEvent();
stateEvent.setType(StateEventType.TASK_STATE_CHANGE);
stateEvent.setProcessInstanceId(ths.processInstance.getId());
stateEvent.setTaskInstanceId(taskInstance.getId());
stateEvent.setExecutonStatus(taskProcessor.taskInstance().getState());
ths.addStateEvent(stateEvent);
}
}
}
publc boolean workFlowFnsh() {
return ths.processInstance.getState().typeIsFnshed();
}
/**
* handlng the lst of tasks to be submtted
*/
publc vod submtStandByTask() throws StateEventHandleExcepton {
nt length = readyToSubmtTaskQueue.sze(); |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,225 | [Feature][Alert] Add close alert method | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
Right now, we can set the alert when workflow executes failure.
In some cases, if the alert plugin is a third party service like [pagerduty](https://www.pagerduty.com), when we send an alert we will call the service to send an alert, and it may send 5 minutes unless we close the alert(We can close the alert by manual too).
But we may hope when the workflow success, the before alert can close automatically, so this issue is want to discuss add a new method closeAlert in alert plugin, we will provide a default method, the default method will do nothing, you can override this.
When the workflow execute success we will check if there exist a alert, and insert a close alert event, the AlertServer side will consume the event and execute the closeMethod.
### Use case
_No response_
### Related issues
_No response_
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11225 | https://github.com/apache/dolphinscheduler/pull/11221 | 01fc6c8a3a3aa68a45879ec5130250c99492a86b | 8e21c38c0041aa2689eb9d9bc703f67292c98e88 | 2022-08-01T03:59:56Z | java | 2022-08-01T05:04:27Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | for (nt = 0; < length; ++) {
TaskInstance task = readyToSubmtTaskQueue.peek();
f (task == null) {
contnue;
}
// stop ta
f (task.taskCanRetry()) {
TaskInstance retryTask = processServce.fndTaskInstanceById(task.getId());
f (retryTask != null && retryTask.getState().equals(ExecutonStatus.FORCED_SUCCESS)) {
task.setState(retryTask.getState());
logger.nfo("task: {} has been forced success, put t nto complete task lst and stop retryng", task.getName());
removeTaskFromStandbyLst(task);
completeTaskMap.put(task.getTaskCode(), task.getId());
taskInstanceMap.put(task.getId(), task);
submtPostNode(Long.toStrng(task.getTaskCode()));
contnue;
}
}
//nt var
f (task.sFrstRun()) {
//get pre
Set<Strng> preTask = dag.getPrevousNodes(Long.toStrng(task.getTaskCode()));
getPreVarPool(task, preTask);
}
DependResult dependResult = getDependResultForTask(task);
f (DependResult.SUCCESS == dependResult) {
Optonal<TaskInstance> taskInstanceOptonal = submtTaskExec(task);
f (!taskInstanceOptonal.sPresent()) {
ths.taskFaledSubmt = true;
// Remove |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,225 | [Feature][Alert] Add close alert method | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
Right now, we can set the alert when workflow executes failure.
In some cases, if the alert plugin is a third party service like [pagerduty](https://www.pagerduty.com), when we send an alert we will call the service to send an alert, and it may send 5 minutes unless we close the alert(We can close the alert by manual too).
But we may hope when the workflow success, the before alert can close automatically, so this issue is want to discuss add a new method closeAlert in alert plugin, we will provide a default method, the default method will do nothing, you can override this.
When the workflow execute success we will check if there exist a alert, and insert a close alert event, the AlertServer side will consume the event and execute the closeMethod.
### Use case
_No response_
### Related issues
_No response_
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11225 | https://github.com/apache/dolphinscheduler/pull/11221 | 01fc6c8a3a3aa68a45879ec5130250c99492a86b | 8e21c38c0041aa2689eb9d9bc703f67292c98e88 | 2022-08-01T03:59:56Z | java | 2022-08-01T05:04:27Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | f (!removeTaskFromStandbyLst(task)) {
logger.error(
"Task submt faled, remove from standby lst faled, workflowInstanceId: {}, taskCode: {}",
processInstance.getId(),
task.getTaskCode());
}
completeTaskMap.put(task.getTaskCode(), task.getId());
taskInstanceMap.put(task.getId(), task);
errorTaskMap.put(task.getTaskCode(), task.getId());
actveTaskProcessorMaps.remove(task.getTaskCode());
logger.error("Task submtted faled, workflowInstanceId: {}, taskInstanceId: {}, taskCode: {}",
task.getProcessInstanceId(),
task.getId(),
task.getTaskCode());
} else {
removeTaskFromStandbyLst(task);
}
} else f (DependResult.FAILED == dependResult) {
// f the
dependFaledTaskMap.put(task.getTaskCode(), task.getId());
removeTaskFromStandbyLst(task);
logger.nfo("Task dependent result s faled, taskInstanceId:{} depend result : {}", task.getId(), dependResult);
} else f (DependResult.NON_EXEC == dependResult) {
// for som
removeTaskFromStandbyLst(task);
logger.nfo("Remove task due to depend result not executed, taskInstanceId:{} depend result : {}", task.getId(), dependResult);
}
}
}
/** |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,225 | [Feature][Alert] Add close alert method | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
Right now, we can set the alert when workflow executes failure.
In some cases, if the alert plugin is a third party service like [pagerduty](https://www.pagerduty.com), when we send an alert we will call the service to send an alert, and it may send 5 minutes unless we close the alert(We can close the alert by manual too).
But we may hope when the workflow success, the before alert can close automatically, so this issue is want to discuss add a new method closeAlert in alert plugin, we will provide a default method, the default method will do nothing, you can override this.
When the workflow execute success we will check if there exist a alert, and insert a close alert event, the AlertServer side will consume the event and execute the closeMethod.
### Use case
_No response_
### Related issues
_No response_
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11225 | https://github.com/apache/dolphinscheduler/pull/11221 | 01fc6c8a3a3aa68a45879ec5130250c99492a86b | 8e21c38c0041aa2689eb9d9bc703f67292c98e88 | 2022-08-01T03:59:56Z | java | 2022-08-01T05:04:27Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | * Get start task nstance lst from recover
*
* @param cmdParam command param
* @return task nstance lst
*/
protected Lst<TaskInstance> getRecoverTaskInstanceLst(Strng cmdParam) {
Map<Strng, Strng> paramMap = JSONUtls.toMap(cmdParam);
// todo: Can we use a better way to set the recover taskInstanceId lst? rather then use the cmdParam
f (paramMap != null && paramMap.contansKey(CMD_PARAM_RECOVERY_START_NODE_STRING)) {
Strng[] dLst = paramMap.get(CMD_PARAM_RECOVERY_START_NODE_STRING).splt(Constants.COMMA);
f (ArrayUtls.sNotEmpty(dLst)) {
Lst<Integer> taskInstanceIds = Arrays.stream(dLst).map(Integer::valueOf).collect(Collectors.toLst());
return processServce.fndTaskInstanceByIdLst(taskInstanceIds);
}
}
return Collectons.emptyLst();
}
/**
* parse "StartNodeNameLst" from cmd param
*
* @param cmdParam command param
* @return start node name lst
*/
prvate Lst<Strng> parseStartNodeName(Strng cmdParam) {
Lst<Strng> startNodeNameLst = new ArrayLst<>();
Map<Strng, Strng> paramMap = JSONUtls.toMap(cmdParam);
f (paramMap == null) {
return startNodeNameLst;
}
f (paramMap.contansKey(CMD_PARAM_START_NODES)) { |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,225 | [Feature][Alert] Add close alert method | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
Right now, we can set the alert when workflow executes failure.
In some cases, if the alert plugin is a third party service like [pagerduty](https://www.pagerduty.com), when we send an alert we will call the service to send an alert, and it may send 5 minutes unless we close the alert(We can close the alert by manual too).
But we may hope when the workflow success, the before alert can close automatically, so this issue is want to discuss add a new method closeAlert in alert plugin, we will provide a default method, the default method will do nothing, you can override this.
When the workflow execute success we will check if there exist a alert, and insert a close alert event, the AlertServer side will consume the event and execute the closeMethod.
### Use case
_No response_
### Related issues
_No response_
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11225 | https://github.com/apache/dolphinscheduler/pull/11221 | 01fc6c8a3a3aa68a45879ec5130250c99492a86b | 8e21c38c0041aa2689eb9d9bc703f67292c98e88 | 2022-08-01T03:59:56Z | java | 2022-08-01T05:04:27Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | startNodeNameLst = Arrays.asLst(paramMap.get(CMD_PARAM_START_NODES).splt(Constants.COMMA));
}
return startNodeNameLst;
}
/**
* generate start node code lst from parsng command param;
* f "StartNodeIdLst" exsts n command param, return StartNodeIdLst
*
* @return recovery node code lst
*/
prvate Lst<Strng> getRecoveryNodeCodeLst(Lst<TaskInstance> recoverNodeLst) {
Lst<Strng> recoveryNodeCodeLst = new ArrayLst<>();
f (CollectonUtls.sNotEmpty(recoverNodeLst)) {
for (TaskInstance task : recoverNodeLst) {
recoveryNodeCodeLst.add(Long.toStrng(task.getTaskCode()));
}
}
return recoveryNodeCodeLst;
}
/**
* generate flow dag
*
* @param totalTaskNodeLst total task node lst
* @param startNodeNameLst start node name lst
* @param recoveryNodeCodeLst recovery node code lst
* @param depNodeType depend node type
* @return ProcessDag process dag
* @throws Excepton excepton
*/
publc ProcessDag generateFlowDag(Lst<TaskNode> totalTaskNodeLst, |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,225 | [Feature][Alert] Add close alert method | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
Right now, we can set the alert when workflow executes failure.
In some cases, if the alert plugin is a third party service like [pagerduty](https://www.pagerduty.com), when we send an alert we will call the service to send an alert, and it may send 5 minutes unless we close the alert(We can close the alert by manual too).
But we may hope when the workflow success, the before alert can close automatically, so this issue is want to discuss add a new method closeAlert in alert plugin, we will provide a default method, the default method will do nothing, you can override this.
When the workflow execute success we will check if there exist a alert, and insert a close alert event, the AlertServer side will consume the event and execute the closeMethod.
### Use case
_No response_
### Related issues
_No response_
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11225 | https://github.com/apache/dolphinscheduler/pull/11221 | 01fc6c8a3a3aa68a45879ec5130250c99492a86b | 8e21c38c0041aa2689eb9d9bc703f67292c98e88 | 2022-08-01T03:59:56Z | java | 2022-08-01T05:04:27Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | Lst<Strng> startNodeNameLst,
Lst<Strng> recoveryNodeCodeLst,
TaskDependType depNodeType) throws Excepton {
return DagHelper.generateFlowDag(totalTaskNodeLst, startNodeNameLst, recoveryNodeCodeLst, depNodeType);
}
/**
* check task queue
*/
prvate boolean checkTaskQueue() {
AtomcBoolean result = new AtomcBoolean(false);
taskInstanceMap.forEach((d, taskInstance) -> {
f (taskInstance != null && taskInstance.getTaskGroupId() > 0) {
result.set(true);
}
});
return result.get();
}
/**
* s new process nstance
*/
prvate boolean sNewProcessInstance() {
f (Flag.YES.equals(processInstance.getRecovery())) {
logger.nfo("Ths workInstance wll be recover by ths executon");
return false;
}
f (ExecutonStatus.RUNNING_EXECUTION == processInstance.getState() && processInstance.getRunTmes() == 1) {
return true;
}
logger.nfo(
"The workflowInstance has been executed before, ths executon s to reRun, processInstance status: {}, runTmes: {}", |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,225 | [Feature][Alert] Add close alert method | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
Right now, we can set the alert when workflow executes failure.
In some cases, if the alert plugin is a third party service like [pagerduty](https://www.pagerduty.com), when we send an alert we will call the service to send an alert, and it may send 5 minutes unless we close the alert(We can close the alert by manual too).
But we may hope when the workflow success, the before alert can close automatically, so this issue is want to discuss add a new method closeAlert in alert plugin, we will provide a default method, the default method will do nothing, you can override this.
When the workflow execute success we will check if there exist a alert, and insert a close alert event, the AlertServer side will consume the event and execute the closeMethod.
### Use case
_No response_
### Related issues
_No response_
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11225 | https://github.com/apache/dolphinscheduler/pull/11221 | 01fc6c8a3a3aa68a45879ec5130250c99492a86b | 8e21c38c0041aa2689eb9d9bc703f67292c98e88 | 2022-08-01T03:59:56Z | java | 2022-08-01T05:04:27Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | processInstance.getState(),
processInstance.getRunTmes());
return false;
}
publc vod resubmt(long taskCode) throws Excepton {
ITaskProcessor taskProcessor = actveTaskProcessorMaps.get(taskCode);
f (taskProcessor != null) {
taskProcessor.acton(TaskActon.RESUBMIT);
logger.debug("RESUBMIT: task code:{}", taskCode);
} else {
throw new Excepton("resubmt error, taskProcessor s null, task code: " + taskCode);
}
}
publc Map<Long, Integer> getCompleteTaskMap() {
return completeTaskMap;
}
publc Map<Long, ITaskProcessor> getActveTaskProcessMap() {
return actveTaskProcessorMaps;
}
publc Map<Long, TaskInstance> getWatToRetryTaskInstanceMap() {
return watToRetryTaskInstanceMap;
}
prvate vod setGlobalParamIfCommanded(ProcessDefnton processDefnton, Map<Strng, Strng> cmdParam) {
// get start params from command param
Map<Strng, Strng> startParamMap = new HashMap<>();
f (cmdParam.contansKey(Constants.CMD_PARAM_START_PARAMS)) {
Strng startParamJson = cmdParam.get(Constants.CMD_PARAM_START_PARAMS);
startParamMap = JSONUtls.toMap(startParamJson);
}
Map<Strng, Strng> fatherParamMap = new HashMap<>(); |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,225 | [Feature][Alert] Add close alert method | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
Right now, we can set the alert when workflow executes failure.
In some cases, if the alert plugin is a third party service like [pagerduty](https://www.pagerduty.com), when we send an alert we will call the service to send an alert, and it may send 5 minutes unless we close the alert(We can close the alert by manual too).
But we may hope when the workflow success, the before alert can close automatically, so this issue is want to discuss add a new method closeAlert in alert plugin, we will provide a default method, the default method will do nothing, you can override this.
When the workflow execute success we will check if there exist a alert, and insert a close alert event, the AlertServer side will consume the event and execute the closeMethod.
### Use case
_No response_
### Related issues
_No response_
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11225 | https://github.com/apache/dolphinscheduler/pull/11221 | 01fc6c8a3a3aa68a45879ec5130250c99492a86b | 8e21c38c0041aa2689eb9d9bc703f67292c98e88 | 2022-08-01T03:59:56Z | java | 2022-08-01T05:04:27Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | f (cmdParam.contansKey(Constants.CMD_PARAM_FATHER_PARAMS)) {
Strng fatherParamJson = cmdParam.get(Constants.CMD_PARAM_FATHER_PARAMS);
fatherParamMap = JSONUtls.toMap(fatherParamJson);
}
startParamMap.putAll(fatherParamMap);
// set start param nto global params
Map<Strng, Strng> globalMap = processDefnton.getGlobalParamMap();
Lst<Property> globalParamLst = processDefnton.getGlobalParamLst();
f (startParamMap.sze() > 0 && globalMap != null) {
//start pa
for (Map.Entry<Strng, Strng> param : globalMap.entrySet()) {
Strng val = startParamMap.get(param.getKey());
f (val != null) {
param.setValue(val);
}
}
//start pa
for (Map.Entry<Strng, Strng> startParam : startParamMap.entrySet()) {
f (!globalMap.contansKey(startParam.getKey())) {
globalMap.put(startParam.getKey(), startParam.getValue());
globalParamLst.add(new Property(startParam.getKey(), IN, VARCHAR, startParam.getValue()));
}
}
}
}
prvate enum WorkflowRunnableStatus {
CREATED, INITIALIZE_DAG, INITIALIZE_QUEUE, STARTED,
;
}
} |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,225 | [Feature][Alert] Add close alert method | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
Right now, we can set the alert when workflow executes failure.
In some cases, if the alert plugin is a third party service like [pagerduty](https://www.pagerduty.com), when we send an alert we will call the service to send an alert, and it may send 5 minutes unless we close the alert(We can close the alert by manual too).
But we may hope when the workflow success, the before alert can close automatically, so this issue is want to discuss add a new method closeAlert in alert plugin, we will provide a default method, the default method will do nothing, you can override this.
When the workflow execute success we will check if there exist a alert, and insert a close alert event, the AlertServer side will consume the event and execute the closeMethod.
### Use case
_No response_
### Related issues
_No response_
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11225 | https://github.com/apache/dolphinscheduler/pull/11221 | 01fc6c8a3a3aa68a45879ec5130250c99492a86b | 8e21c38c0041aa2689eb9d9bc703f67292c98e88 | 2022-08-01T03:59:56Z | java | 2022-08-01T05:04:27Z | dolphinscheduler-service/src/main/java/org/apache/dolphinscheduler/service/alert/ProcessAlertManager.java | /*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package org.apache.dolphinscheduler.service.alert;
import org.apache.dolphinscheduler.common.enums.AlertType;
import org.apache.dolphinscheduler.common.enums.CommandType;
import org.apache.dolphinscheduler.common.enums.Flag;
import org.apache.dolphinscheduler.common.enums.WarningType;
import org.apache.dolphinscheduler.common.utils.JSONUtils;
import org.apache.dolphinscheduler.dao.AlertDao; |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,225 | [Feature][Alert] Add close alert method | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
Right now, we can set the alert when workflow executes failure.
