Bug ID
int64 961
1.91M
| Comment ID
int64 3.98k
17.1M
| Author
stringlengths 8
48
| Comment Text
stringlengths 1
64.3k
β |
|---|---|---|---|
1,358,069 | 12,250,326 |
Created attachment 8859946
BugzillaAttempSelectPlatformVersionProblem.png
User Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:52.0) Gecko/20100101 Firefox/52.0
Build ID: 20170323105023
Steps to reproduce:
selected product: support.mozilla.org - Lithium
selected component: Knowlege Base Software
Actual results:
Was able to enter all the info required on the input screen, but when I clicked "Submit Bug" I got a screen that insisted I provide OS version. Went back to input screen and could not find a field to enter this data. "Version" input field is absent.
Expected results:
Once I entered all the required data on all available inputs, I should have been able to click on Submit Bug and successfully post the bug.
|
|
1,368,868 | 12,359,619 |
+++ This bug was initially created as a clone of Bug #1368433 +++
bug 1368433 (and many, many duplicates) report being unable to reach a vast swath of Microsoft cloud-hosted servers due to incorrect stapled OCSP responses. The quick fix in that bug is to disable stapling. To avoid cross-conversations in that largely operational "stop the pain" bug I'm opening this to consider what we should do about stapling in Firefox 54 (shipping in two weeks!) and beyond.
In bug 1368433 comment 26 ekr suggests making stapling "soft-fail" as Chrome does, so let's start with that. The advantage is that stapling can succeed and then we get the performance and privacy benefits we sought. If it fails because the server set their automation up incorrectly it fails over to the CA's server who presumably are better at that kind of thing.
The disadvantage is that if stapling is broken the server likely won't ever know because the site will just work. We should at the very least spit an error into the web console to give sites a fighting chance.
Do we need to distinguish between error reasons? If the stapled response is an outright "REVOKED" status and otherwise completely validly signed can we hard-fail on that, at least? AIUI that's not the kind of error MS had.
|
|
1,363,378 | 12,304,000 |
[Affected versions]:
Firefox 55.0a1 (Build Id:20170508100218)
Firefox 54.0b6 (Build Id:20170508210007)
[Unaffected versions]
Firefox 53.0.2 (Build Id:20170504175505)
Firefox 52.1.1 esr (Build Id:20170504112025)
[Affected platforms]:
Ubuntu 16.04 LTS 64bit
[Steps to reproduce]:
1.Launch Firefox with a clean profile.
2.Enable e10s (if it is not already enabled).
3.Got to https://www.youtube.com webpage.
4.Open 2 random videos.
5.Install Crash Me Now add-on.
6.Crash the browser.
[Expected result]:
After crashing Firefox, the audio stops playback and the opened processes are killed.
[Actual result]:
After crashing Firefox, the audio continues to playback and the opened processes are not killed. The audio can be heard even if starting Firefox again after submitting the crash report.
[Regression range]:
I think that this issue requires finding the regression manually.
I will get back to this as soon as time permits.
[Additional notes]
Please note that this issue occurs only when e10s is enabled.
Please note that the Web Content process is not killed after Firefox crashed.
For more information regarding this issue please observe the following video: https://drive.google.com/open?id=0B94iuIVMr-TER21BbTlncURPRGM
|
|
1,349,320 | 12,164,816 |
I've built an example of a bug I found at: http://labs.jensimmons.com/2017/01-012-bug.html
Firefox doesn't seem to be handling `align-items: end` or `align-self: end` properly when there is more than one item placed in the same area. When those Grid containers are themselves items in a grandparent grid. The alignment seems to leak from the one Grid to the containing Grid, affecting it's alignment. Or something.