In some cases, if the alert plugin is a third party service like [pagerduty](https://www.pagerduty.com), when we send an alert we will call the service to send an alert, and it may send 5 minutes unless we close the alert(We can close the alert by manual too).
But we may hope when the workflow success, the before alert can close automatically, so this issue is want to discuss add a new method closeAlert in alert plugin, we will provide a default method, the default method will do nothing, you can override this.
When the workflow execute success we will check if there exist a alert, and insert a close alert event, the AlertServer side will consume the event and execute the closeMethod.
### Use case
_No response_
### Related issues
_No response_
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11225 | https://github.com/apache/dolphinscheduler/pull/11221 | 01fc6c8a3a3aa68a45879ec5130250c99492a86b | 8e21c38c0041aa2689eb9d9bc703f67292c98e88 | 2022-08-01T03:59:56Z | java | 2022-08-01T05:04:27Z | dolphinscheduler-service/src/main/java/org/apache/dolphinscheduler/service/alert/ProcessAlertManager.java | import org.apache.dolphinscheduler.dao.entity.Alert;
import org.apache.dolphinscheduler.dao.entity.DqExecuteResult;
import org.apache.dolphinscheduler.dao.entity.DqExecuteResultAlertContent;
import org.apache.dolphinscheduler.dao.entity.ProcessAlertContent;
import org.apache.dolphinscheduler.dao.entity.ProcessInstance;
import org.apache.dolphinscheduler.dao.entity.ProjectUser;
import org.apache.dolphinscheduler.dao.entity.TaskAlertContent;
import org.apache.dolphinscheduler.dao.entity.TaskInstance;
import org.apache.dolphinscheduler.plugin.task.api.enums.dp.DqTaskState;
import java.util.ArrayList;
import java.util.Date;
import java.util.List;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.stereotype.Component;
/**
* process alert manager
*/
@Component
public class ProcessAlertManager {
/**
* logger of AlertManager
*/
private static final Logger logger = LoggerFactory.getLogger(ProcessAlertManager.class);
/**
* alert dao
*/
@Autowired
private AlertDao alertDao; |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,225 | [Feature][Alert] Add close alert method | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
Right now, we can set the alert when workflow executes failure.
In some cases, if the alert plugin is a third party service like [pagerduty](https://www.pagerduty.com), when we send an alert we will call the service to send an alert, and it may send 5 minutes unless we close the alert(We can close the alert by manual too).
But we may hope when the workflow success, the before alert can close automatically, so this issue is want to discuss add a new method closeAlert in alert plugin, we will provide a default method, the default method will do nothing, you can override this.
When the workflow execute success we will check if there exist a alert, and insert a close alert event, the AlertServer side will consume the event and execute the closeMethod.
### Use case
_No response_
### Related issues
_No response_
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11225 | https://github.com/apache/dolphinscheduler/pull/11221 | 01fc6c8a3a3aa68a45879ec5130250c99492a86b | 8e21c38c0041aa2689eb9d9bc703f67292c98e88 | 2022-08-01T03:59:56Z | java | 2022-08-01T05:04:27Z | dolphinscheduler-service/src/main/java/org/apache/dolphinscheduler/service/alert/ProcessAlertManager.java | /**
* command type convert chinese
*
* @param commandType command type
* @return command name
*/
private String getCommandCnName(CommandType commandType) {
switch (commandType) {
case RECOVER_TOLERANCE_FAULT_PROCESS:
return "recover tolerance fault process";
case RECOVER_SUSPENDED_PROCESS:
return "recover suspended process";
case START_CURRENT_TASK_PROCESS:
return "start current task process";
case START_FAILURE_TASK_PROCESS:
return "start failure task process";
case START_PROCESS:
return "start process";
case REPEAT_RUNNING:
return "repeat running";
case SCHEDULER:
return "scheduler";
case COMPLEMENT_DATA:
return "complement data";
case PAUSE:
return "pause";
case STOP:
return "stop";
default:
return "unknown type"; |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,225 | [Feature][Alert] Add close alert method | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
Right now, we can set the alert when workflow executes failure.
In some cases, if the alert plugin is a third party service like [pagerduty](https://www.pagerduty.com), when we send an alert we will call the service to send an alert, and it may send 5 minutes unless we close the alert(We can close the alert by manual too).
But we may hope when the workflow success, the before alert can close automatically, so this issue is want to discuss add a new method closeAlert in alert plugin, we will provide a default method, the default method will do nothing, you can override this.
When the workflow execute success we will check if there exist a alert, and insert a close alert event, the AlertServer side will consume the event and execute the closeMethod.
### Use case
_No response_
### Related issues
_No response_
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11225 | https://github.com/apache/dolphinscheduler/pull/11221 | 01fc6c8a3a3aa68a45879ec5130250c99492a86b | 8e21c38c0041aa2689eb9d9bc703f67292c98e88 | 2022-08-01T03:59:56Z | java | 2022-08-01T05:04:27Z | dolphinscheduler-service/src/main/java/org/apache/dolphinscheduler/service/alert/ProcessAlertManager.java | }
}
/**
* get process instance content
*
* @param processInstance process instance
* @param taskInstances task instance list
* @return process instance format content
*/
public String getContentProcessInstance(ProcessInstance processInstance,
List<TaskInstance> taskInstances,
ProjectUser projectUser) {
String res = "";
if (processInstance.getState().typeIsSuccess()) {
List<ProcessAlertContent> successTaskList = new ArrayList<>(1);
ProcessAlertContent processAlertContent = ProcessAlertContent.newBuilder()
.projectCode(projectUser.getProjectCode())
.projectName(projectUser.getProjectName())
.owner(projectUser.getUserName())
.processId(processInstance.getId())
.processDefinitionCode(processInstance.getProcessDefinitionCode())
.processName(processInstance.getName())
.processType(processInstance.getCommandType())
.processState(processInstance.getState())
.recovery(processInstance.getRecovery())
.runTimes(processInstance.getRunTimes())
.processStartTime(processInstance.getStartTime())
.processEndTime(processInstance.getEndTime())
.processHost(processInstance.getHost())
.build(); |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,225 | [Feature][Alert] Add close alert method | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
Right now, we can set the alert when workflow executes failure.
In some cases, if the alert plugin is a third party service like [pagerduty](https://www.pagerduty.com), when we send an alert we will call the service to send an alert, and it may send 5 minutes unless we close the alert(We can close the alert by manual too).
But we may hope when the workflow success, the before alert can close automatically, so this issue is want to discuss add a new method closeAlert in alert plugin, we will provide a default method, the default method will do nothing, you can override this.
When the workflow execute success we will check if there exist a alert, and insert a close alert event, the AlertServer side will consume the event and execute the closeMethod.
### Use case
_No response_
### Related issues
_No response_
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11225 | https://github.com/apache/dolphinscheduler/pull/11221 | 01fc6c8a3a3aa68a45879ec5130250c99492a86b | 8e21c38c0041aa2689eb9d9bc703f67292c98e88 | 2022-08-01T03:59:56Z | java | 2022-08-01T05:04:27Z | dolphinscheduler-service/src/main/java/org/apache/dolphinscheduler/service/alert/ProcessAlertManager.java | successTaskList.add(processAlertContent);
res = JSONUtils.toJsonString(successTaskList);
} else if (processInstance.getState().typeIsFailure()) {
List<ProcessAlertContent> failedTaskList = new ArrayList<>();
for (TaskInstance task : taskInstances) {
if (task.getState().typeIsSuccess()) {
continue;
}
ProcessAlertContent processAlertContent = ProcessAlertContent.newBuilder()
.projectCode(projectUser.getProjectCode())
.projectName(projectUser.getProjectName())
.owner(projectUser.getUserName())
.processId(processInstance.getId())
.processDefinitionCode(processInstance.getProcessDefinitionCode())
.processName(processInstance.getName())
.taskCode(task.getTaskCode())
.taskName(task.getName())
.taskType(task.getTaskType())
.taskState(task.getState())
.taskStartTime(task.getStartTime())
.taskEndTime(task.getEndTime())
.taskHost(task.getHost())
.logPath(task.getLogPath())
.build();
failedTaskList.add(processAlertContent);
}
res = JSONUtils.toJsonString(failedTaskList);
}
return res;
} |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,225 | [Feature][Alert] Add close alert method | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
Right now, we can set the alert when workflow executes failure.
In some cases, if the alert plugin is a third party service like [pagerduty](https://www.pagerduty.com), when we send an alert we will call the service to send an alert, and it may send 5 minutes unless we close the alert(We can close the alert by manual too).
But we may hope when the workflow success, the before alert can close automatically, so this issue is want to discuss add a new method closeAlert in alert plugin, we will provide a default method, the default method will do nothing, you can override this.
When the workflow execute success we will check if there exist a alert, and insert a close alert event, the AlertServer side will consume the event and execute the closeMethod.
### Use case
_No response_
### Related issues
_No response_
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11225 | https://github.com/apache/dolphinscheduler/pull/11221 | 01fc6c8a3a3aa68a45879ec5130250c99492a86b | 8e21c38c0041aa2689eb9d9bc703f67292c98e88 | 2022-08-01T03:59:56Z | java | 2022-08-01T05:04:27Z | dolphinscheduler-service/src/main/java/org/apache/dolphinscheduler/service/alert/ProcessAlertManager.java | /**
* getting worker fault tolerant content
*
* @param processInstance process instance
* @param toleranceTaskList tolerance task list
* @return worker tolerance content
*/
private String getWorkerToleranceContent(ProcessInstance processInstance, List<TaskInstance> toleranceTaskList) {
List<ProcessAlertContent> toleranceTaskInstanceList = new ArrayList<>();
for (TaskInstance taskInstance : toleranceTaskList) {
ProcessAlertContent processAlertContent = ProcessAlertContent.newBuilder()
.processId(processInstance.getId())
.processDefinitionCode(processInstance.getProcessDefinitionCode())
.processName(processInstance.getName())
.taskCode(taskInstance.getTaskCode())
.taskName(taskInstance.getName())
.taskHost(taskInstance.getHost())
.retryTimes(taskInstance.getRetryTimes())
.build();
toleranceTaskInstanceList.add(processAlertContent);
}
return JSONUtils.toJsonString(toleranceTaskInstanceList);
}
/**
* send worker alert fault tolerance
*
* @param processInstance process instance
* @param toleranceTaskList tolerance task list
*/
public void sendAlertWorkerToleranceFault(ProcessInstance processInstance, List<TaskInstance> toleranceTaskList) { |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,225 | [Feature][Alert] Add close alert method | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
Right now, we can set the alert when workflow executes failure.
In some cases, if the alert plugin is a third party service like [pagerduty](https://www.pagerduty.com), when we send an alert we will call the service to send an alert, and it may send 5 minutes unless we close the alert(We can close the alert by manual too).
But we may hope when the workflow success, the before alert can close automatically, so this issue is want to discuss add a new method closeAlert in alert plugin, we will provide a default method, the default method will do nothing, you can override this.
When the workflow execute success we will check if there exist a alert, and insert a close alert event, the AlertServer side will consume the event and execute the closeMethod.
### Use case
_No response_
### Related issues
_No response_
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11225 | https://github.com/apache/dolphinscheduler/pull/11221 | 01fc6c8a3a3aa68a45879ec5130250c99492a86b | 8e21c38c0041aa2689eb9d9bc703f67292c98e88 | 2022-08-01T03:59:56Z | java | 2022-08-01T05:04:27Z | dolphinscheduler-service/src/main/java/org/apache/dolphinscheduler/service/alert/ProcessAlertManager.java | try {
Alert alert = new Alert();
alert.setTitle("worker fault tolerance");
String content = getWorkerToleranceContent(processInstance, toleranceTaskList);
alert.setContent(content);
alert.setWarningType(WarningType.FAILURE);
alert.setCreateTime(new Date());
alert.setAlertGroupId(processInstance.getWarningGroupId() == null ? 1 : processInstance.getWarningGroupId());
alert.setAlertType(AlertType.FAULT_TOLERANCE_WARNING);
alertDao.addAlert(alert);
logger.info("add alert to db , alert : {}", alert);
} catch (Exception e) {
logger.error("send alert failed:{} ", e.getMessage());
}
}
/**
* send process instance alert
*
* @param processInstance process instance
* @param taskInstances task instance list
*/
public void sendAlertProcessInstance(ProcessInstance processInstance,
List<TaskInstance> taskInstances,
ProjectUser projectUser) {
if (!isNeedToSendWarning(processInstance)) {
return;
}
Alert alert = new Alert();
String cmdName = getCommandCnName(processInstance.getCommandType());
String success = processInstance.getState().typeIsSuccess() ? "success" : "failed"; |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,225 | [Feature][Alert] Add close alert method | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
Right now, we can set the alert when workflow executes failure.
In some cases, if the alert plugin is a third party service like [pagerduty](https://www.pagerduty.com), when we send an alert we will call the service to send an alert, and it may send 5 minutes unless we close the alert(We can close the alert by manual too).
But we may hope when the workflow success, the before alert can close automatically, so this issue is want to discuss add a new method closeAlert in alert plugin, we will provide a default method, the default method will do nothing, you can override this.
When the workflow execute success we will check if there exist a alert, and insert a close alert event, the AlertServer side will consume the event and execute the closeMethod.
### Use case
_No response_
### Related issues
_No response_
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11225 | https://github.com/apache/dolphinscheduler/pull/11221 | 01fc6c8a3a3aa68a45879ec5130250c99492a86b | 8e21c38c0041aa2689eb9d9bc703f67292c98e88 | 2022-08-01T03:59:56Z | java | 2022-08-01T05:04:27Z | dolphinscheduler-service/src/main/java/org/apache/dolphinscheduler/service/alert/ProcessAlertManager.java | alert.setTitle(cmdName + " " + success);
alert.setWarningType(processInstance.getState().typeIsSuccess() ? WarningType.SUCCESS : WarningType.FAILURE);
String content = getContentProcessInstance(processInstance, taskInstances, projectUser);
alert.setContent(content);
alert.setAlertGroupId(processInstance.getWarningGroupId());
alert.setCreateTime(new Date());
alert.setProjectCode(projectUser.getProjectCode());
alert.setProcessDefinitionCode(processInstance.getProcessDefinitionCode());
alert.setProcessInstanceId(processInstance.getId());
alert.setAlertType(processInstance.getState().typeIsSuccess() ? AlertType.PROCESS_INSTANCE_SUCCESS : AlertType.PROCESS_INSTANCE_FAILURE);
alertDao.addAlert(alert);
logger.info("add alert to db , alert: {}", alert);
}
/**
* check if need to be send warning
*
* @param processInstance
* @return
*/
public boolean isNeedToSendWarning(ProcessInstance processInstance) {
if (Flag.YES == processInstance.getIsSubProcess()) {
return false;
}
boolean sendWarning = false;
WarningType warningType = processInstance.getWarningType();
switch (warningType) {
case ALL:
if (processInstance.getState().typeIsFinished()) {
sendWarning = true;
} |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,225 | [Feature][Alert] Add close alert method | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
Right now, we can set the alert when workflow executes failure.
In some cases, if the alert plugin is a third party service like [pagerduty](https://www.pagerduty.com), when we send an alert we will call the service to send an alert, and it may send 5 minutes unless we close the alert(We can close the alert by manual too).
But we may hope when the workflow success, the before alert can close automatically, so this issue is want to discuss add a new method closeAlert in alert plugin, we will provide a default method, the default method will do nothing, you can override this.
When the workflow execute success we will check if there exist a alert, and insert a close alert event, the AlertServer side will consume the event and execute the closeMethod.
### Use case
_No response_
### Related issues
_No response_
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11225 | https://github.com/apache/dolphinscheduler/pull/11221 | 01fc6c8a3a3aa68a45879ec5130250c99492a86b | 8e21c38c0041aa2689eb9d9bc703f67292c98e88 | 2022-08-01T03:59:56Z | java | 2022-08-01T05:04:27Z | dolphinscheduler-service/src/main/java/org/apache/dolphinscheduler/service/alert/ProcessAlertManager.java | break;
case SUCCESS:
if (processInstance.getState().typeIsSuccess()) {
sendWarning = true;
}
break;
case FAILURE:
if (processInstance.getState().typeIsFailure()) {
sendWarning = true;
}
break;
default:
}
return sendWarning;
}
/**
* send process timeout alert
*
* @param processInstance process instance
* @param projectUser projectUser
*/
public void sendProcessTimeoutAlert(ProcessInstance processInstance, ProjectUser projectUser) {
alertDao.sendProcessTimeoutAlert(processInstance, projectUser);
}
/**
* send data quality task alert
*/
public void sendDataQualityTaskExecuteResultAlert(DqExecuteResult result, ProcessInstance processInstance) {
Alert alert = new Alert();
String state = DqTaskState.of(result.getState()).getDescription(); |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,225 | [Feature][Alert] Add close alert method | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
Right now, we can set the alert when workflow executes failure.
In some cases, if the alert plugin is a third party service like [pagerduty](https://www.pagerduty.com), when we send an alert we will call the service to send an alert, and it may send 5 minutes unless we close the alert(We can close the alert by manual too).
But we may hope when the workflow success, the before alert can close automatically, so this issue is want to discuss add a new method closeAlert in alert plugin, we will provide a default method, the default method will do nothing, you can override this.
When the workflow execute success we will check if there exist a alert, and insert a close alert event, the AlertServer side will consume the event and execute the closeMethod.
### Use case
_No response_
### Related issues
_No response_
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11225 | https://github.com/apache/dolphinscheduler/pull/11221 | 01fc6c8a3a3aa68a45879ec5130250c99492a86b | 8e21c38c0041aa2689eb9d9bc703f67292c98e88 | 2022-08-01T03:59:56Z | java | 2022-08-01T05:04:27Z | dolphinscheduler-service/src/main/java/org/apache/dolphinscheduler/service/alert/ProcessAlertManager.java | alert.setTitle("DataQualityResult [" + result.getTaskName() + "] " + state);
String content = getDataQualityAlterContent(result);
alert.setContent(content);
alert.setAlertGroupId(processInstance.getWarningGroupId());
alert.setCreateTime(new Date());
alert.setProjectCode(result.getProjectCode());
alert.setProcessDefinitionCode(processInstance.getProcessDefinitionCode());
alert.setProcessInstanceId(processInstance.getId());
alert.setAlertType(processInstance.getState().typeIsSuccess() ? AlertType.PROCESS_INSTANCE_SUCCESS : AlertType.PROCESS_INSTANCE_FAILURE);
alertDao.addAlert(alert);
logger.info("add alert to db , alert: {}", alert);
}
/**
* send data quality task error alert
*/
public void sendTaskErrorAlert(TaskInstance taskInstance,ProcessInstance processInstance) {
Alert alert = new Alert();
alert.setTitle("Task [" + taskInstance.getName() + "] Failure Warning");
String content = getTaskAlterContent(taskInstance);
alert.setContent(content);
alert.setAlertGroupId(processInstance.getWarningGroupId());
alert.setCreateTime(new Date());
alert.setProcessDefinitionCode(processInstance.getProcessDefinitionCode());
alert.setProcessInstanceId(processInstance.getId());
alert.setAlertType(AlertType.TASK_FAILURE);
alertDao.addAlert(alert);
logger.info("add alert to db , alert: {}", alert);
}
/** |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,225 | [Feature][Alert] Add close alert method | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
Right now, we can set the alert when workflow executes failure.
In some cases, if the alert plugin is a third party service like [pagerduty](https://www.pagerduty.com), when we send an alert we will call the service to send an alert, and it may send 5 minutes unless we close the alert(We can close the alert by manual too).
But we may hope when the workflow success, the before alert can close automatically, so this issue is want to discuss add a new method closeAlert in alert plugin, we will provide a default method, the default method will do nothing, you can override this.
When the workflow execute success we will check if there exist a alert, and insert a close alert event, the AlertServer side will consume the event and execute the closeMethod.
### Use case
_No response_
### Related issues
_No response_
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11225 | https://github.com/apache/dolphinscheduler/pull/11221 | 01fc6c8a3a3aa68a45879ec5130250c99492a86b | 8e21c38c0041aa2689eb9d9bc703f67292c98e88 | 2022-08-01T03:59:56Z | java | 2022-08-01T05:04:27Z | dolphinscheduler-service/src/main/java/org/apache/dolphinscheduler/service/alert/ProcessAlertManager.java | * getDataQualityAlterContent
* @param result DqExecuteResult
* @return String String
*/
public String getDataQualityAlterContent(DqExecuteResult result) {
DqExecuteResultAlertContent content = DqExecuteResultAlertContent.newBuilder()
.processDefinitionId(result.getProcessDefinitionId())
.processDefinitionName(result.getProcessDefinitionName())
.processInstanceId(result.getProcessInstanceId())
.processInstanceName(result.getProcessInstanceName())
.taskInstanceId(result.getTaskInstanceId())
.taskName(result.getTaskName())
.ruleType(result.getRuleType())
.ruleName(result.getRuleName())
.statisticsValue(result.getStatisticsValue())
.comparisonValue(result.getComparisonValue())
.checkType(result.getCheckType())
.threshold(result.getThreshold())
.operator(result.getOperator())
.failureStrategy(result.getFailureStrategy())
.userId(result.getUserId())
.userName(result.getUserName())
.state(result.getState())
.errorDataPath(result.getErrorOutputPath())
.build();
return JSONUtils.toJsonString(content);
}
/**
* getTaskAlterContent
* @param taskInstance TaskInstance |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,225 | [Feature][Alert] Add close alert method | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
Right now, we can set the alert when workflow executes failure.
In some cases, if the alert plugin is a third party service like [pagerduty](https://www.pagerduty.com), when we send an alert we will call the service to send an alert, and it may send 5 minutes unless we close the alert(We can close the alert by manual too).
But we may hope when the workflow success, the before alert can close automatically, so this issue is want to discuss add a new method closeAlert in alert plugin, we will provide a default method, the default method will do nothing, you can override this.
When the workflow execute success we will check if there exist a alert, and insert a close alert event, the AlertServer side will consume the event and execute the closeMethod.
### Use case
_No response_
### Related issues
_No response_
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11225 | https://github.com/apache/dolphinscheduler/pull/11221 | 01fc6c8a3a3aa68a45879ec5130250c99492a86b | 8e21c38c0041aa2689eb9d9bc703f67292c98e88 | 2022-08-01T03:59:56Z | java | 2022-08-01T05:04:27Z | dolphinscheduler-service/src/main/java/org/apache/dolphinscheduler/service/alert/ProcessAlertManager.java | * @return String String
*/
public String getTaskAlterContent(TaskInstance taskInstance) {
TaskAlertContent content = TaskAlertContent.newBuilder()
.processInstanceName(taskInstance.getProcessInstanceName())
.processInstanceId(taskInstance.getProcessInstanceId())
.taskInstanceId(taskInstance.getId())
.taskName(taskInstance.getName())
.taskType(taskInstance.getTaskType())
.state(taskInstance.getState())
.startTime(taskInstance.getStartTime())
.endTime(taskInstance.getEndTime())
.host(taskInstance.getHost())
.logPath(taskInstance.getLogPath())
.build();
return JSONUtils.toJsonString(content);
}
public void sendTaskTimeoutAlert(ProcessInstance processInstance, TaskInstance taskInstance, ProjectUser projectUser) {
alertDao.sendTaskTimeoutAlert(processInstance, taskInstance, projectUser);
}
/**
*
* check node type and process blocking flag, then insert a block record into db
*
* @param processInstance process instance
* @param projectUser the project owner
*/
public void sendProcessBlockingAlert(ProcessInstance processInstance,
ProjectUser projectUser) {
Alert alert = new Alert(); |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,225 | [Feature][Alert] Add close alert method | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
Right now, we can set the alert when workflow executes failure.
In some cases, if the alert plugin is a third party service like [pagerduty](https://www.pagerduty.com), when we send an alert we will call the service to send an alert, and it may send 5 minutes unless we close the alert(We can close the alert by manual too).
But we may hope when the workflow success, the before alert can close automatically, so this issue is want to discuss add a new method closeAlert in alert plugin, we will provide a default method, the default method will do nothing, you can override this.
When the workflow execute success we will check if there exist a alert, and insert a close alert event, the AlertServer side will consume the event and execute the closeMethod.
### Use case
_No response_
### Related issues
_No response_
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11225 | https://github.com/apache/dolphinscheduler/pull/11221 | 01fc6c8a3a3aa68a45879ec5130250c99492a86b | 8e21c38c0041aa2689eb9d9bc703f67292c98e88 | 2022-08-01T03:59:56Z | java | 2022-08-01T05:04:27Z | dolphinscheduler-service/src/main/java/org/apache/dolphinscheduler/service/alert/ProcessAlertManager.java | String cmdName = getCommandCnName(processInstance.getCommandType());
List<ProcessAlertContent> blockingNodeList = new ArrayList<>(1);
ProcessAlertContent processAlertContent = ProcessAlertContent.newBuilder()
.projectCode(projectUser.getProjectCode())
.projectName(projectUser.getProjectName())
.owner(projectUser.getUserName())
.processId(processInstance.getId())
.processName(processInstance.getName())
.processType(processInstance.getCommandType())
.processState(processInstance.getState())
.runTimes(processInstance.getRunTimes())
.processStartTime(processInstance.getStartTime())
.processEndTime(processInstance.getEndTime())
.processHost(processInstance.getHost())
.build();
blockingNodeList.add(processAlertContent);
String content = JSONUtils.toJsonString(blockingNodeList);
alert.setTitle(cmdName + " Blocked");
alert.setContent(content);
alert.setAlertGroupId(processInstance.getWarningGroupId());
alert.setCreateTime(new Date());
alert.setProjectCode(projectUser.getProjectCode());
alert.setProcessDefinitionCode(processInstance.getProcessDefinitionCode());
alert.setProcessInstanceId(processInstance.getId());
alert.setAlertType(AlertType.PROCESS_INSTANCE_BLOCKED);
alertDao.addAlert(alert);
logger.info("add alert to db, alert: {}",alert);
}
} |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,214 | [Improvement][Worker] Avoid worker oom when you find appid from task logs | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
We deploy app on yarn by spark or flink , if you need to check app status on yarn , you must know appid. In dolphin, the appid is from worker's task logs. when task logs's size is very large, worker's memory will occuted by task logs.
eg...
<img width="766" alt="98b6e379d63b718e6aef3612eba1b79" src="https://user-images.githubusercontent.com/20475112/182008581-662eae4e-421b-4986-86ca-59b1b5081143.png">
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11214 | https://github.com/apache/dolphinscheduler/pull/11224 | 0141aa511b6077fca610d6964b4727bfabeda52a | 9a53c6ac3430c9edb1ded218d5a0190214fadc80 | 2022-07-31T03:29:28Z | java | 2022-08-01T07:28:06Z | dolphinscheduler-task-plugin/dolphinscheduler-task-api/src/main/java/org/apache/dolphinscheduler/plugin/task/api/AbstractCommandExecutor.java | /*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package org.apache.dolphinscheduler.plugin.task.api;
import static org.apache.dolphinscheduler.plugin.task.api.TaskConstants.EXIT_CODE_FAILURE;
import static org.apache.dolphinscheduler.plugin.task.api.TaskConstants.EXIT_CODE_KILL;
import org.apache.dolphinscheduler.plugin.task.api.model.TaskResponse; |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,214 | [Improvement][Worker] Avoid worker oom when you find appid from task logs | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
We deploy app on yarn by spark or flink , if you need to check app status on yarn , you must know appid. In dolphin, the appid is from worker's task logs. when task logs's size is very large, worker's memory will occuted by task logs.
eg...
<img width="766" alt="98b6e379d63b718e6aef3612eba1b79" src="https://user-images.githubusercontent.com/20475112/182008581-662eae4e-421b-4986-86ca-59b1b5081143.png">
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11214 | https://github.com/apache/dolphinscheduler/pull/11224 | 0141aa511b6077fca610d6964b4727bfabeda52a | 9a53c6ac3430c9edb1ded218d5a0190214fadc80 | 2022-07-31T03:29:28Z | java | 2022-08-01T07:28:06Z | dolphinscheduler-task-plugin/dolphinscheduler-task-api/src/main/java/org/apache/dolphinscheduler/plugin/task/api/AbstractCommandExecutor.java | import org.apache.dolphinscheduler.plugin.task.api.utils.AbstractCommandExecutorConstants;
import org.apache.dolphinscheduler.plugin.task.api.utils.OSUtils;
import org.apache.dolphinscheduler.spi.utils.PropertyUtils;
import org.apache.dolphinscheduler.spi.utils.StringUtils;
import org.apache.commons.lang3.SystemUtils;
import java.io.BufferedReader;
import java.io.File;
import java.io.FileInputStream;
import java.io.IOException;
import java.io.InputStreamReader;
import java.lang.reflect.Field;
import java.nio.charset.StandardCharsets;
import java.util.ArrayList;
import java.util.Collections;
import java.util.LinkedList;
import java.util.List;
import java.util.concurrent.ExecutorService;
import java.util.concurrent.Executors;
import java.util.concurrent.LinkedBlockingQueue;
import java.util.concurrent.ThreadFactory;
import java.util.concurrent.TimeUnit;
import java.util.function.Consumer;
import java.util.regex.Matcher;
import java.util.regex.Pattern;
import org.slf4j.Logger;
import com.google.common.util.concurrent.ThreadFactoryBuilder;
/**
* abstract command executor
*/
public abstract class AbstractCommandExecutor { |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,214 | [Improvement][Worker] Avoid worker oom when you find appid from task logs | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
We deploy app on yarn by spark or flink , if you need to check app status on yarn , you must know appid. In dolphin, the appid is from worker's task logs. when task logs's size is very large, worker's memory will occuted by task logs.
eg...
<img width="766" alt="98b6e379d63b718e6aef3612eba1b79" src="https://user-images.githubusercontent.com/20475112/182008581-662eae4e-421b-4986-86ca-59b1b5081143.png">
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11214 | https://github.com/apache/dolphinscheduler/pull/11224 | 0141aa511b6077fca610d6964b4727bfabeda52a | 9a53c6ac3430c9edb1ded218d5a0190214fadc80 | 2022-07-31T03:29:28Z | java | 2022-08-01T07:28:06Z | dolphinscheduler-task-plugin/dolphinscheduler-task-api/src/main/java/org/apache/dolphinscheduler/plugin/task/api/AbstractCommandExecutor.java | /**
* rules for extracting application ID
*/
protected static final Pattern APPLICATION_REGEX = Pattern.compile(TaskConstants.APPLICATION_REGEX);
/**
* rules for extracting Var Pool
*/
protected static final Pattern SETVALUE_REGEX = Pattern.compile(TaskConstants.SETVALUE_REGEX);
protected StringBuilder varPool = new StringBuilder();
/**
* process
*/
private Process process;
/**
* log handler
*/
protected Consumer<LinkedBlockingQueue<String>> logHandler;
/**
* logger
*/
protected Logger logger;
/** |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,214 | [Improvement][Worker] Avoid worker oom when you find appid from task logs | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
We deploy app on yarn by spark or flink , if you need to check app status on yarn , you must know appid. In dolphin, the appid is from worker's task logs. when task logs's size is very large, worker's memory will occuted by task logs.
eg...
<img width="766" alt="98b6e379d63b718e6aef3612eba1b79" src="https://user-images.githubusercontent.com/20475112/182008581-662eae4e-421b-4986-86ca-59b1b5081143.png">
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11214 | https://github.com/apache/dolphinscheduler/pull/11224 | 0141aa511b6077fca610d6964b4727bfabeda52a | 9a53c6ac3430c9edb1ded218d5a0190214fadc80 | 2022-07-31T03:29:28Z | java | 2022-08-01T07:28:06Z | dolphinscheduler-task-plugin/dolphinscheduler-task-api/src/main/java/org/apache/dolphinscheduler/plugin/task/api/AbstractCommandExecutor.java | * log list
*/
protected LinkedBlockingQueue<String> logBuffer;
protected boolean logOutputIsSuccess = false;
/*
* SHELL result string
*/
protected String taskResultString;
/**
* taskRequest
*/
protected TaskExecutionContext taskRequest;
public AbstractCommandExecutor(Consumer<LinkedBlockingQueue<String>> logHandler,
TaskExecutionContext taskRequest,
Logger logger) {
this.logHandler = logHandler;
this.taskRequest = taskRequest;
this.logger = logger;
this.logBuffer = new LinkedBlockingQueue<>();
}
public AbstractCommandExecutor(LinkedBlockingQueue<String> logBuffer) {
this.logBuffer = logBuffer;
}
/**
* build process
*
* @param commandFile command file
* @throws IOException IO Exception
*/
private void buildProcess(String commandFile) throws IOException { |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,214 | [Improvement][Worker] Avoid worker oom when you find appid from task logs | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
We deploy app on yarn by spark or flink , if you need to check app status on yarn , you must know appid. In dolphin, the appid is from worker's task logs. when task logs's size is very large, worker's memory will occuted by task logs.
eg...
<img width="766" alt="98b6e379d63b718e6aef3612eba1b79" src="https://user-images.githubusercontent.com/20475112/182008581-662eae4e-421b-4986-86ca-59b1b5081143.png">
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11214 | https://github.com/apache/dolphinscheduler/pull/11224 | 0141aa511b6077fca610d6964b4727bfabeda52a | 9a53c6ac3430c9edb1ded218d5a0190214fadc80 | 2022-07-31T03:29:28Z | java | 2022-08-01T07:28:06Z | dolphinscheduler-task-plugin/dolphinscheduler-task-api/src/main/java/org/apache/dolphinscheduler/plugin/task/api/AbstractCommandExecutor.java | List<String> command = new LinkedList<>();
ProcessBuilder processBuilder = new ProcessBuilder();
processBuilder.directory(new File(taskRequest.getExecutePath()));
processBuilder.redirectErrorStream(true);
if (OSUtils.isSudoEnable()) {
if (SystemUtils.IS_OS_LINUX && PropertyUtils.getBoolean(AbstractCommandExecutorConstants.TASK_RESOURCE_LIMIT_STATE)) {
generateCgroupCommand(command);
} else {
command.add("sudo");
command.add("-u");
command.add(taskRequest.getTenantCode());
}
}
command.add(commandInterpreter());
command.addAll(Collections.emptyList());
command.add(commandFile);
processBuilder.command(command);
process = processBuilder.start();
printCommand(command);
}
/**
* generate systemd command.
* eg: sudo systemd-run -q --scope -p CPUQuota=100% -p MemoryMax=200M --uid=root
* @param command command |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,214 | [Improvement][Worker] Avoid worker oom when you find appid from task logs | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
We deploy app on yarn by spark or flink , if you need to check app status on yarn , you must know appid. In dolphin, the appid is from worker's task logs. when task logs's size is very large, worker's memory will occuted by task logs.
eg...