See comparison with results in Chrome:
https://monosnap.com/file/bEj3Tm1PxuAI9z51lfK1UYMh1EGSPn.png
|
|
1,349,421 | 12,165,935 |
The following code performs unchecked arithmetic operations on integral data types with the result being used in memset:
https://dxr.mozilla.org/mozilla-central/rev/7ebcd45634eef3711dccf68e4e1390134d48b63b/dom/media/encoder/OpusTrackEncoder.cpp#348
Checking the limits on the variables, it seems that mChannels can have a value of 2, sizeof(AudioDataValue) a value of 2 or 4 depending on the datatype used, and frameToCopy can be INT32_MAX on 32-bit. The last variable seems to be capped by chunk duration, which is StreamTime:
https://dxr.mozilla.org/mozilla-central/rev/7ebcd45634eef3711dccf68e4e1390134d48b63b/dom/media/AudioSegment.h#164
That again seems to be limited to
> const int64_t TRACK_RATE_MAX_BITS = 20;
> const int64_t TRACK_TICKS_MAX = INT64_MAX >> TRACK_RATE_MAX_BITS;
> const int64_t MEDIA_TIME_MAX = TRACK_TICKS_MAX;
So that should be enough to cause an overflow in the memset at least on 32 bit. That again leaves some of the data uninitialized, so I guess it is possible that some of that uninitialized data can be read through the encoder API?
I also wonder why the code uses int to get a duration that is potentially larger than INT32_MAX, but that is a different bug most likely.
We should probably use CheckedInt to ensure that there is no overflow happening.
If the analysis here isn't correct and this code is indeed safe, then we should add an annotation for the static analysis to ignore this particular code.
(This is part of a custom static analysis we have been working on in bug 1279569).
|
|
1,349,991 | 12,172,513 |
I'm seeing *all* child processes (content and gfx) saturating 1 core each and the gfx process consuming more and more memory until it crashes or I stop playback. Upon stopping playback the memory usage ramps down again over the course of several seconds.
The video playing in a background tab does not exhibit this behavior, it has to be visible.
GPU (pid 12156)
Explicit Allocations
6,885.54 MB (100.0%) -- explicit
βββ6,824.60 MB (99.11%) ββ heap-unclassified
ββββββ60.94 MB (00.89%) -- (5 tiny)
βββ60.40 MB (00.88%) ++ heap-overhead
ββββ0.17 MB (00.00%) ββ profiler/profiler-state
ββββ0.14 MB (00.00%) ++ xpcom
ββββ0.13 MB (00.00%) ++ atom-tables
ββββ0.11 MB (00.00%) ββ telemetry
Other Measurements
134,217,727.94 MB (100.0%) -- address-space
βββ134,210,283.62 MB (99.99%) ββ free(segments=132)
βββββββββ7,444.32 MB (00.01%) -- (2 tiny)
βββ7,349.94 MB (00.01%) -- commit
β βββ7,087.55 MB (00.01%) -- private
β β βββ7,079.70 MB (00.01%) ββ readwrite(segments=7232)
β β βββββββ7.59 MB (00.00%) ββ readwrite+writecombine(segments=9)
β β βββββββ0.15 MB (00.00%) ββ readwrite+stack(segments=9)
β β βββββββ0.11 MB (00.00%) ββ readwrite+guard(segments=9)
β β βββββββ0.00 MB (00.00%) ββ execute-readwrite(segments=1)
β β βββββββ0.00 MB (00.00%) ββ readonly(segments=1)
β βββββ153.52 MB (00.00%) ++ image
β βββββ108.87 MB (00.00%) ++ mapped
ββββββ94.38 MB (00.00%) ++ reserved
6,885.54 MB (100.0%) -- heap-committed
βββ6,825.14 MB (99.12%) ββ allocated
ββββββ60.40 MB (00.88%) ββ overhead
10 (100.0%) -- observer-service
βββ10 (100.0%) -- referent
βββ10 (100.0%) ββ strong
ββββ0 (00.00%) ++ weak
6,825.14 MB ββ heap-allocated
1.00 MB ββ heap-chunksize
6,890.00 MB ββ heap-mapped
7,191.41 MB ββ private
7,002.82 MB ββ resident
6,940.80 MB ββ resident-unique
2.30 MB ββ system-heap-allocated
7,524.25 MB ββ vsize
132,587,666.44 MB ββ vsize-max-contiguous
|
|
1,341,409 | 12,085,762 |
STR:
1. Open https://jsfiddle.net/jib1/9f6ztq8v/ and share camera.
Expected result: 1280x720 or 1024x576 or 640x360 (assuming your camera supports it)
Actual result: 640x480
|
|
1,342,000 | 12,091,423 |
User Agent: Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:51.0) Gecko/20100101 Firefox/51.0
Build ID: 20170201180315
Steps to reproduce:
1) Set TimeZone as Europe/Moscow
User Preferences >> General Preferences >> "Timezone used to display dates and times" >> Europe/Moscow
Actual results:
Actual Moscow time == 12:00
Bugzilla.mozilla.org time == 13:00
|
|
1,355,520 | 12,225,566 |
[Affected versions]:
Firefox 52.1.0esr (Build Id:20170410145022)
Firefox 53.0b10(Build Id:20170407033734)
[Unaffected versions]:
Firefox 54.0a2 (Build Id:20170410004019)
Firefox 55.0a1 (Build Id:20170410030221)
[Affected platforms]:
Mac 10.11.6
[Steps to reproduce]:
1.Launch Firefox.
2.Access the https://www.youtube.com/watch?v=ymNFyxvIdaM weblink.
3.Press the Pause button.
4.Unplug headphones.
5.Plug headphones.
6.Click the Play button.
7.Observe that media gets stuck and no longer plays.
[Expected result]:
The video continues to play.
[Actual result]:
The media gets stuck and no longer plays even if switching to another video.
[Regression range]:
I will come back with a regression range asap, if any.
[Additional information]:
Please access the following link: https://drive.google.com/open?id=0B94iuIVMr-TEcktkZjdpc1dpSE0 in order to observe the issue.
Please note that you may need to repeat steps 3, 4 and 5 a couple of times in order to reproduce this issue.
|
|
1,348,432 | 12,155,664 |
User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:55.0) Gecko/20100101 Firefox/55.0
Build ID: 20170317111607
Steps to reproduce:
Visit YouTube and play any video. It plays fine. Click to play another video.
Actual results:
That, and all future videos play without audio, unless the page is refreshed. The page must be refreshed for every new video. Started after 2017-03-17 Nightly. Most likely after the second (~2PM PST) update - not the morning one. Tested with new profile. Not related to YouTube mute button.
Expected results:
You should be able to request a video in the same tab without having to refresh the tab, to play the audio.
|
|
1,343,691 | 12,108,564 |
RTCP stats are missing for both video and audio in the following fiddle.
STRs:
1. Open https://jsfiddle.net/jib1/eboqozdy/ and share camera.
2. Wait ~5 seconds
Expected result:
Sender side:
Timestamp: 18:43:24 GMT-0500 (EST) Type: outbound-rtp
SSRC: 783676504 Sent: 12165 packets (12.50 MB)
Dropped frames: 124
RTCP: Timestamp: 18:43:23 GMT-0500 (EST) Type: inbound-rtp
SSRC: 783676504 Received: 12165 packets (12.14 MB) Lost: 0
Discarded packets: undefined Jitter: 0.375
Receiver side:
Timestamp: 18:43:24 GMT-0500 (EST) Type: inbound-rtp
SSRC: 783676504 Received: 12165 packets (12.38 MB) Lost: 0
Discarded packets: 0 Jitter: 0.375
RTCP: Timestamp: 18:43:23 GMT-0500 (EST) Type: outbound-rtp
SSRC: 783676504 Sent: 12165 packets (12.14 MB)
Dropped frames: undefined
Actual result:
Sender side
Timestamp: 18:43:24 GMT-0500 (EST) Type: outbound-rtp
SSRC: 783676504 Sent: 12165 packets (12.50 MB)
Dropped frames: 124
Receiver side
Timestamp: 18:43:24 GMT-0500 (EST) Type: inbound-rtp
SSRC: 783676504 Received: 12165 packets (12.38 MB) Lost: 0
Discarded packets: 0 Jitter: 0.375
Regression range:
15:12.90 INFO: Last good revision: 935e36fde31c6ecd8321beb29d896e42a70aecd0
15:12.90 INFO: First bad revision: 126348e718d03dec640b30b5def70fce8aa71527
15:12.90 INFO: Pushlog:
https://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=935e36fde31c6ecd8321beb29d896e42a70aecd0&tochange=126348e718d03dec640b30b5def70fce8aa71527
|
|
1,343,920 | 12,110,717 |
[Affected versions]:
Firefox 52.0b2 (Build Id:20170130065342)
Firefox 52.0b9 (Build Id:20170223185858)
Firefox 54.0a1 (Build Id:20170302030206)
Firefox old Nightly 52.0a1 (Build Id:20161022030204)
Firefox 52.0b1 (Build Id:20170124094647)
Firefox ESR 52.0 (Build Id:20170227085837)
[Affected platforms]:
MacOS
[Steps to reproduce]:
1.Launch Firefox.
2.Access the https://www.youtube.com/watch?v=gjQmTkMjIBg weblink (or any YouTube video).
3.Mute the video.
4.Unplug headphones.
5.Plug headphones.
6.Click on the Play button.
7.Observe that the video resumes and the audio is outputted in the headphones even tho the video is muted.
[Expected result]:
The video continues to play but the audio is muted.
[Actual result]:
The video continues to play but the audio is outputted even tho the video is displayed as muted.
[Regression range]:
This is an old regression:
The latest good build is old Nightly from 2016-10-21
The first bad build is old Nightly from 2016-10-22
Pushlog:https://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=f0f1aaf051d6798e1e73d1feee07ca847333167a&tochange=5639a9f476d08f300c079117e61697f5026b6367
[Additional information]:
This issue is reproducible using headphones that are connected via jack or USB.
This issue is not reproducible on Windows and Linux.
Please note that this issue does not affect Firefox 51.0.1,Firefox old Nightly from 2016-10-21 and Firefox ESR 45.8.0.
|
|
1,345,612 | 12,127,602 |
User Agent: Mozilla/5.0 (X11; Linux x86_64; rv:55.0) Gecko/20100101 Firefox/55.0
Build ID: 20170308110300
Steps to reproduce:
not possible to add certificate exeptions on https://[::1]:1337 or https://[fd00::9ec7:a6ff:fe70:cb36]/ (=https://fritz.box/ consumer router)
I am using Nightly. I have this problem for weeks (or maybe longer undetected), but reported it only via the feedback form. Can you reproduce this?
Actual results:
if I confirm the popup to add the exception, nothing happens. If I connect to https://localhost:1337 or https://fritz.box (consumer router) it is possible to add certificate exceptions.
I was unsure if I should comment on this 6 year old bug which was about the same, but was fixed 4 years ago: https://bugzilla.mozilla.org/show_bug.cgi?id=633001
Expected results:
It should be possible to add certificate exceptions for https://[ipv6-address] as I can do it with https://ipv4-address/ and https://ipv6-domain/
|
|
1,323,628 | 11,920,335 |
Created attachment 8818774
log fragment
I don't see a timebomb set for that time anywhere in https://hg.mozilla.org/mozilla-central/rev/edcd2fb87124 but there must be one, since it began failing on mozilla-inbound, mozilla-central, and autoland, but not on mozilla-aurora, and retriggering on a week-old mozilla-central push it now fails 100%.
https://treeherder.mozilla.org/logviewer.html#?job_id=7917409&repo=autoland
|
|
1,318,839 | 11,872,884 |
Firefox 50.0 64-bit on Windows 8.1 64-bit on a system with Intel i7-4700MQ (4 cores, 8 hyperthreads)
After upgrading my default Firefox profile from Firefox 49.0.2 64-bit to 50.0 64-bit, the firefox.exe (e10s disabled due to addons) has now a thread which always consumes a whole thread. Filing it against Telemetry because TelemetryAccumulate is four times part of the stack according to Process Explorer > firefox.exe > Properties > Threads > ucrtbase.DLL > Stack
Please redirect to the appropriate component if necessary.