<img width="766" alt="98b6e379d63b718e6aef3612eba1b79" src="https://user-images.githubusercontent.com/20475112/182008581-662eae4e-421b-4986-86ca-59b1b5081143.png">
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11214 | https://github.com/apache/dolphinscheduler/pull/11224 | 0141aa511b6077fca610d6964b4727bfabeda52a | 9a53c6ac3430c9edb1ded218d5a0190214fadc80 | 2022-07-31T03:29:28Z | java | 2022-08-01T07:28:06Z | dolphinscheduler-task-plugin/dolphinscheduler-task-api/src/main/java/org/apache/dolphinscheduler/plugin/task/api/AbstractCommandExecutor.java | */
private void generateCgroupCommand(List<String> command) {
Integer cpuQuota = taskRequest.getCpuQuota();
Integer memoryMax = taskRequest.getMemoryMax();
command.add("sudo");
command.add("systemd-run");
command.add("-q");
command.add("--scope");
if (cpuQuota == -1) {
command.add("-p");
command.add("CPUQuota=");
} else {
command.add("-p");
command.add(String.format("CPUQuota=%s%%", taskRequest.getCpuQuota()));
}
if (memoryMax == -1) {
command.add("-p");
command.add(String.format("MemoryMax=%s", "infinity"));
} else {
command.add("-p");
command.add(String.format("MemoryMax=%sM", taskRequest.getMemoryMax()));
}
command.add(String.format("--uid=%s", taskRequest.getTenantCode()));
}
public TaskResponse run(String execCommand) throws IOException, InterruptedException {
TaskResponse result = new TaskResponse();
int taskInstanceId = taskRequest.getTaskInstanceId();
if (null == TaskExecutionContextCacheManager.getByTaskInstanceId(taskInstanceId)) {
result.setExitStatusCode(EXIT_CODE_KILL);
return result; |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,214 | [Improvement][Worker] Avoid worker oom when you find appid from task logs | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
We deploy app on yarn by spark or flink , if you need to check app status on yarn , you must know appid. In dolphin, the appid is from worker's task logs. when task logs's size is very large, worker's memory will occuted by task logs.
eg...
<img width="766" alt="98b6e379d63b718e6aef3612eba1b79" src="https://user-images.githubusercontent.com/20475112/182008581-662eae4e-421b-4986-86ca-59b1b5081143.png">
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11214 | https://github.com/apache/dolphinscheduler/pull/11224 | 0141aa511b6077fca610d6964b4727bfabeda52a | 9a53c6ac3430c9edb1ded218d5a0190214fadc80 | 2022-07-31T03:29:28Z | java | 2022-08-01T07:28:06Z | dolphinscheduler-task-plugin/dolphinscheduler-task-api/src/main/java/org/apache/dolphinscheduler/plugin/task/api/AbstractCommandExecutor.java | }
if (StringUtils.isEmpty(execCommand)) {
TaskExecutionContextCacheManager.removeByTaskInstanceId(taskInstanceId);
return result;
}
String commandFilePath = buildCommandFilePath();
createCommandFileIfNotExists(execCommand, commandFilePath);
buildProcess(commandFilePath);
parseProcessOutput(process);
int processId = getProcessId(process);
result.setProcessId(processId);
taskRequest.setProcessId(processId);
boolean updateTaskExecutionContextStatus = TaskExecutionContextCacheManager.updateTaskExecutionContext(taskRequest);
if (Boolean.FALSE.equals(updateTaskExecutionContextStatus)) {
ProcessUtils.kill(taskRequest);
result.setExitStatusCode(EXIT_CODE_KILL);
return result;
}
logger.info("process start, process id is: {}", processId);
long remainTime = getRemainTime();
boolean status = process.waitFor(remainTime, TimeUnit.SECONDS);
if (status) { |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,214 | [Improvement][Worker] Avoid worker oom when you find appid from task logs | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
We deploy app on yarn by spark or flink , if you need to check app status on yarn , you must know appid. In dolphin, the appid is from worker's task logs. when task logs's size is very large, worker's memory will occuted by task logs.
eg...
<img width="766" alt="98b6e379d63b718e6aef3612eba1b79" src="https://user-images.githubusercontent.com/20475112/182008581-662eae4e-421b-4986-86ca-59b1b5081143.png">
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11214 | https://github.com/apache/dolphinscheduler/pull/11224 | 0141aa511b6077fca610d6964b4727bfabeda52a | 9a53c6ac3430c9edb1ded218d5a0190214fadc80 | 2022-07-31T03:29:28Z | java | 2022-08-01T07:28:06Z | dolphinscheduler-task-plugin/dolphinscheduler-task-api/src/main/java/org/apache/dolphinscheduler/plugin/task/api/AbstractCommandExecutor.java | List<String> appIds = getAppIds(taskRequest.getLogPath());
result.setAppIds(String.join(TaskConstants.COMMA, appIds));
result.setExitStatusCode(process.exitValue());
} else {
logger.error("process has failure , exitStatusCode:{}, processExitValue:{}, ready to kill ...",
result.getExitStatusCode(), process.exitValue());
ProcessUtils.kill(taskRequest);
result.setExitStatusCode(EXIT_CODE_FAILURE);
}
logger.info("process has exited, execute path:{}, processId:{} ,exitStatusCode:{} ,processWaitForStatus:{} ,processExitValue:{}",
taskRequest.getExecutePath(), processId, result.getExitStatusCode(), status, process.exitValue());
return result;
}
public String getVarPool() {
return varPool.toString();
}
/**
* cancel application
*
* @throws Exception exception
*/
public void cancelApplication() throws Exception {
if (process == null) {
return;
}
clear();
int processId = getProcessId(process); |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,214 | [Improvement][Worker] Avoid worker oom when you find appid from task logs | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
We deploy app on yarn by spark or flink , if you need to check app status on yarn , you must know appid. In dolphin, the appid is from worker's task logs. when task logs's size is very large, worker's memory will occuted by task logs.
eg...
<img width="766" alt="98b6e379d63b718e6aef3612eba1b79" src="https://user-images.githubusercontent.com/20475112/182008581-662eae4e-421b-4986-86ca-59b1b5081143.png">
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11214 | https://github.com/apache/dolphinscheduler/pull/11224 | 0141aa511b6077fca610d6964b4727bfabeda52a | 9a53c6ac3430c9edb1ded218d5a0190214fadc80 | 2022-07-31T03:29:28Z | java | 2022-08-01T07:28:06Z | dolphinscheduler-task-plugin/dolphinscheduler-task-api/src/main/java/org/apache/dolphinscheduler/plugin/task/api/AbstractCommandExecutor.java | logger.info("cancel process: {}", processId);
boolean killed = softKill(processId);
if (!killed) {
hardKill(processId);
process.destroy();
process = null;
}
}
/**
* soft kill
*
* @param processId process id
* @return process is alive
*/
private boolean softKill(int processId) {
if (processId != 0 && process.isAlive()) {
try {
String cmd = String.format("kill %d", processId);
cmd = OSUtils.getSudoCmd(taskRequest.getTenantCode(), cmd);
logger.info("soft kill task:{}, process id:{}, cmd:{}", taskRequest.getTaskAppId(), processId, cmd);
Runtime.getRuntime().exec(cmd);
} catch (IOException e) {
logger.info("kill attempt failed", e);
}
}
return process.isAlive(); |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,214 | [Improvement][Worker] Avoid worker oom when you find appid from task logs | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
We deploy app on yarn by spark or flink , if you need to check app status on yarn , you must know appid. In dolphin, the appid is from worker's task logs. when task logs's size is very large, worker's memory will occuted by task logs.
eg...
<img width="766" alt="98b6e379d63b718e6aef3612eba1b79" src="https://user-images.githubusercontent.com/20475112/182008581-662eae4e-421b-4986-86ca-59b1b5081143.png">
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11214 | https://github.com/apache/dolphinscheduler/pull/11224 | 0141aa511b6077fca610d6964b4727bfabeda52a | 9a53c6ac3430c9edb1ded218d5a0190214fadc80 | 2022-07-31T03:29:28Z | java | 2022-08-01T07:28:06Z | dolphinscheduler-task-plugin/dolphinscheduler-task-api/src/main/java/org/apache/dolphinscheduler/plugin/task/api/AbstractCommandExecutor.java | }
/**
* hard kill
*
* @param processId process id
*/
private void hardKill(int processId) {
if (processId != 0 && process.isAlive()) {
try {
String cmd = String.format("kill -9 %d", processId);
cmd = OSUtils.getSudoCmd(taskRequest.getTenantCode(), cmd);
logger.info("hard kill task:{}, process id:{}, cmd:{}", taskRequest.getTaskAppId(), processId, cmd);
Runtime.getRuntime().exec(cmd);
} catch (IOException e) {
logger.error("kill attempt failed ", e);
}
}
}
private void printCommand(List<String> commands) {
logger.info("task run command: {}", String.join(" ", commands));
}
/**
* clear
*/
private void clear() {
LinkedBlockingQueue<String> markerLog = new LinkedBlockingQueue<>(1);
markerLog.add(ch.qos.logback.classic.ClassicConstants.FINALIZE_SESSION_MARKER.toString());
if (!logBuffer.isEmpty()) {
logHandler.accept(logBuffer); |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,214 | [Improvement][Worker] Avoid worker oom when you find appid from task logs | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
We deploy app on yarn by spark or flink , if you need to check app status on yarn , you must know appid. In dolphin, the appid is from worker's task logs. when task logs's size is very large, worker's memory will occuted by task logs.
eg...
<img width="766" alt="98b6e379d63b718e6aef3612eba1b79" src="https://user-images.githubusercontent.com/20475112/182008581-662eae4e-421b-4986-86ca-59b1b5081143.png">
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11214 | https://github.com/apache/dolphinscheduler/pull/11224 | 0141aa511b6077fca610d6964b4727bfabeda52a | 9a53c6ac3430c9edb1ded218d5a0190214fadc80 | 2022-07-31T03:29:28Z | java | 2022-08-01T07:28:06Z | dolphinscheduler-task-plugin/dolphinscheduler-task-api/src/main/java/org/apache/dolphinscheduler/plugin/task/api/AbstractCommandExecutor.java | logBuffer.clear();
}
logHandler.accept(markerLog);
}
/**
* get the standard output of the process
*
* @param process process
*/
private void parseProcessOutput(Process process) {
String threadLoggerInfoName = taskRequest.getTaskLogName();
ExecutorService getOutputLogService = newDaemonSingleThreadExecutor(threadLoggerInfoName);
getOutputLogService.submit(() -> {
try (BufferedReader inReader = new BufferedReader(new InputStreamReader(process.getInputStream()))) {
String line;
while ((line = inReader.readLine()) != null) {
if (line.startsWith("${setValue(") || line.startsWith("#{setValue(")) {
varPool.append(findVarPool(line));
varPool.append("$VarPool$");
} else {
logBuffer.add(line);
taskResultString = line;
}
}
logOutputIsSuccess = true;
} catch (Exception e) {
logger.error(e.getMessage(), e);
logOutputIsSuccess = true;
}
}); |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,214 | [Improvement][Worker] Avoid worker oom when you find appid from task logs | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
We deploy app on yarn by spark or flink , if you need to check app status on yarn , you must know appid. In dolphin, the appid is from worker's task logs. when task logs's size is very large, worker's memory will occuted by task logs.
eg...
<img width="766" alt="98b6e379d63b718e6aef3612eba1b79" src="https://user-images.githubusercontent.com/20475112/182008581-662eae4e-421b-4986-86ca-59b1b5081143.png">
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11214 | https://github.com/apache/dolphinscheduler/pull/11224 | 0141aa511b6077fca610d6964b4727bfabeda52a | 9a53c6ac3430c9edb1ded218d5a0190214fadc80 | 2022-07-31T03:29:28Z | java | 2022-08-01T07:28:06Z | dolphinscheduler-task-plugin/dolphinscheduler-task-api/src/main/java/org/apache/dolphinscheduler/plugin/task/api/AbstractCommandExecutor.java | getOutputLogService.shutdown();
ExecutorService parseProcessOutputExecutorService = newDaemonSingleThreadExecutor(threadLoggerInfoName);
parseProcessOutputExecutorService.submit(() -> {
try {
long lastFlushTime = System.currentTimeMillis();
while (logBuffer.size() > 0 || !logOutputIsSuccess) {
if (logBuffer.size() > 0) {
lastFlushTime = flush(lastFlushTime);
} else {
Thread.sleep(TaskConstants.DEFAULT_LOG_FLUSH_INTERVAL);
}
}
} catch (Exception e) {
Thread.currentThread().interrupt();
logger.error(e.getMessage(), e);
} finally {
clear();
}
});
parseProcessOutputExecutorService.shutdown();
}
/**
* get app links
*
* @param logPath log path
* @return app id list
*/
private List<String> getAppIds(String logPath) {
List<String> logs = convertFile2List(logPath);
List<String> appIds = new ArrayList<>(); |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,214 | [Improvement][Worker] Avoid worker oom when you find appid from task logs | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
We deploy app on yarn by spark or flink , if you need to check app status on yarn , you must know appid. In dolphin, the appid is from worker's task logs. when task logs's size is very large, worker's memory will occuted by task logs.
eg...
<img width="766" alt="98b6e379d63b718e6aef3612eba1b79" src="https://user-images.githubusercontent.com/20475112/182008581-662eae4e-421b-4986-86ca-59b1b5081143.png">
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11214 | https://github.com/apache/dolphinscheduler/pull/11224 | 0141aa511b6077fca610d6964b4727bfabeda52a | 9a53c6ac3430c9edb1ded218d5a0190214fadc80 | 2022-07-31T03:29:28Z | java | 2022-08-01T07:28:06Z | dolphinscheduler-task-plugin/dolphinscheduler-task-api/src/main/java/org/apache/dolphinscheduler/plugin/task/api/AbstractCommandExecutor.java | /*
* analysis log?get submited yarn application id
*/
for (String log : logs) {
String appId = findAppId(log);
if (StringUtils.isNotEmpty(appId) && !appIds.contains(appId)) {
logger.info("find app id: {}", appId);
appIds.add(appId);
}
}
return appIds;
}
/**
* convert file to list
*
* @param filename file name
* @return line list
*/
private List<String> convertFile2List(String filename) {
List<String> lineList = new ArrayList<>(100);
File file = new File(filename);
if (!file.exists()) {
return lineList;
}
try (BufferedReader br = new BufferedReader(new InputStreamReader(new FileInputStream(filename), StandardCharsets.UTF_8))) {
String line;
while ((line = br.readLine()) != null) {
lineList.add(line);
}
} catch (Exception e) { |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,214 | [Improvement][Worker] Avoid worker oom when you find appid from task logs | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
We deploy app on yarn by spark or flink , if you need to check app status on yarn , you must know appid. In dolphin, the appid is from worker's task logs. when task logs's size is very large, worker's memory will occuted by task logs.
eg...
<img width="766" alt="98b6e379d63b718e6aef3612eba1b79" src="https://user-images.githubusercontent.com/20475112/182008581-662eae4e-421b-4986-86ca-59b1b5081143.png">
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11214 | https://github.com/apache/dolphinscheduler/pull/11224 | 0141aa511b6077fca610d6964b4727bfabeda52a | 9a53c6ac3430c9edb1ded218d5a0190214fadc80 | 2022-07-31T03:29:28Z | java | 2022-08-01T07:28:06Z | dolphinscheduler-task-plugin/dolphinscheduler-task-api/src/main/java/org/apache/dolphinscheduler/plugin/task/api/AbstractCommandExecutor.java | logger.error(String.format("read file: %s failed : ", filename), e);
}
return lineList;
}
/**
* find var pool
* @param line
* @return
*/
private String findVarPool(String line) {
Matcher matcher = SETVALUE_REGEX.matcher(line);
if (matcher.find()) {
return matcher.group(1);
}
return null;
}
/**
* find app id
*
* @param line line
* @return appid
*/
private String findAppId(String line) {
Matcher matcher = APPLICATION_REGEX.matcher(line);
if (matcher.find()) {
return matcher.group();
}
return null;
} |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,214 | [Improvement][Worker] Avoid worker oom when you find appid from task logs | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
We deploy app on yarn by spark or flink , if you need to check app status on yarn , you must know appid. In dolphin, the appid is from worker's task logs. when task logs's size is very large, worker's memory will occuted by task logs.
eg...
<img width="766" alt="98b6e379d63b718e6aef3612eba1b79" src="https://user-images.githubusercontent.com/20475112/182008581-662eae4e-421b-4986-86ca-59b1b5081143.png">
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11214 | https://github.com/apache/dolphinscheduler/pull/11224 | 0141aa511b6077fca610d6964b4727bfabeda52a | 9a53c6ac3430c9edb1ded218d5a0190214fadc80 | 2022-07-31T03:29:28Z | java | 2022-08-01T07:28:06Z | dolphinscheduler-task-plugin/dolphinscheduler-task-api/src/main/java/org/apache/dolphinscheduler/plugin/task/api/AbstractCommandExecutor.java | /**
* get remain time（s）
*
* @return remain time
*/
private long getRemainTime() {
long usedTime = (System.currentTimeMillis() - taskRequest.getStartTime().getTime()) / 1000;
long remainTime = taskRequest.getTaskTimeout() - usedTime;
if (remainTime < 0) {
throw new RuntimeException("task execution time out");
}
return remainTime;
}
/**
* get process id
*
* @param process process
* @return process id
*/
private int getProcessId(Process process) {
int processId = 0;
try {
Field f = process.getClass().getDeclaredField(TaskConstants.PID);
f.setAccessible(true);
processId = f.getInt(process);
} catch (Throwable e) {
logger.error(e.getMessage(), e);
}
return processId;
} |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,214 | [Improvement][Worker] Avoid worker oom when you find appid from task logs | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar feature requirement.
### Description
We deploy app on yarn by spark or flink , if you need to check app status on yarn , you must know appid. In dolphin, the appid is from worker's task logs. when task logs's size is very large, worker's memory will occuted by task logs.
eg...