Here are stacks as returned by Process Explorer:
xul.dll!XRE_IsContentProcess+0x24aba
xul.dll!XRE_TelemetryAccumulate+0x36f55
xul.dll!XRE_TelemetryAccumulate+0x36e37
xul.dll!XRE_TelemetryAccumulate+0x36dfa
xul.dll!XRE_TelemetryAccumulate+0x3777f
nss3.dll!PR_JoinThread+0x1a6
nss3.dll!PR_NativeCreateThread+0x15a
ucrtbase.DLL!crt_at_quick_exit+0x7d
KERNEL32.DLL!BaseThreadInitThunk+0x22
ntdll.dll!RtlUserThreadStart+0x34
USER32.dll!PeekMessageW+0xfa
USER32.dll!PeekMessageW+0x116
xul.dll!mozilla_dump_image+0x60f660
xul.dll!XRE_IsContentProcess+0x2344a
xul.dll!XRE_IsContentProcess+0x2327e
xul.dll!XRE_TelemetryAccumulate+0x36f01
xul.dll!XRE_TelemetryAccumulate+0x36e37
xul.dll!XRE_TelemetryAccumulate+0x36dfa
xul.dll!XRE_TelemetryAccumulate+0x3777f
nss3.dll!PR_JoinThread+0x1a6
nss3.dll!PR_NativeCreateThread+0x15a
ucrtbase.DLL!crt_at_quick_exit+0x7d
KERNEL32.DLL!BaseThreadInitThunk+0x22
ntdll.dll!RtlUserThreadStart+0x34
ntdll.dll!RtlLeaveCriticalSection
nss3.dll!PR_Unlock+0x2c
xul.dll!XRE_IsContentProcess+0x24ac8
xul.dll!XRE_TelemetryAccumulate+0x36f55
xul.dll!XRE_TelemetryAccumulate+0x36e37
xul.dll!XRE_TelemetryAccumulate+0x36dfa
xul.dll!XRE_TelemetryAccumulate+0x3777f
nss3.dll!PR_JoinThread+0x1a6
nss3.dll!PR_NativeCreateThread+0x15a
ucrtbase.DLL!crt_at_quick_exit+0x7d
KERNEL32.DLL!BaseThreadInitThunk+0x22
ntdll.dll!RtlUserThreadStart+0x34
xul.dll!XRE_IsContentProcess+0x24b2e
xul.dll!XRE_IsContentProcess+0x24aba
xul.dll!XRE_TelemetryAccumulate+0x36f55
xul.dll!XRE_TelemetryAccumulate+0x36e37
xul.dll!XRE_TelemetryAccumulate+0x36dfa
xul.dll!XRE_TelemetryAccumulate+0x3777f
nss3.dll!PR_JoinThread+0x1a6
nss3.dll!PR_NativeCreateThread+0x15a
ucrtbase.DLL!crt_at_quick_exit+0x7d
KERNEL32.DLL!BaseThreadInitThunk+0x22
ntdll.dll!RtlUserThreadStart+0x34
The issue and stack persist if turn off Telemetry in the settings and restart Firefox.
|
|
1,319,242 | 11,877,395 |
function f(x) {
return ((-4294967295 | 0) + (x | 0) | 0) + (x, -4);
}
f(Infinity);
print(f(-2147483649));
$ ./js-dbg-64-dm-clang-darwin-0534254e9a40 --fuzzing-safe --no-threads --ion-eager testcase.js
2147483644
$ ./js-dbg-64-dm-clang-darwin-0534254e9a40 --fuzzing-safe --no-threads --baseline-eager --no-ion testcase.js
-2147483652
Tested this on m-c rev 0534254e9a40.
My configure flags are:
CC="clang -Qunused-arguments" CXX="clang++ -Qunused-arguments" AR=ar AUTOCONF=/usr/local/Cellar/autoconf213/2.13/bin/autoconf213 sh /Users/skywalker/trees/mozilla-central/js/src/configure --target=x86_64-apple-darwin14.5.0 --disable-jemalloc --enable-debug --enable-more-deterministic --with-ccache --enable-gczeal --enable-debug-symbols --disable-tests
python -u ~/funfuzz/js/compileShell.py -b "--enable-debug --enable-more-deterministic" -r 0534254e9a40
autoBisect shows this is probably related to the following changeset:
The first bad revision is:
changeset: https://hg.mozilla.org/mozilla-central/rev/07358be0ec02
user: Sander Mathijs van Veen
date: Tue Oct 11 07:04:00 2016 +0200
summary: Bug 1295130 - Fold AddI opcode with constant into other AddI with constant r=nbp
Jan/Nicolas, is bug 1295130 a likely regressor?
Note that this is quite difficult to ignore as this seems to involve addition, so setting [fuzzblocker] as it blocks differential testing.
|
|
1,320,170 | 11,886,183 |
The getUserMedia call made by webrtcUI.jsm fails and calls the error callback.
Steps to reproduce:
1. Load https://mozilla.github.io/webrtc-landing/gum_test.html
2. Click "Screen"
3. In the door hanger, select a screen instead of "No Screen"
Expected result: a preview of the screen should appear
Actual result: no preview, but clicking the "Allow" button works, and the page receives a stream as expected.
mozregression says https://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=39efc7bfb839484ef121a2f275481770906d9021&tochange=5230f39d8f21270414aea3cdfb97914fc6acf4c1 -> bug 1319045
One thing I can't explain is that the preview works fine in the browser-chrome mochitest I'm writing to cover the screensharing UI. I would like us to have test coverage for this.
|
|
1,311,508 | 11,799,004 |
Created attachment 8802663
embed.xpi
Steps to reproduce:
* I am using nightly.
* XPI consists of - `function startup(aData) { console.log('aData.webExtension:', aData.webExtension) }`
* Disable signing so you can install a XPI permanently.
* Drag XPI and install it
* Restart Firefox, it is undefined, looping it with a nsITimer and you will see aData.webExtension never becomes available
Demo attached.
|
|
1,316,324 | 11,847,514 |
Build: 6.0b2
Device: iPad Air 2 (10.0.2)
Prerequisites: FxA with synced data (History, Bookmarks)
Steps to reproduce:
1. Log in to FxA and perform Synchronization
2. Open a new tab
3. Background Firefox
4. Restore Firefox
5. Switch between Top Sites, Bookmarks, History tabs
Actual results:
- The tabs are empty. No synced data is displayed. After closing Firefox and re-opening the Bookmarks, History and other sync data will be displayed.
Expected results:
- Bookmarks History and other synced data should be displayed after sync action is performed.
See video of the issue: https://www.youtube.com/watch?v=NDFs5SghQ_g
Note:
- Issue encountered on all devices.
|
|
1,316,404 | 11,848,487 |
+++ This bug was initially created as a clone of Bug #1313418 +++
[Tracking Requested - why for this release]: Regression
STR:
1. Open this website page - https://bugzilla.mozilla.org/enter_bug.cgi?product=Core
2. Open second "Platform" dropdown menu list
3. See that font in text on dropdown menu list is using other font family than on button. It's consistency and compatibility issue, comparing to other browsers.
|
|
1,317,281 | 11,857,358 |
[Preconditions]:
- SonicWALL Firewall Appliance has to be installed and active:
https://www.sonicwall.com/products/sonicwall-nsa/
- The "Gateway Anti-Virus" option needs to be enabled for SonicWALL.