<img width="766" alt="98b6e379d63b718e6aef3612eba1b79" src="https://user-images.githubusercontent.com/20475112/182008581-662eae4e-421b-4986-86ca-59b1b5081143.png">
### Are you willing to submit a PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11214 | https://github.com/apache/dolphinscheduler/pull/11224 | 0141aa511b6077fca610d6964b4727bfabeda52a | 9a53c6ac3430c9edb1ded218d5a0190214fadc80 | 2022-07-31T03:29:28Z | java | 2022-08-01T07:28:06Z | dolphinscheduler-task-plugin/dolphinscheduler-task-api/src/main/java/org/apache/dolphinscheduler/plugin/task/api/AbstractCommandExecutor.java | /**
* when log buffer siz or flush time reach condition , then flush
*
* @param lastFlushTime last flush time
* @return last flush time
*/
private long flush(long lastFlushTime) {
long now = System.currentTimeMillis();
/*
* when log buffer siz or flush time reach condition , then flush
*/
if (logBuffer.size() >= TaskConstants.DEFAULT_LOG_ROWS_NUM || now - lastFlushTime > TaskConstants.DEFAULT_LOG_FLUSH_INTERVAL) {
lastFlushTime = now;
logHandler.accept(logBuffer);
logBuffer.clear();
}
return lastFlushTime;
}
protected abstract String buildCommandFilePath();
protected abstract void createCommandFileIfNotExists(String execCommand, String commandFile) throws IOException;
ExecutorService newDaemonSingleThreadExecutor(String threadName) {
ThreadFactory threadFactory = new ThreadFactoryBuilder()
.setDaemon(true)
.setNameFormat(threadName)
.build();
return Executors.newSingleThreadExecutor(threadFactory);
}
protected abstract String commandInterpreter();
} |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | /*
* Lcensed to the Apache Software Foundaton (ASF) under one or more
* contrbutor lcense agreements. See the NOTICE fle dstrbuted wth
* ths work for addtonal nformaton regardng copyrght ownershp.
* The ASF lcenses ths fle to You under the Apache Lcense, Verson 2.0
* (the "Lcense"); you may not use ths fle except n complance wth
* the Lcense. You may obtan a copy of the Lcense at
*
* http://www.apache.org/lcenses/LICENSE-2.0
*
* Unless requred by applcable law or agreed to n wrtng, software
* dstrbuted under the Lcense s dstrbuted on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, ether express or mpled.
* See the Lcense for the specfc language governng permssons and
* lmtatons under the Lcense.
*/
package org.apache.dolphnscheduler.server.master.runner;
mport statc org.apache.dolphnscheduler.common.Constants.CMDPARAM_COMPLEMENT_DATA_END_DATE; |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | mport statc org.apache.dolphnscheduler.common.Constants.CMDPARAM_COMPLEMENT_DATA_SCHEDULE_DATE_LIST;
mport statc org.apache.dolphnscheduler.common.Constants.CMDPARAM_COMPLEMENT_DATA_START_DATE;
mport statc org.apache.dolphnscheduler.common.Constants.CMD_PARAM_RECOVERY_START_NODE_STRING;
mport statc org.apache.dolphnscheduler.common.Constants.CMD_PARAM_RECOVER_PROCESS_ID_STRING;
mport statc org.apache.dolphnscheduler.common.Constants.CMD_PARAM_START_NODES;
mport statc org.apache.dolphnscheduler.common.Constants.COMMA;
mport statc org.apache.dolphnscheduler.common.Constants.DEFAULT_WORKER_GROUP;
mport statc org.apache.dolphnscheduler.common.Constants.YYYY_MM_DD_HH_MM_SS;
mport statc org.apache.dolphnscheduler.plugn.task.ap.TaskConstants.TASK_TYPE_BLOCKING;
mport statc org.apache.dolphnscheduler.plugn.task.ap.enums.DataType.VARCHAR;
mport statc org.apache.dolphnscheduler.plugn.task.ap.enums.Drect.IN;
mport org.apache.dolphnscheduler.common.Constants;
mport org.apache.dolphnscheduler.common.enums.CommandType;
mport org.apache.dolphnscheduler.common.enums.FalureStrategy;
mport org.apache.dolphnscheduler.common.enums.Flag;
mport org.apache.dolphnscheduler.common.enums.Prorty;
mport org.apache.dolphnscheduler.common.enums.StateEventType;
mport org.apache.dolphnscheduler.common.enums.TaskDependType;
mport org.apache.dolphnscheduler.common.enums.TaskGroupQueueStatus;
mport org.apache.dolphnscheduler.common.graph.DAG;
mport org.apache.dolphnscheduler.common.model.TaskNode;
mport org.apache.dolphnscheduler.common.model.TaskNodeRelaton;
mport org.apache.dolphnscheduler.common.process.ProcessDag;
mport org.apache.dolphnscheduler.common.thread.ThreadUtls;
mport org.apache.dolphnscheduler.common.utls.DateUtls;
mport org.apache.dolphnscheduler.common.utls.JSONUtls;
mport org.apache.dolphnscheduler.common.utls.LoggerUtls;
mport org.apache.dolphnscheduler.common.utls.NetUtls;
mport org.apache.dolphnscheduler.dao.entty.Command;
mport org.apache.dolphnscheduler.dao.entty.Envronment; |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | mport org.apache.dolphnscheduler.dao.entty.ProcessDefnton;
mport org.apache.dolphnscheduler.dao.entty.ProcessInstance;
mport org.apache.dolphnscheduler.dao.entty.ProcessTaskRelaton;
mport org.apache.dolphnscheduler.dao.entty.ProjectUser;
mport org.apache.dolphnscheduler.dao.entty.Schedule;
mport org.apache.dolphnscheduler.dao.entty.TaskDefntonLog;
mport org.apache.dolphnscheduler.dao.entty.TaskGroupQueue;
mport org.apache.dolphnscheduler.dao.entty.TaskInstance;
mport org.apache.dolphnscheduler.dao.utls.DagHelper;
mport org.apache.dolphnscheduler.plugn.task.ap.enums.DependResult;
mport org.apache.dolphnscheduler.plugn.task.ap.enums.Drect;
mport org.apache.dolphnscheduler.plugn.task.ap.enums.ExecutonStatus;
mport org.apache.dolphnscheduler.plugn.task.ap.model.Property;
mport org.apache.dolphnscheduler.remote.command.HostUpdateCommand;
mport org.apache.dolphnscheduler.remote.utls.Host;
mport org.apache.dolphnscheduler.server.master.confg.MasterConfg;
mport org.apache.dolphnscheduler.server.master.dspatch.executor.NettyExecutorManager;
mport org.apache.dolphnscheduler.server.master.event.StateEvent;
mport org.apache.dolphnscheduler.server.master.event.StateEventHandleError;
mport org.apache.dolphnscheduler.server.master.event.StateEventHandleExcepton;
mport org.apache.dolphnscheduler.server.master.event.StateEventHandler;
mport org.apache.dolphnscheduler.server.master.event.StateEventHandlerManager;
mport org.apache.dolphnscheduler.server.master.metrcs.TaskMetrcs;
mport org.apache.dolphnscheduler.server.master.runner.task.ITaskProcessor;
mport org.apache.dolphnscheduler.server.master.runner.task.TaskActon;
mport org.apache.dolphnscheduler.server.master.runner.task.TaskProcessorFactory;
mport org.apache.dolphnscheduler.servce.alert.ProcessAlertManager;
mport org.apache.dolphnscheduler.servce.cron.CronUtls;
mport org.apache.dolphnscheduler.servce.exceptons.CronParseExcepton;
mport org.apache.dolphnscheduler.servce.expand.CurngParamsServce; |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | mport org.apache.dolphnscheduler.servce.process.ProcessServce;
mport org.apache.dolphnscheduler.servce.queue.PeerTaskInstancePrortyQueue;
mport org.apache.commons.collectons.CollectonUtls;
mport org.apache.commons.lang3.ArrayUtls;
mport org.apache.commons.lang3.StrngUtls;
mport org.apache.commons.lang3.math.NumberUtls;
mport java.utl.ArrayLst;
mport java.utl.Arrays;
mport java.utl.Collecton;
mport java.utl.Collectons;
mport java.utl.Date;
mport java.utl.HashMap;
mport java.utl.HashSet;
mport java.utl.Iterator;
mport java.utl.Lst;
mport java.utl.Map;
mport java.utl.Objects;
mport java.utl.Optonal;
mport java.utl.Set;
mport java.utl.concurrent.Callable;
mport java.utl.concurrent.ConcurrentHashMap;
mport java.utl.concurrent.ConcurrentLnkedQueue;
mport java.utl.concurrent.atomc.AtomcBoolean;
mport java.utl.stream.Collectors;
mport org.slf4j.Logger;
mport org.slf4j.LoggerFactory;
mport org.sprngframework.beans.BeanUtls;
mport com.google.common.collect.Lsts;
mport lombok.NonNull;
/** |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | * Workflow execute task, used to execute a workflow nstance.
*/
publc class WorkflowExecuteRunnable mplements Callable<WorkflowSubmtStatue> {
prvate statc fnal Logger logger = LoggerFactory.getLogger(WorkflowExecuteRunnable.class);
prvate fnal ProcessServce processServce;
prvate fnal ProcessAlertManager processAlertManager;
prvate fnal NettyExecutorManager nettyExecutorManager;
prvate fnal ProcessInstance processInstance;
prvate ProcessDefnton processDefnton;
prvate DAG<Strng, TaskNode, TaskNodeRelaton> dag;
/**
* unque key of workflow
*/
prvate Strng key;
prvate WorkflowRunnableStatus workflowRunnableStatus = WorkflowRunnableStatus.CREATED;
/**
* submt falure nodes
*/
prvate boolean taskFaledSubmt = false;
/**
* task nstance hash map, taskId as key
*/
prvate fnal Map<Integer, TaskInstance> taskInstanceMap = new ConcurrentHashMap<>();
/**
* runnng taskProcessor, taskCode as key, taskProcessor as value
* only on taskProcessor per taskCode
*/
prvate fnal Map<Long, ITaskProcessor> actveTaskProcessorMaps = new ConcurrentHashMap<>();
/**
* vald task map, taskCode as key, taskId as value |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | * n a DAG, only one taskInstance per taskCode s vald
*/
prvate fnal Map<Long, Integer> valdTaskMap = new ConcurrentHashMap<>();
/**
* error task map, taskCode as key, taskInstanceId as value
* n a DAG, only one taskInstance per taskCode s vald
*/
prvate fnal Map<Long, Integer> errorTaskMap = new ConcurrentHashMap<>();
/**
* complete task map, taskCode as key, taskInstanceId as value
* n a DAG, only one taskInstance per taskCode s vald
*/
prvate fnal Map<Long, Integer> completeTaskMap = new ConcurrentHashMap<>();
/**
* depend faled task map, taskCode as key, taskId as value
*/
prvate fnal Map<Long, Integer> dependFaledTaskMap = new ConcurrentHashMap<>();
/**
* forbdden task map, code as key
*/
prvate fnal Map<Long, TaskNode> forbddenTaskMap = new ConcurrentHashMap<>();
/**
* skp task map, code as key
*/
prvate fnal Map<Strng, TaskNode> skpTaskNodeMap = new ConcurrentHashMap<>();
/**
* complement date lst
*/
prvate Lst<Date> complementLstDate = Lsts.newLnkedLst();
/** |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | * state event queue
*/
prvate fnal ConcurrentLnkedQueue<StateEvent> stateEvents = new ConcurrentLnkedQueue<>();
/**
* The StandBy task lst, wll be executed, need to know, the taskInstance n ths queue may doesn't have d.
*/
prvate fnal PeerTaskInstancePrortyQueue readyToSubmtTaskQueue = new PeerTaskInstancePrortyQueue();
/**
* wat to retry taskInstance map, taskCode as key, taskInstance as value
* before retry, the taskInstance d s 0
*/
prvate fnal Map<Long, TaskInstance> watToRetryTaskInstanceMap = new ConcurrentHashMap<>();
prvate fnal StateWheelExecuteThread stateWheelExecuteThread;
prvate fnal CurngParamsServce curngParamsServce;
prvate fnal Strng masterAddress;
/**
* @param processInstance processInstance
* @param processServce processServce
* @param nettyExecutorManager nettyExecutorManager
* @param processAlertManager processAlertManager
* @param masterConfg masterConfg
* @param stateWheelExecuteThread stateWheelExecuteThread
*/
publc WorkflowExecuteRunnable(
@NonNull ProcessInstance processInstance,
@NonNull ProcessServce processServce,
@NonNull NettyExecutorManager nettyExecutorManager,
@NonNull ProcessAlertManager processAlertManager,
@NonNull MasterConfg masterConfg,
@NonNull StateWheelExecuteThread stateWheelExecuteThread, |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | @NonNull CurngParamsServce curngParamsServce) {
ths.processServce = processServce;
ths.processInstance = processInstance;
ths.nettyExecutorManager = nettyExecutorManager;
ths.processAlertManager = processAlertManager;
ths.stateWheelExecuteThread = stateWheelExecuteThread;
ths.curngParamsServce = curngParamsServce;
ths.masterAddress = NetUtls.getAddr(masterConfg.getLstenPort());
TaskMetrcs.regsterTaskPrepared(readyToSubmtTaskQueue::sze);
}
/**
* the process start nodes are submtted completely.
*/
publc boolean sStart() {
return WorkflowRunnableStatus.STARTED == workflowRunnableStatus;
}
/**
* handle event
*/
publc vod handleEvents() {
f (!sStart()) {
logger.nfo(
"The workflow nstance s not started, wll not handle ts state event, current state event sze: {}",
stateEvents);
return;
}
StateEvent stateEvent = null;
whle (!ths.stateEvents.sEmpty()) {
try {
stateEvent = ths.stateEvents.peek(); |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | LoggerUtls.setWorkflowAndTaskInstanceIDMDC(stateEvent.getProcessInstanceId(),
stateEvent.getTaskInstanceId());
checkProcessInstance(stateEvent);
StateEventHandler stateEventHandler =
StateEventHandlerManager.getStateEventHandler(stateEvent.getType())
.orElseThrow(() -> new StateEventHandleError("Cannot fnd handler for the gven state event"));
logger.nfo("Begn to handle state event, {}", stateEvent);
f (stateEventHandler.handleStateEvent(ths, stateEvent)) {
ths.stateEvents.remove(stateEvent);
}
} catch (StateEventHandleError stateEventHandleError) {
logger.error("State event handle error, wll remove ths event: {}", stateEvent, stateEventHandleError);
ths.stateEvents.remove(stateEvent);
ThreadUtls.sleep(Constants.SLEEP_TIME_MILLIS);
} catch (StateEventHandleExcepton stateEventHandleExcepton) {
logger.error("State event handle error, wll retry ths event: {}",
stateEvent,
stateEventHandleExcepton);
ThreadUtls.sleep(Constants.SLEEP_TIME_MILLIS);
} catch (Excepton e) {
logger.error("State event handle error, get a unknown excepton, wll retry ths event: {}",
stateEvent,
e);
ThreadUtls.sleep(Constants.SLEEP_TIME_MILLIS);
} fnally {
LoggerUtls.removeWorkflowAndTaskInstanceIdMDC();
} |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | }
}
publc Strng getKey() {
f (StrngUtls.sNotEmpty(key) || ths.processDefnton == null) {
return key;
}
key = Strng.format("%d_%d_%d",
ths.processDefnton.getCode(),
ths.processDefnton.getVerson(),
ths.processInstance.getId());
return key;
}
publc boolean addStateEvent(StateEvent stateEvent) {
f (processInstance.getId() != stateEvent.getProcessInstanceId()) {
logger.nfo("state event would be abounded :{}", stateEvent);
return false;
}
ths.stateEvents.add(stateEvent);
return true;
}
publc nt eventSze() {
return ths.stateEvents.sze();
}
publc ProcessInstance getProcessInstance() {
return ths.processInstance;
}
publc boolean checkForceStartAndWakeUp(StateEvent stateEvent) {
TaskGroupQueue taskGroupQueue = ths.processServce.loadTaskGroupQueue(stateEvent.getTaskInstanceId());
f (taskGroupQueue.getForceStart() == Flag.YES.getCode()) {
TaskInstance taskInstance = ths.processServce.fndTaskInstanceById(stateEvent.getTaskInstanceId()); |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | ITaskProcessor taskProcessor = actveTaskProcessorMaps.get(taskInstance.getTaskCode());
taskProcessor.acton(TaskActon.DISPATCH);
ths.processServce.updateTaskGroupQueueStatus(taskGroupQueue.getTaskId(),
TaskGroupQueueStatus.ACQUIRE_SUCCESS.getCode());
return true;
}
f (taskGroupQueue.getInQueue() == Flag.YES.getCode()) {
boolean acqureTaskGroup = processServce.acqureTaskGroupAgan(taskGroupQueue);
f (acqureTaskGroup) {
TaskInstance taskInstance = ths.processServce.fndTaskInstanceById(stateEvent.getTaskInstanceId());
ITaskProcessor taskProcessor = actveTaskProcessorMaps.get(taskInstance.getTaskCode());
taskProcessor.acton(TaskActon.DISPATCH);
return true;
}
}
return false;
}
publc vod processTmeout() {
ProjectUser projectUser = processServce.queryProjectWthUserByProcessInstanceId(processInstance.getId());
ths.processAlertManager.sendProcessTmeoutAlert(ths.processInstance, projectUser);
}
publc vod taskTmeout(TaskInstance taskInstance) {
ProjectUser projectUser = processServce.queryProjectWthUserByProcessInstanceId(processInstance.getId());
processAlertManager.sendTaskTmeoutAlert(processInstance, taskInstance, projectUser);
}
publc vod taskFnshed(TaskInstance taskInstance) throws StateEventHandleExcepton {
logger.nfo("TaskInstance fnshed task code:{} state:{}", taskInstance.getTaskCode(), taskInstance.getState());
try {
actveTaskProcessorMaps.remove(taskInstance.getTaskCode());
stateWheelExecuteThread.removeTask4TmeoutCheck(processInstance, taskInstance); |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | stateWheelExecuteThread.removeTask4RetryCheck(processInstance, taskInstance);
stateWheelExecuteThread.removeTask4StateCheck(processInstance, taskInstance);
f (taskInstance.getState().typeIsSuccess()) {
completeTaskMap.put(taskInstance.getTaskCode(), taskInstance.getId());
processInstance.setVarPool(taskInstance.getVarPool());
processServce.saveProcessInstance(processInstance);
f (!processInstance.sBlocked()) {
submtPostNode(Long.toStrng(taskInstance.getTaskCode()));
}
} else f (taskInstance.taskCanRetry() && processInstance.getState() != ExecutonStatus.READY_STOP) {
logger.nfo("Retry taskInstance taskInstance state: {}", taskInstance.getState());
retryTaskInstance(taskInstance);
} else f (taskInstance.getState().typeIsFalure()) {
completeTaskMap.put(taskInstance.getTaskCode(), taskInstance.getId());
f (processInstance.getFalureStrategy() == FalureStrategy.CONTINUE && DagHelper.haveAllNodeAfterNode(
Long.toStrng(taskInstance.getTaskCode()),
dag)) {
submtPostNode(Long.toStrng(taskInstance.getTaskCode()));
} else {
errorTaskMap.put(taskInstance.getTaskCode(), taskInstance.getId());
f (processInstance.getFalureStrategy() == FalureStrategy.END) {
kllAllTasks();
}
}
} else f (taskInstance.getState().typeIsFnshed()) {
completeTaskMap.put(taskInstance.getTaskCode(), taskInstance.getId()); |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | }
logger.nfo("TaskInstance fnshed wll try to update the workflow nstance state, task code:{} state:{}",
taskInstance.getTaskCode(),
taskInstance.getState());
ths.updateProcessInstanceState();
} catch (Excepton ex) {
logger.error("Task fnsh faled, get a excepton, wll remove ths taskInstance from completeTaskMap", ex);
completeTaskMap.remove(taskInstance.getTaskCode());
throw ex;
}
}
/**
* release task group
*
* @param taskInstance
*/
publc vod releaseTaskGroup(TaskInstance taskInstance) {
f (taskInstance.getTaskGroupId() > 0) {
TaskInstance nextTaskInstance = ths.processServce.releaseTaskGroup(taskInstance);
f (nextTaskInstance != null) {
f (nextTaskInstance.getProcessInstanceId() == taskInstance.getProcessInstanceId()) {
StateEvent nextEvent = new StateEvent();
nextEvent.setProcessInstanceId(ths.processInstance.getId());
nextEvent.setTaskInstanceId(nextTaskInstance.getId());
nextEvent.setType(StateEventType.WAIT_TASK_GROUP);
ths.stateEvents.add(nextEvent);
} else {
ProcessInstance processInstance = ths.processServce.fndProcessInstanceById(nextTaskInstance.getProcessInstanceId());
ths.processServce.sendStartTask2Master(processInstance, nextTaskInstance.getId(), org.apache.dolphnscheduler.remote.command.CommandType.TASK_WAKEUP_EVENT_REQUEST); |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | }
}
}
}
/**
* crate new task nstance to retry, dfferent objects from the orgnal
*
* @param taskInstance
*/
prvate vod retryTaskInstance(TaskInstance taskInstance) throws StateEventHandleExcepton {
f (!taskInstance.taskCanRetry()) {
return;
}
TaskInstance newTaskInstance = cloneRetryTaskInstance(taskInstance);
f (newTaskInstance == null) {
logger.error("retry fal, new taskInstance s null, task code:{}, task d:{}",
taskInstance.getTaskCode(),
taskInstance.getId());
return;
}
watToRetryTaskInstanceMap.put(newTaskInstance.getTaskCode(), newTaskInstance);
f (!taskInstance.retryTaskIntervalOverTme()) {
logger.nfo("falure task wll be submtted: process d: {}, task nstance code: {} state:{} retry tmes:{} / {}, nterval:{}", processInstance.getId(), newTaskInstance.getTaskCode(),
newTaskInstance.getState(), newTaskInstance.getRetryTmes(), newTaskInstance.getMaxRetryTmes(), newTaskInstance.getRetryInterval());
stateWheelExecuteThread.addTask4TmeoutCheck(processInstance, newTaskInstance);
stateWheelExecuteThread.addTask4RetryCheck(processInstance, newTaskInstance);
} else {
addTaskToStandByLst(newTaskInstance);
submtStandByTask();
watToRetryTaskInstanceMap.remove(newTaskInstance.getTaskCode()); |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | }
}
/**
* update process nstance
*/
publc vod refreshProcessInstance(nt processInstanceId) {
logger.nfo("process nstance update: {}", processInstanceId);
ProcessInstance newProcessInstance = processServce.fndProcessInstanceById(processInstanceId);
BeanUtls.copyPropertes(newProcessInstance, processInstance);
processDefnton = processServce.fndProcessDefnton(processInstance.getProcessDefntonCode(),
processInstance.getProcessDefntonVerson());
processInstance.setProcessDefnton(processDefnton);
}
/**
* update task nstance
*/
publc vod refreshTaskInstance(nt taskInstanceId) {
logger.nfo("task nstance update: {} ", taskInstanceId);
TaskInstance taskInstance = processServce.fndTaskInstanceById(taskInstanceId);
f (taskInstance == null) {
logger.error("can not fnd task nstance, d:{}", taskInstanceId);
return;
}
processServce.packageTaskInstance(taskInstance, processInstance);
taskInstanceMap.put(taskInstance.getId(), taskInstance);
valdTaskMap.remove(taskInstance.getTaskCode());
f (Flag.YES == taskInstance.getFlag()) {
valdTaskMap.put(taskInstance.getTaskCode(), taskInstance.getId());
} |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | }
/**
* check process nstance by state event
*/
publc vod checkProcessInstance(StateEvent stateEvent) throws StateEventHandleError {
f (ths.processInstance.getId() != stateEvent.getProcessInstanceId()) {
throw new StateEventHandleError("The event doesn't contans process nstance d");
}
}
/**
* check f task nstance exst by state event
*/
publc vod checkTaskInstanceByStateEvent(StateEvent stateEvent) throws StateEventHandleError {
f (stateEvent.getTaskInstanceId() == 0) {
throw new StateEventHandleError("The taskInstanceId s 0");
}
f (!taskInstanceMap.contansKey(stateEvent.getTaskInstanceId())) {
throw new StateEventHandleError("Cannot fnd the taskInstance from taskInstanceMap");
}
}
/**
* check f task nstance exst by d
*/
publc boolean checkTaskInstanceById(nt taskInstanceId) {
f (taskInstanceMap.sEmpty()) {
return false;
}
return taskInstanceMap.contansKey(taskInstanceId);
}
/** |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | * get task nstance from memory
*/
publc Optonal<TaskInstance> getTaskInstance(nt taskInstanceId) {
f (taskInstanceMap.contansKey(taskInstanceId)) {
return Optonal.ofNullable(taskInstanceMap.get(taskInstanceId));
}
return Optonal.empty();
}
publc Optonal<TaskInstance> getTaskInstance(long taskCode) {
f (taskInstanceMap.sEmpty()) {
return Optonal.empty();
}
for (TaskInstance taskInstance : taskInstanceMap.values()) {
f (taskInstance.getTaskCode() == taskCode) {
return Optonal.of(taskInstance);
}
}
return Optonal.empty();
}
publc Optonal<TaskInstance> getActveTaskInstanceByTaskCode(long taskCode) {
Integer taskInstanceId = valdTaskMap.get(taskCode);
f (taskInstanceId != null) {
return Optonal.ofNullable(taskInstanceMap.get(taskInstanceId));
}
return Optonal.empty();
}
publc Optonal<TaskInstance> getRetryTaskInstanceByTaskCode(long taskCode) {
f (watToRetryTaskInstanceMap.contansKey(taskCode)) {
return Optonal.ofNullable(watToRetryTaskInstanceMap.get(taskCode));
} |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | return Optonal.empty();
}
publc vod processBlock() {
ProjectUser projectUser = processServce.queryProjectWthUserByProcessInstanceId(processInstance.getId());
processAlertManager.sendProcessBlockngAlert(processInstance, projectUser);
logger.nfo("processInstance {} block alert send successful!", processInstance.getId());
}
publc boolean processComplementData() {
f (!needComplementProcess()) {
return false;
}
f (processInstance.getState() == ExecutonStatus.READY_STOP || !processInstance.getState().typeIsFnshed()) {
return false;
}
Date scheduleDate = processInstance.getScheduleTme();
f (scheduleDate == null) {
scheduleDate = complementLstDate.get(0);
} else f (processInstance.getState().typeIsFnshed()) {
endProcess();
f (complementLstDate.sEmpty()) {
logger.nfo("process complement end. process d:{}", processInstance.getId());
return true;
}
nt ndex = complementLstDate.ndexOf(scheduleDate);
f (ndex >= complementLstDate.sze() - 1 || !processInstance.getState().typeIsSuccess()) {
logger.nfo("process complement end. process d:{}", processInstance.getId());
return true; |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | }
logger.nfo("process complement contnue. process d:{}, schedule tme:{} complementLstDate:{}", processInstance.getId(), processInstance.getScheduleTme(), complementLstDate);
scheduleDate = complementLstDate.get(ndex + 1);
}
nt create = ths.createComplementDataCommand(scheduleDate);
f (create > 0) {
logger.nfo("create complement data command successfully.");
}
return true;
}
prvate nt createComplementDataCommand(Date scheduleDate) {
Command command = new Command();
command.setScheduleTme(scheduleDate);
command.setCommandType(CommandType.COMPLEMENT_DATA);
command.setProcessDefntonCode(processInstance.getProcessDefntonCode());
Map<Strng, Strng> cmdParam = JSONUtls.toMap(processInstance.getCommandParam());
f (cmdParam.contansKey(Constants.CMD_PARAM_RECOVERY_START_NODE_STRING)) {
cmdParam.remove(Constants.CMD_PARAM_RECOVERY_START_NODE_STRING);
}
f (cmdParam.contansKey(CMDPARAM_COMPLEMENT_DATA_SCHEDULE_DATE_LIST)) {
cmdParam.replace(CMDPARAM_COMPLEMENT_DATA_SCHEDULE_DATE_LIST,
cmdParam.get(CMDPARAM_COMPLEMENT_DATA_SCHEDULE_DATE_LIST)
.substrng(cmdParam.get(CMDPARAM_COMPLEMENT_DATA_SCHEDULE_DATE_LIST).ndexOf(COMMA) + 1));
}
f (cmdParam.contansKey(CMDPARAM_COMPLEMENT_DATA_START_DATE)) {
cmdParam.replace(CMDPARAM_COMPLEMENT_DATA_START_DATE, DateUtls.format(scheduleDate, YYYY_MM_DD_HH_MM_SS, null));
}
command.setCommandParam(JSONUtls.toJsonStrng(cmdParam));
command.setTaskDependType(processInstance.getTaskDependType()); |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | command.setFalureStrategy(processInstance.getFalureStrategy());
command.setWarnngType(processInstance.getWarnngType());
command.setWarnngGroupId(processInstance.getWarnngGroupId());
command.setStartTme(new Date());
command.setExecutorId(processInstance.getExecutorId());
command.setUpdateTme(new Date());
command.setProcessInstanceProrty(processInstance.getProcessInstanceProrty());
command.setWorkerGroup(processInstance.getWorkerGroup());
command.setEnvronmentCode(processInstance.getEnvronmentCode());
command.setDryRun(processInstance.getDryRun());
command.setProcessInstanceId(0);
command.setProcessDefntonVerson(processInstance.getProcessDefntonVerson());
return processServce.createCommand(command);
}
prvate boolean needComplementProcess() {
f (processInstance.sComplementData() && Flag.NO == processInstance.getIsSubProcess()) {
return true;
}
return false;
}
/**
* ProcessInstance start entrypont.
*/
@Overrde
publc WorkflowSubmtStatue call() {
f (sStart()) {
logger.warn("[WorkflowInstance-{}] The workflow has already been started", processInstance.getId());
return WorkflowSubmtStatue.DUPLICATED_SUBMITTED;
} |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | try {
LoggerUtls.setWorkflowInstanceIdMDC(processInstance.getId());
f (workflowRunnableStatus == WorkflowRunnableStatus.CREATED) {
buldFlowDag();
workflowRunnableStatus = WorkflowRunnableStatus.INITIALIZE_DAG;
logger.nfo("workflowStatue changed to :{}", workflowRunnableStatus);
}
f (workflowRunnableStatus == WorkflowRunnableStatus.INITIALIZE_DAG) {
ntTaskQueue();
workflowRunnableStatus = WorkflowRunnableStatus.INITIALIZE_QUEUE;
logger.nfo("workflowStatue changed to :{}", workflowRunnableStatus);
}
f (workflowRunnableStatus == WorkflowRunnableStatus.INITIALIZE_QUEUE) {
submtPostNode(null);
workflowRunnableStatus = WorkflowRunnableStatus.STARTED;
logger.nfo("workflowStatue changed to :{}", workflowRunnableStatus);
}
return WorkflowSubmtStatue.SUCCESS;
} catch (Excepton e) {
logger.error("Start workflow error", e);
return WorkflowSubmtStatue.FAILED;
} fnally {
LoggerUtls.removeWorkflowInstanceIdMDC();
}
}
/**
* process end handle
*/
publc vod endProcess() {
ths.stateEvents.clear(); |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | f (processDefnton.getExecutonType().typeIsSeralWat() || processDefnton.getExecutonType()
.typeIsSeralProrty()) {
checkSeralProcess(processDefnton);
}
f (processInstance.getState().typeIsWatngThread()) {
processServce.createRecoveryWatngThreadCommand(null, processInstance);
}
ProjectUser projectUser = processServce.queryProjectWthUserByProcessInstanceId(processInstance.getId());
f (processAlertManager.sNeedToSendWarnng(processInstance)) {
processAlertManager.sendAlertProcessInstance(processInstance, getValdTaskLst(), projectUser);
}
f (processInstance.getState().typeIsSuccess()) {
processAlertManager.closeAlert(processInstance);
}
f (checkTaskQueue()) {
processServce.releaseAllTaskGroup(processInstance.getId());
}
}
publc vod checkSeralProcess(ProcessDefnton processDefnton) {
nt nextInstanceId = processInstance.getNextProcessInstanceId();
f (nextInstanceId == 0) {
ProcessInstance nextProcessInstance =
ths.processServce.loadNextProcess4Seral(processInstance.getProcessDefnton().getCode(), ExecutonStatus.SERIAL_WAIT.getCode(), processInstance.getId());
f (nextProcessInstance == null) {
return;
}
ProcessInstance nextReadyStopProcessInstance =
ths.processServce.loadNextProcess4Seral(processInstance.getProcessDefnton().getCode(), ExecutonStatus.READY_STOP.getCode(), processInstance.getId());
f (processDefnton.getExecutonType().typeIsSeralProrty() && nextReadyStopProcessInstance != null) { |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | return;
}
nextInstanceId = nextProcessInstance.getId();
}
ProcessInstance nextProcessInstance = ths.processServce.fndProcessInstanceById(nextInstanceId);
f (nextProcessInstance.getState().typeIsFnshed() || nextProcessInstance.getState().typeIsRunnng()) {
return;
}
Map<Strng, Object> cmdParam = new HashMap<>();
cmdParam.put(CMD_PARAM_RECOVER_PROCESS_ID_STRING, nextInstanceId);
Command command = new Command();
command.setCommandType(CommandType.RECOVER_SERIAL_WAIT);
command.setProcessInstanceId(nextProcessInstance.getId());
command.setProcessDefntonCode(processDefnton.getCode());
command.setProcessDefntonVerson(processDefnton.getVerson());
command.setCommandParam(JSONUtls.toJsonStrng(cmdParam));
processServce.createCommand(command);
}
/**
* Generate process dag
*
* @throws Excepton excepton
*/
prvate vod buldFlowDag() throws Excepton {
processDefnton = processServce.fndProcessDefnton(processInstance.getProcessDefntonCode(), processInstance.getProcessDefntonVerson());
processInstance.setProcessDefnton(processDefnton);
Lst<TaskInstance> recoverNodeLst = getRecoverTaskInstanceLst(processInstance.getCommandParam());
Lst<ProcessTaskRelaton> processTaskRelatons = processServce.fndRelatonByCode(processDefnton.getCode(), processDefnton.getVerson());
Lst<TaskDefntonLog> taskDefntonLogs = processServce.getTaskDefneLogLstByRelaton(processTaskRelatons);
Lst<TaskNode> taskNodeLst = processServce.transformTask(processTaskRelatons, taskDefntonLogs); |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | forbddenTaskMap.clear();
taskNodeLst.forEach(taskNode -> {
f (taskNode.sForbdden()) {
forbddenTaskMap.put(taskNode.getCode(), taskNode);
}
});
Lst<Strng> recoveryNodeCodeLst = getRecoveryNodeCodeLst(recoverNodeLst);
Lst<Strng> startNodeNameLst = parseStartNodeName(processInstance.getCommandParam());
ProcessDag processDag = generateFlowDag(taskNodeLst, startNodeNameLst, recoveryNodeCodeLst, processInstance.getTaskDependType());
f (processDag == null) {