[Affected versions]:
all
[Affected platforms]:
Windows (all)
[Steps to reproduce]:
1. Download the following 64-bit stub installer:
https://bugzilla.mozilla.org/attachment.cgi?id=8805254
2. Run the stub installer as administrator.
3. Follow the setup wizard's steps to install Firefox.
[Expected result]:
Firefox is successfully installed via stub installer.
[Actual result]:
- The stub installer hangs in the "Downloading Firefox" stage for a
while and then the following alert is displayed:
Gateway Anti-Virus Alert [...] This request is blocked by the
Firewall Gateway Anti-Virus Service. Name: InstallCore.A_237
(Adware) blocked.
- As a result, the stub installer is blocked from downloading
Firefox and the install process is cancelled.
[Regression range]:
This issue is not a regression.
[Additional notes]:
- We are using SonicWALL Firewall Appliance in all of our offices,
but for one of them we've recently enabled a specific option:
Gateway Anti-Virus.
- Based on a few tests across different offices, we concluded that
only the office that had Gateway Anti-Virus enabled was showing
this problem.
- Initially, we've noticed that suddenly both 32-bit and 64-bit
stub installers were being blocked.
- Starting last Thursday and without any changes being made to our
office's firewall rules, we noticed that 32-bit installers were
no longer being blocked and I've double-checked that again today.
As of this moment, only 64-bit stub installers get blocked, with
the same alert notification.
- These are the stub installers we've used to test the firewall's
behavior:
- 64-bit (blocked):
* https://bugzilla.mozilla.org/attachment.cgi?id=8805254
- 32-bit (initially blocked, but now they're not):
* https://mzl.la/2eXc5VD
* https://mzl.la/19V7f56
* https://mzl.la/2fyuVAG
* https://mzl.la/2g7bQcV
- We tested the same firewall setup with other stub/web installers,
to see whether it's something isolated to the Firefox stub:
- Google Chrome: not blocked
- Opera: blocked, reason: MalAgent.I_12705 (Trojan)
- Torch: not blocked
- Pale Moon: not blocked
- Vivaldi: n/a, no web installer available
- Maxthon: n/a, no web installer available
- Yandex: n/a, no web installer available
- Dropbox: not blocked
- Google Drive: not blocked
|
|
1,322,121 | 11,906,534 |
Build 6.0b6
Device iPad Air 2
iOS 10.1.1
Steps to reproduce:
1. Open a new tab and tap on the bookmark panel
2. Open a new tab and go to youtube.com
3. Close Firefox
4. Restore Firefox
5. Go to tab that was opened in the bookmark panel
Actual results:
- Bookmarks are not displayed.
Expected results:
- Bookmarks should be displayed.
See video: https://www.youtube.com/watch?v=Cc661Z0GmO4
|
|
1,325,049 | 11,934,423 |
Steps to reproduce:
1. Enable e10s multi (set dom.ipc.processCount to 2 in about:config)
2. Load https://people-mozilla.org/~fqueze2/webrtc/ and share the camera.
3. Load http://mozilla.github.io/webrtc-landing/gum_test.html in a second tab, and share the camera.
4. Click the "Stop" button.
5. (Mac) Click the Camera icon in the menubar.
Expected result:
The menu should offer to control sharing for the first tab.
Actual result:
The menu has only one disable item, saying "Sharing Camera with 0 tabs".