logger.error("processDag s null");
return;
}
dag = DagHelper.buldDagGraph(processDag);
logger.nfo("Buld dag success, dag: {}", dag);
}
/**
* nt task queue
*/
prvate vod ntTaskQueue() throws StateEventHandleExcepton, CronParseExcepton {
taskFaledSubmt = false;
actveTaskProcessorMaps.clear();
dependFaledTaskMap.clear();
completeTaskMap.clear();
errorTaskMap.clear();
f (!sNewProcessInstance()) {
logger.nfo("The workflowInstance s not a newly runnng nstance, runtmes: {}, recover flag: {}",
processInstance.getRunTmes(), |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | processInstance.getRecovery());
Lst<TaskInstance> valdTaskInstanceLst =
processServce.fndValdTaskLstByProcessId(processInstance.getId());
for (TaskInstance task : valdTaskInstanceLst) {
try {
LoggerUtls.setWorkflowAndTaskInstanceIDMDC(task.getProcessInstanceId(), task.getId());
logger.nfo(
"Check the taskInstance from a exst workflowInstance, exstTaskInstanceCode: {}, taskInstanceStatus: {}",
task.getTaskCode(),
task.getState());
f (valdTaskMap.contansKey(task.getTaskCode())) {
nt oldTaskInstanceId = valdTaskMap.get(task.getTaskCode());
TaskInstance oldTaskInstance = taskInstanceMap.get(oldTaskInstanceId);
f (!oldTaskInstance.getState().typeIsFnshed() && task.getState().typeIsFnshed()) {
task.setFlag(Flag.NO);
processServce.updateTaskInstance(task);
contnue;
}
logger.warn("have same taskCode taskInstance when nt task queue, taskCode:{}",
task.getTaskCode());
}
valdTaskMap.put(task.getTaskCode(), task.getId());
taskInstanceMap.put(task.getId(), task);
f (task.sTaskComplete()) {
completeTaskMap.put(task.getTaskCode(), task.getId());
contnue;
}
f (task.sCondtonsTask() || DagHelper.haveCondtonsAfterNode(Long.toStrng(task.getTaskCode()),
dag)) {
contnue; |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | }
f (task.taskCanRetry()) {
f (task.getState() == ExecutonStatus.NEED_FAULT_TOLERANCE) {
TaskInstance tolerantTaskInstance = cloneTolerantTaskInstance(task);
addTaskToStandByLst(tolerantTaskInstance);
} else {
retryTaskInstance(task);
}
contnue;
}
f (task.getState().typeIsFalure()) {
errorTaskMap.put(task.getTaskCode(), task.getId());
}
} fnally {
LoggerUtls.removeWorkflowAndTaskInstanceIdMDC();
}
}
} else {
logger.nfo("The current workflowInstance s a newly runnng workflowInstance");
}
f (processInstance.sComplementData() && complementLstDate.sEmpty()) {
Map<Strng, Strng> cmdParam = JSONUtls.toMap(processInstance.getCommandParam());
f (cmdParam != null) {
setGlobalParamIfCommanded(processDefnton, cmdParam);
Date start = null;
Date end = null;
f (cmdParam.contansKey(CMDPARAM_COMPLEMENT_DATA_START_DATE) && cmdParam.contansKey(CMDPARAM_COMPLEMENT_DATA_END_DATE)) {
start = DateUtls.strngToDate(cmdParam.get(CMDPARAM_COMPLEMENT_DATA_START_DATE)); |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | end = DateUtls.strngToDate(cmdParam.get(CMDPARAM_COMPLEMENT_DATA_END_DATE));
}
Lst<Schedule> schedules = processServce.queryReleaseSchedulerLstByProcessDefntonCode(processInstance.getProcessDefntonCode());
f (complementLstDate.sEmpty() && needComplementProcess()) {
f (start != null && end != null) {
complementLstDate = CronUtls.getSelfFreDateLst(start, end, schedules);
}
f (cmdParam.contansKey(CMDPARAM_COMPLEMENT_DATA_SCHEDULE_DATE_LIST)) {
complementLstDate = CronUtls.getSelfScheduleDateLst(cmdParam);
}
logger.nfo(" process defnton code:{} complement data: {}", processInstance.getProcessDefntonCode(), complementLstDate);
f (!complementLstDate.sEmpty() && Flag.NO == processInstance.getIsSubProcess()) {
processInstance.setScheduleTme(complementLstDate.get(0));
Strng globalParams = curngParamsServce.curngGlobalParams(processInstance.getId(),
processDefnton.getGlobalParamMap(),
processDefnton.getGlobalParamLst(),
CommandType.COMPLEMENT_DATA,
processInstance.getScheduleTme(),
cmdParam.get(Constants.SCHEDULE_TIMEZONE));
processInstance.setGlobalParams(globalParams);
processServce.updateProcessInstance(processInstance);
}
}
}
}
logger.nfo("Intalze task queue, dependFaledTaskMap: {}, completeTaskMap: {}, errorTaskMap: {}",
dependFaledTaskMap,
completeTaskMap,
errorTaskMap);
} |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | /**
* submt task to execute
*
* @param taskInstance task nstance
* @return TaskInstance
*/
prvate Optonal<TaskInstance> submtTaskExec(TaskInstance taskInstance) {
try {
processServce.packageTaskInstance(taskInstance, processInstance);
ITaskProcessor taskProcessor = TaskProcessorFactory.getTaskProcessor(taskInstance.getTaskType());
taskProcessor.nt(taskInstance, processInstance);
f (taskInstance.getState() == ExecutonStatus.RUNNING_EXECUTION && taskProcessor.getType().equalsIgnoreCase(Constants.COMMON_TASK_TYPE)) {
notfyProcessHostUpdate(taskInstance);
}
boolean submt = taskProcessor.acton(TaskActon.SUBMIT);
f (!submt) {
logger.error("process d:{} name:{} submt standby task d:{} name:{} faled!",
processInstance.getId(),
processInstance.getName(),
taskInstance.getId(),
taskInstance.getName());
return Optonal.empty();
}
f (valdTaskMap.contansKey(taskInstance.getTaskCode())) {
nt oldTaskInstanceId = valdTaskMap.get(taskInstance.getTaskCode());
f (taskInstance.getId() != oldTaskInstanceId) {
TaskInstance oldTaskInstance = taskInstanceMap.get(oldTaskInstanceId);
oldTaskInstance.setFlag(Flag.NO); |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | processServce.updateTaskInstance(oldTaskInstance);
valdTaskMap.remove(taskInstance.getTaskCode());
actveTaskProcessorMaps.remove(taskInstance.getTaskCode());
}
}
valdTaskMap.put(taskInstance.getTaskCode(), taskInstance.getId());
taskInstanceMap.put(taskInstance.getId(), taskInstance);
actveTaskProcessorMaps.put(taskInstance.getTaskCode(), taskProcessor);
boolean dspatchSuccess = taskProcessor.acton(TaskActon.DISPATCH);
f (!dspatchSuccess) {
logger.error("process d:{} name:{} dspatch standby task d:{} name:{} faled!",
processInstance.getId(),
processInstance.getName(),
taskInstance.getId(),
taskInstance.getName());
return Optonal.empty();
}
taskProcessor.acton(TaskActon.RUN);
stateWheelExecuteThread.addTask4TmeoutCheck(processInstance, taskInstance);
stateWheelExecuteThread.addTask4StateCheck(processInstance, taskInstance);
f (taskProcessor.taskInstance().getState().typeIsFnshed()) {
f (processInstance.sBlocked()) {
StateEvent processBlockEvent = new StateEvent();
processBlockEvent.setProcessInstanceId(ths.processInstance.getId());
processBlockEvent.setTaskInstanceId(taskInstance.getId());
processBlockEvent.setExecutonStatus(taskProcessor.taskInstance().getState());
processBlockEvent.setType(StateEventType.PROCESS_BLOCKED);
ths.stateEvents.add(processBlockEvent);
}
StateEvent taskStateChangeEvent = new StateEvent(); |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | taskStateChangeEvent.setProcessInstanceId(ths.processInstance.getId());
taskStateChangeEvent.setTaskInstanceId(taskInstance.getId());
taskStateChangeEvent.setExecutonStatus(taskProcessor.taskInstance().getState());
taskStateChangeEvent.setType(StateEventType.TASK_STATE_CHANGE);
ths.stateEvents.add(taskStateChangeEvent);
}
return Optonal.of(taskInstance);
} catch (Excepton e) {
logger.error("submt standby task error, taskCode: {}, taskInstanceId: {}",
taskInstance.getTaskCode(),
taskInstance.getId(),
e);
return Optonal.empty();
}
}
prvate vod notfyProcessHostUpdate(TaskInstance taskInstance) {
f (StrngUtls.sEmpty(taskInstance.getHost())) {
return;
}
try {
HostUpdateCommand hostUpdateCommand = new HostUpdateCommand();
hostUpdateCommand.setProcessHost(masterAddress);
hostUpdateCommand.setTaskInstanceId(taskInstance.getId());
Host host = new Host(taskInstance.getHost());
nettyExecutorManager.doExecute(host, hostUpdateCommand.convert2Command());
} catch (Excepton e) {
logger.error("notfy process host update", e);
}
} |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | /**
* fnd task nstance n db.
* n case submt more than one same name task n the same tme.
*
* @param taskCode task code
* @param taskVerson task verson
* @return TaskInstance
*/
prvate TaskInstance fndTaskIfExsts(Long taskCode, nt taskVerson) {
Lst<TaskInstance> valdTaskInstanceLst = getValdTaskLst();
for (TaskInstance taskInstance : valdTaskInstanceLst) {
f (taskInstance.getTaskCode() == taskCode && taskInstance.getTaskDefntonVerson() == taskVerson) {
return taskInstance;
}
}
return null;
}
/**
* encapsulaton task, ths method wll only create a new task nstance, the return task nstance wll not contan d.
*
* @param processInstance process nstance
* @param taskNode taskNode
* @return TaskInstance
*/
prvate TaskInstance createTaskInstance(ProcessInstance processInstance, TaskNode taskNode) {
TaskInstance taskInstance = fndTaskIfExsts(taskNode.getCode(), taskNode.getVerson());
f (taskInstance != null) {
return taskInstance;
}
return newTaskInstance(processInstance, taskNode); |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | }
/**
* clone a new taskInstance for retry and reset some logc felds
*
* @return
*/
publc TaskInstance cloneRetryTaskInstance(TaskInstance taskInstance) {
TaskNode taskNode = dag.getNode(Long.toStrng(taskInstance.getTaskCode()));
f (taskNode == null) {
logger.error("taskNode s null, code:{}", taskInstance.getTaskCode());
return null;
}
TaskInstance newTaskInstance = newTaskInstance(processInstance, taskNode);
newTaskInstance.setTaskDefne(taskInstance.getTaskDefne());
newTaskInstance.setProcessDefne(taskInstance.getProcessDefne());
newTaskInstance.setProcessInstance(processInstance);
newTaskInstance.setRetryTmes(taskInstance.getRetryTmes() + 1);
newTaskInstance.setState(taskInstance.getState());
newTaskInstance.setEndTme(taskInstance.getEndTme());
return newTaskInstance;
}
/**
* clone a new taskInstance for tolerant and reset some logc felds
*
* @return
*/
publc TaskInstance cloneTolerantTaskInstance(TaskInstance taskInstance) {
TaskNode taskNode = dag.getNode(Long.toStrng(taskInstance.getTaskCode()));
f (taskNode == null) { |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | logger.error("taskNode s null, code:{}", taskInstance.getTaskCode());
return null;
}
TaskInstance newTaskInstance = newTaskInstance(processInstance, taskNode);
newTaskInstance.setTaskDefne(taskInstance.getTaskDefne());
newTaskInstance.setProcessDefne(taskInstance.getProcessDefne());
newTaskInstance.setProcessInstance(processInstance);
newTaskInstance.setRetryTmes(taskInstance.getRetryTmes());
newTaskInstance.setState(taskInstance.getState());
return newTaskInstance;
}
/**
* new a taskInstance
*
* @param processInstance
* @param taskNode
* @return
*/
publc TaskInstance newTaskInstance(ProcessInstance processInstance, TaskNode taskNode) {
TaskInstance taskInstance = new TaskInstance();
taskInstance.setTaskCode(taskNode.getCode());
taskInstance.setTaskDefntonVerson(taskNode.getVerson());
taskInstance.setName(taskNode.getName());
taskInstance.setState(ExecutonStatus.SUBMITTED_SUCCESS);
taskInstance.setProcessInstanceId(processInstance.getId());
taskInstance.setTaskType(taskNode.getType().toUpperCase()); |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | taskInstance.setAlertFlag(Flag.NO);
taskInstance.setStartTme(null);
taskInstance.setFlag(Flag.YES);
taskInstance.setDryRun(processInstance.getDryRun());
taskInstance.setRetryTmes(0);
taskInstance.setMaxRetryTmes(taskNode.getMaxRetryTmes());
taskInstance.setRetryInterval(taskNode.getRetryInterval());
taskInstance.setTaskParams(taskNode.getTaskParams());
taskInstance.setTaskGroupId(taskNode.getTaskGroupId());
taskInstance.setTaskGroupProrty(taskNode.getTaskGroupProrty());
taskInstance.setCpuQuota(taskNode.getCpuQuota());
taskInstance.setMemoryMax(taskNode.getMemoryMax());
f (taskNode.getTaskInstanceProrty() == null) {
taskInstance.setTaskInstanceProrty(Prorty.MEDIUM);
} else {
taskInstance.setTaskInstanceProrty(taskNode.getTaskInstanceProrty());
}
Strng processWorkerGroup = processInstance.getWorkerGroup();
processWorkerGroup = StrngUtls.sBlank(processWorkerGroup) ? DEFAULT_WORKER_GROUP : processWorkerGroup; |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | Strng taskWorkerGroup = StrngUtls.sBlank(taskNode.getWorkerGroup()) ? processWorkerGroup : taskNode.getWorkerGroup();
Long processEnvronmentCode = Objects.sNull(processInstance.getEnvronmentCode()) ? -1 : processInstance.getEnvronmentCode();
Long taskEnvronmentCode = Objects.sNull(taskNode.getEnvronmentCode()) ? processEnvronmentCode : taskNode.getEnvronmentCode();
f (!processWorkerGroup.equals(DEFAULT_WORKER_GROUP) && taskWorkerGroup.equals(DEFAULT_WORKER_GROUP)) {
taskInstance.setWorkerGroup(processWorkerGroup);
taskInstance.setEnvronmentCode(processEnvronmentCode);
} else {
taskInstance.setWorkerGroup(taskWorkerGroup);
taskInstance.setEnvronmentCode(taskEnvronmentCode);
}
f (!taskInstance.getEnvronmentCode().equals(-1L)) {
Envronment envronment = processServce.fndEnvronmentByCode(taskInstance.getEnvronmentCode());
f (Objects.nonNull(envronment) && StrngUtls.sNotEmpty(envronment.getConfg())) {
taskInstance.setEnvronmentConfg(envronment.getConfg());
}
}
taskInstance.setDelayTme(taskNode.getDelayTme());
return taskInstance;
}
publc vod getPreVarPool(TaskInstance taskInstance, Set<Strng> preTask) {
Map<Strng, Property> allProperty = new HashMap<>();
Map<Strng, TaskInstance> allTaskInstance = new HashMap<>();
f (CollectonUtls.sNotEmpty(preTask)) {
for (Strng preTaskCode : preTask) {
Integer taskId = completeTaskMap.get(Long.parseLong(preTaskCode));
f (taskId == null) {
contnue;
}
TaskInstance preTaskInstance = taskInstanceMap.get(taskId); |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | f (preTaskInstance == null) {
contnue;
}
Strng preVarPool = preTaskInstance.getVarPool();
f (StrngUtls.sNotEmpty(preVarPool)) {
Lst<Property> propertes = JSONUtls.toLst(preVarPool, Property.class);
for (Property nfo : propertes) {
setVarPoolValue(allProperty, allTaskInstance, preTaskInstance, nfo);
}
}
}
f (allProperty.sze() > 0) {
taskInstance.setVarPool(JSONUtls.toJsonStrng(allProperty.values()));
}
} else {
f (StrngUtls.sNotEmpty(processInstance.getVarPool())) {
taskInstance.setVarPool(processInstance.getVarPool());
}
}
}
publc Collecton<TaskInstance> getAllTaskInstances() {
return taskInstanceMap.values();
}
prvate vod setVarPoolValue(Map<Strng, Property> allProperty, Map<Strng, TaskInstance> allTaskInstance, TaskInstance preTaskInstance, Property thsProperty) {
thsProperty.setDrect(Drect.IN);
Strng proName = thsProperty.getProp();
f (allProperty.contansKey(proName)) { |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | Property otherPro = allProperty.get(proName);
f (StrngUtls.sEmpty(thsProperty.getValue())) {
allProperty.put(proName, otherPro);
} else f (StrngUtls.sNotEmpty(otherPro.getValue())) {
TaskInstance otherTask = allTaskInstance.get(proName);
f (otherTask.getEndTme().getTme() > preTaskInstance.getEndTme().getTme()) {
allProperty.put(proName, thsProperty);
allTaskInstance.put(proName, preTaskInstance);
} else {
allProperty.put(proName, otherPro);
}
} else {
allProperty.put(proName, thsProperty);
allTaskInstance.put(proName, preTaskInstance);
}
} else {
allProperty.put(proName, thsProperty);
allTaskInstance.put(proName, preTaskInstance);
}
}
/**
* get complete task nstance map, taskCode as key
*/
prvate Map<Strng, TaskInstance> getCompleteTaskInstanceMap() {
Map<Strng, TaskInstance> completeTaskInstanceMap = new HashMap<>();
for (Map.Entry<Long, Integer> entry : completeTaskMap.entrySet()) {
Long taskConde = entry.getKey(); |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | Integer taskInstanceId = entry.getValue();
TaskInstance taskInstance = taskInstanceMap.get(taskInstanceId);
f (taskInstance == null) {
logger.warn("Cannot fnd the taskInstance from taskInstanceMap, taskInstanceId: {}, taskConde: {}",
taskInstanceId,
taskConde);
contnue;
}
completeTaskInstanceMap.put(Long.toStrng(taskInstance.getTaskCode()), taskInstance);
}
return completeTaskInstanceMap;
}
/**
* get vald task lst
*/
prvate Lst<TaskInstance> getValdTaskLst() {
Lst<TaskInstance> valdTaskInstanceLst = new ArrayLst<>();
for (Integer taskInstanceId : valdTaskMap.values()) {
valdTaskInstanceLst.add(taskInstanceMap.get(taskInstanceId));
}
return valdTaskInstanceLst;
}
prvate vod submtPostNode(Strng parentNodeCode) throws StateEventHandleExcepton {
Set<Strng> submtTaskNodeLst =
DagHelper.parsePostNodes(parentNodeCode, skpTaskNodeMap, dag, getCompleteTaskInstanceMap());
Lst<TaskInstance> taskInstances = new ArrayLst<>();
for (Strng taskNode : submtTaskNodeLst) {
TaskNode taskNodeObject = dag.getNode(taskNode);
Optonal<TaskInstance> exstTaskInstanceOptonal = getTaskInstance(taskNodeObject.getCode()); |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | f (exstTaskInstanceOptonal.sPresent()) {
taskInstances.add(exstTaskInstanceOptonal.get());
contnue;
}
TaskInstance task = createTaskInstance(processInstance, taskNodeObject);
taskInstances.add(task);
}
f (StrngUtls.sNotEmpty(parentNodeCode) && dag.getEndNode().contans(parentNodeCode)) {
TaskInstance endTaskInstance = taskInstanceMap.get(completeTaskMap.get(NumberUtls.toLong(parentNodeCode)));
Strng taskInstanceVarPool = endTaskInstance.getVarPool();
f (StrngUtls.sNotEmpty(taskInstanceVarPool)) {
Set<Property> taskPropertes = new HashSet<>(JSONUtls.toLst(taskInstanceVarPool, Property.class));
Strng processInstanceVarPool = processInstance.getVarPool();
f (StrngUtls.sNotEmpty(processInstanceVarPool)) {
Set<Property> propertes = new HashSet<>(JSONUtls.toLst(processInstanceVarPool, Property.class));
propertes.addAll(taskPropertes);
processInstance.setVarPool(JSONUtls.toJsonStrng(propertes));
} else {
processInstance.setVarPool(JSONUtls.toJsonStrng(taskPropertes));
}
}
}
for (TaskInstance task : taskInstances) {
f (readyToSubmtTaskQueue.contans(task)) {
logger.warn("Task s already at submt queue, taskInstanceId: {}", task.getId());
contnue;
}
f (task.getId() > 0 && completeTaskMap.contansKey(task.getTaskCode())) { |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | logger.nfo("task {} has already run success", task.getName());
contnue;
}
f (task.getState().typeIsPause() || task.getState().typeIsCancel()) {
logger.nfo("task {} stopped, the state s {}", task.getName(), task.getState());
contnue;
}
addTaskToStandByLst(task);
}
submtStandByTask();
updateProcessInstanceState();
}
/**
* determne whether the dependences of the task node are complete
*
* @return DependResult
*/
prvate DependResult sTaskDepsComplete(Strng taskCode) {
Collecton<Strng> startNodes = dag.getBegnNode();
f (startNodes.contans(taskCode)) {
return DependResult.SUCCESS;
}
TaskNode taskNode = dag.getNode(taskCode);
Lst<Strng> ndrectDepCodeLst = new ArrayLst<>();
setIndrectDepLst(taskCode, ndrectDepCodeLst);
for (Strng depsNode : ndrectDepCodeLst) {
f (dag.contansNode(depsNode) && !skpTaskNodeMap.contansKey(depsNode)) {
Long despNodeTaskCode = Long.parseLong(depsNode); |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | f (!completeTaskMap.contansKey(despNodeTaskCode)) {
return DependResult.WAITING;
}
Integer depsTaskId = completeTaskMap.get(despNodeTaskCode);
ExecutonStatus depTaskState = taskInstanceMap.get(depsTaskId).getState();
f (depTaskState.typeIsPause() || depTaskState.typeIsCancel()) {
return DependResult.NON_EXEC;
}
f (taskNode.sBlockngTask()) {
contnue;
}
f (taskNode.sCondtonsTask()) {
contnue;
}
f (!dependTaskSuccess(depsNode, taskCode)) {
return DependResult.FAILED;
}
}
}
logger.nfo("taskCode: {} completeDependTaskLst: {}", taskCode, Arrays.toStrng(completeTaskMap.keySet().toArray()));
return DependResult.SUCCESS;
}
/**
* Ths functon s specally used to handle the dependency stuaton where the parent node s a prohbted node.
* When the parent node s a forbdden node, the dependency relatonshp should contnue to be traced
*
* @param taskCode taskCode
* @param ndrectDepCodeLst All ndrectly dependent nodes |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | */
prvate vod setIndrectDepLst(Strng taskCode, Lst<Strng> ndrectDepCodeLst) {
TaskNode taskNode = dag.getNode(taskCode);
Lst<Strng> depCodeLst = taskNode.getDepLst();
for (Strng depsNode : depCodeLst) {
f (forbddenTaskMap.contansKey(Long.parseLong(depsNode))) {
setIndrectDepLst(depsNode, ndrectDepCodeLst);
} else {
ndrectDepCodeLst.add(depsNode);
}
}
}
/**
* depend node s completed, but here need check the condton task branch s the next node
*/
prvate boolean dependTaskSuccess(Strng dependNodeName, Strng nextNodeName) {
f (dag.getNode(dependNodeName).sCondtonsTask()) {
Lst<Strng> nextTaskLst = DagHelper.parseCondtonTask(dependNodeName, skpTaskNodeMap, dag, getCompleteTaskInstanceMap());
f (!nextTaskLst.contans(nextNodeName)) {
return false;
}
} else {
long taskCode = Long.parseLong(dependNodeName);
Integer taskInstanceId = completeTaskMap.get(taskCode);
ExecutonStatus depTaskState = taskInstanceMap.get(taskInstanceId).getState();
f (depTaskState.typeIsFalure()) {
return false;
}
} |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | return true;
}
/**
* query task nstance by complete state
*
* @param state state
* @return task nstance lst
*/
prvate Lst<TaskInstance> getCompleteTaskByState(ExecutonStatus state) {
Lst<TaskInstance> resultLst = new ArrayLst<>();
for (Integer taskInstanceId : completeTaskMap.values()) {
TaskInstance taskInstance = taskInstanceMap.get(taskInstanceId);
f (taskInstance != null && taskInstance.getState() == state) {
resultLst.add(taskInstance);
}
}
return resultLst;
}
/**
* where there are ongong tasks
*
* @param state state
* @return ExecutonStatus
*/
prvate ExecutonStatus runnngState(ExecutonStatus state) {
f (state == ExecutonStatus.READY_STOP || state == ExecutonStatus.READY_PAUSE || state == ExecutonStatus.WAITING_THREAD || state == ExecutonStatus.READY_BLOCK ||
state == ExecutonStatus.DELAY_EXECUTION) {
return state;
} else { |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | return ExecutonStatus.RUNNING_EXECUTION;
}
}
/**
* exsts falure task,contans submt falure、dependency falure,execute falure(retry after)
*
* @return Boolean whether has faled task
*/
prvate boolean hasFaledTask() {
f (ths.taskFaledSubmt) {
return true;
}
f (ths.errorTaskMap.sze() > 0) {
return true;
}
return ths.dependFaledTaskMap.sze() > 0;
}
/**
* process nstance falure
*
* @return Boolean whether process nstance faled
*/
prvate boolean processFaled() {
f (hasFaledTask()) {
logger.nfo("The current process has faled task, the current process faled");
f (processInstance.getFalureStrategy() == FalureStrategy.END) {
return true;
}
f (processInstance.getFalureStrategy() == FalureStrategy.CONTINUE) {
return readyToSubmtTaskQueue.sze() == 0 && actveTaskProcessorMaps.sze() == 0 && watToRetryTaskInstanceMap.sze() == 0; |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | }
}
return false;
}
/**
* whether task for watng thread
*
* @return Boolean whether has watng thread task
*/
prvate boolean hasWatngThreadTask() {
Lst<TaskInstance> watngLst = getCompleteTaskByState(ExecutonStatus.WAITING_THREAD);
return CollectonUtls.sNotEmpty(watngLst);
}
/**
* prepare for pause
* 1，faled retry task n the preparaton queue , returns to falure drectly
* 2，exsts pause task，complement not completed, pendng submsson of tasks, return to suspenson
* 3，success
*
* @return ExecutonStatus
*/
prvate ExecutonStatus processReadyPause() {
f (hasRetryTaskInStandBy()) {
return ExecutonStatus.FAILURE;
}
Lst<TaskInstance> pauseLst = getCompleteTaskByState(ExecutonStatus.PAUSE);
f (CollectonUtls.sNotEmpty(pauseLst) || processInstance.sBlocked() || !sComplementEnd() || readyToSubmtTaskQueue.sze() > 0) {
return ExecutonStatus.PAUSE;
} else {
return ExecutonStatus.SUCCESS; |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | }
}
/**
* prepare for block
* f process has tasks stll runnng, pause them
* f readyToSubmtTaskQueue s not empty, kll them
* else return block status drectly
*
* @return ExecutonStatus
*/
prvate ExecutonStatus processReadyBlock() {
f (actveTaskProcessorMaps.sze() > 0) {
for (ITaskProcessor taskProcessor : actveTaskProcessorMaps.values()) {
f (!TASK_TYPE_BLOCKING.equals(taskProcessor.getType())) {
taskProcessor.acton(TaskActon.PAUSE);
}
}
}
f (readyToSubmtTaskQueue.sze() > 0) {
for (Iterator<TaskInstance> ter = readyToSubmtTaskQueue.terator(); ter.hasNext(); ) {
ter.next().setState(ExecutonStatus.KILL);
}
}
return ExecutonStatus.BLOCK;
}
/**
* generate the latest process nstance status by the tasks state
*
* @return process nstance executon status
*/ |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | prvate ExecutonStatus getProcessInstanceState(ProcessInstance nstance) {
ExecutonStatus state = nstance.getState();
f (actveTaskProcessorMaps.sze() > 0 || hasRetryTaskInStandBy()) {
// actve
ExecutonStatus executonStatus = runnngState(state);
logger.nfo("The workflowInstance has task runnng, the workflowInstance status s {}", executonStatus);
return executonStatus;
}
// block
f (state == ExecutonStatus.READY_BLOCK) {
ExecutonStatus executonStatus = processReadyBlock();
logger.nfo("The workflowInstance s ready to block, the workflowInstance status s {}", executonStatus);
return executonStatus;
}
// watng thread
f (hasWatngThreadTask()) {
logger.nfo("The workflowInstance has watng thread task, the workflow status s {}",
ExecutonStatus.WAITING_THREAD);
return ExecutonStatus.WAITING_THREAD;
}
// pause
f (state == ExecutonStatus.READY_PAUSE) {
ExecutonStatus executonStatus = processReadyPause();
logger.nfo("The workflowInstance s ready to pause, the workflow status s {}", executonStatus);
return executonStatus;
}
// stop
f (state == ExecutonStatus.READY_STOP) {
Lst<TaskInstance> stopLst = getCompleteTaskByState(ExecutonStatus.STOP);
Lst<TaskInstance> kllLst = getCompleteTaskByState(ExecutonStatus.KILL); |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | Lst<TaskInstance> falLst = getCompleteTaskByState(ExecutonStatus.FAILURE);
ExecutonStatus executonStatus;
f (CollectonUtls.sNotEmpty(stopLst) || CollectonUtls.sNotEmpty(kllLst) || CollectonUtls.sNotEmpty(falLst) || !sComplementEnd()) {
executonStatus = ExecutonStatus.STOP;
} else {
executonStatus = ExecutonStatus.SUCCESS;
}
logger.nfo("The workflowInstance s ready to stop, the workflow status s {}", executonStatus);
return executonStatus;
}
// process falure
f (processFaled()) {
logger.nfo("The workflowInstance s faled, the workflow status s {}", ExecutonStatus.FAILURE);
return ExecutonStatus.FAILURE;
}
// success
f (state == ExecutonStatus.RUNNING_EXECUTION) {
Lst<TaskInstance> kllTasks = getCompleteTaskByState(ExecutonStatus.KILL);
f (readyToSubmtTaskQueue.sze() > 0 || watToRetryTaskInstanceMap.sze() > 0) {
//tasks cu
return ExecutonStatus.RUNNING_EXECUTION;
} else f (CollectonUtls.sNotEmpty(kllTasks)) {
// tasks m
return ExecutonStatus.FAILURE;
} else {
// f the
return ExecutonStatus.SUCCESS;
}
}
return state; |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | }
/**
* whether complement end
*
* @return Boolean whether s complement end
*/
prvate boolean sComplementEnd() {
f (!processInstance.sComplementData()) {
return true;
}
Map<Strng, Strng> cmdParam = JSONUtls.toMap(processInstance.getCommandParam());
Date endTme = DateUtls.strngToDate(cmdParam.get(CMDPARAM_COMPLEMENT_DATA_END_DATE));
return processInstance.getScheduleTme().equals(endTme);
}
/**
* updateProcessInstance process nstance state
* after each batch of tasks s executed, of the process nstance s updated
*/
prvate vod updateProcessInstanceState() throws StateEventHandleExcepton {
ExecutonStatus state = getProcessInstanceState(processInstance);
f (processInstance.getState() != state) {
logger.nfo("Update workflowInstance states, orgn state: {}, target state: {}",
processInstance.getState(),
state);
updateWorkflowInstanceStatesToDB(state);
StateEvent stateEvent = new StateEvent();
stateEvent.setExecutonStatus(processInstance.getState());
stateEvent.setProcessInstanceId(ths.processInstance.getId());
stateEvent.setType(StateEventType.PROCESS_STATE_CHANGE);
// replace |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | ths.stateEvents.add(stateEvent);
} else {
logger.nfo("There s no need to update the workflow nstance state, orgn state: {}, target state: {}",
processInstance.getState(),
state);
}
}
/**
* stateEvent's executon status as process nstance state
*/
publc vod updateProcessInstanceState(StateEvent stateEvent) throws StateEventHandleExcepton {
ExecutonStatus state = stateEvent.getExecutonStatus();
updateWorkflowInstanceStatesToDB(state);
}
prvate vod updateWorkflowInstanceStatesToDB(ExecutonStatus newStates) throws StateEventHandleExcepton {
ExecutonStatus orgnStates = processInstance.getState();
f (orgnStates != newStates) {
logger.nfo("Begn to update workflow nstance state , state wll change from {} to {}",
orgnStates,
newStates);
processInstance.setState(newStates);
f (newStates.typeIsFnshed()) {
processInstance.setEndTme(new Date());
}
try {
processServce.updateProcessInstance(processInstance);
} catch (Excepton ex) {
// recover
processInstance.setState(orgnStates);
processInstance.setEndTme(null); |
closed | apache/dolphinscheduler | https://github.com/apache/dolphinscheduler | 11,238 | [Bug] [Master] Cannot recover from failed tasks when force success task | ### Search before asking
- [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues.
### What happened
I find when the task failed, if I force success the failed task, and then recovery the workflow instance, the master will go into dead loop. This is caused it cannot find the failed tasks.
### What you expected to happen
The workflow instance can go into success.
### How to reproduce
1. Create a workflow with one task.
2. The task execute failed.
3. Force success the taskinstance.
4. Recovery the workflow instance.
### Anything else
_No response_
### Version
dev
### Are you willing to submit PR?
- [X] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
| https://github.com/apache/dolphinscheduler/issues/11238 | https://github.com/apache/dolphinscheduler/pull/11239 | caa6287d7fb9d73e18cdad2850f1ec08e602ad38 | 04f3aa97135d79469daf7c21c935029faff827b2 | 2022-08-01T11:48:04Z | java | 2022-08-02T01:30:18Z | dolphinscheduler-master/src/main/java/org/apache/dolphinscheduler/server/master/runner/WorkflowExecuteRunnable.java | throw new StateEventHandleExcepton("Update process nstance status to DB error", ex);
}
}
}
/**
* get task dependency result
*
* @param taskInstance task nstance
* @return DependResult
*/
prvate DependResult getDependResultForTask(TaskInstance taskInstance) {
return sTaskDepsComplete(Long.toStrng(taskInstance.getTaskCode()));
}
/**
* add task to standby lst
*
* @param taskInstance task nstance
*/
publc vod addTaskToStandByLst(TaskInstance taskInstance) {
f (readyToSubmtTaskQueue.contans(taskInstance)) {
logger.warn("task was found n ready submt queue, task code:{}", taskInstance.getTaskCode());
return;
}
logger.nfo("add task to stand by lst, task name:{}, task d:{}, task code:{}",
taskInstance.getName(),
taskInstance.getId(),
taskInstance.getTaskCode());
TaskMetrcs.ncTaskInstanceByState("submt");
readyToSubmtTaskQueue.put(taskInstance);
} |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.