This is because when receiving the "webrtc:UpdatingIndicators" message from any content process, webrtcUI.jsm clears the webrtcUI._streams array that contained info about active streams from all processes: http://searchfox.org/mozilla-central/rev/cc2a84852bd4e6f6d8d4d5b17b8382bb5d005749/browser/modules/webrtcUI.jsm#217
|
|
1,325,695 | 11,941,036 |
Created attachment 8821624
Testcase
[Tracking Requested - why for this release]: new feature
It seems like the max-width used for wrapping in the insecure field warning isn't recalculated when the focus is moved to a new field (having a different width). This can look very bad and lead to clipping of text in the warning or in the opposite direction leads to useless ugly space below or to the right of the warning.
Bug 1317882 made this recalculate in the XBL constructor but perhaps that constructor isn't called when moving between fields.
Tim/Sean, since you both investigated bug 1317882, do you have ideas to fix this or would you be able to help out for 52? Thanks. Perhaps the other approach in that bug would have fixed this?
|
|
1,388,129 | 12,544,076 |
STRs:
1. Open https://blog.mozilla.org/webrtc/fiddle-week-downscale-video-peerconnection/
2. Click the "Result" tab of the embedded fiddle, and share camera.
Expected result:
- Video elements appear at 640x480 and 320x240 respectively, and stay there.
Actual result:
- Video elements appear at 640x480 and 320x240 initially, but after a second or two,
the latter shrinks to 120x90, and then after a sec or two more, shrinks again to 80x60.
Tested on OSX and Win10.
Regression range:
4306:00.22 INFO: Narrowed inbound regression window from [23ad1168, 5bbdb7d3] (4 revisions) to [c7428449, 5bbdb7d3] (2 revisions) (~1 steps left)
4306:00.22 INFO: No more inbound revisions, bisection finished.
4306:00.22 INFO: Last good revision: c7428449127566105bdd94b2823b01e0e5e007d5
4306:00.22 INFO: First bad revision: 5bbdb7d36ee3c136a0ed03be9d5b012d05dfd08e
4306:00.22 INFO: Pushlog:
https://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=c7428449127566105bdd94b2823b01e0e5e007d5&tochange=5bbdb7d36ee3c136a0ed03be9d5b012d05dfd08e
|
|
1,388,990 | 12,553,089 |
User Agent: Mozilla/5.0 (X11; Linux x86_64; rv:57.0) Gecko/20100101 Firefox/57.0
Build ID: 20170809100326
Steps to reproduce:
Step 1: Go to https://en.wikipedia.org/
Step 2: In that page , click on the "English" link it will redirect the page to https://en.wikipedia.org/wiki/Main_Page and the Go back one page button gets activated
Step 3: Now click hamburger menu icon of firefox and choose Customize..
Step 4: Drag and Drop the sidebar icon to the left of the Go back one page button and click Done
Actual results:
The Go back one page button which was activated becomes inactive and it is not clickable.
Expected results:
The Go back one page button should remain active and one clicking it , the page should be redirected to the https://en.wikipedia.org/
|
|
1,378,420 | 12,448,725 |
[Affected versions]:
- latest Nightly 56.0a1
[Affected platforms]:
- Mac OS X 10.11.6
- Ubuntu 16.04 x64 LTS
[Steps to reproduce]:
1. Launch Firefox.
2. Go to: https://en.wikipedia.org/wiki/Main_Page
3. Load the above site in Netmonitor Launchpad.
4. Select a request and open context menu via right click.
[Expected result]:
- The context menu is properly opened.
[Actual result]:
- The context menu is not opened.
[Regression range]:
- I will follow up with a regression range if any.
[Additional notes]:
- Here is a screencast with this issue: https://drive.google.com/file/d/0B6qEQD4BO4AlMjFkQ05OZDZ4Y1U/view
- This was reproduced with "devtools-launchpad": "=0.0.75" version
- Please note that context menu is working as expected on Windows
- I can see the following error being displayed in browser console: Error: could not find pref branch devtools.styleeditor.enabled netmonitor.js:35848:16
|
|
1,394,455 | 12,606,411 |
Created attachment 8901868
Tab hovering.gif
[Affected versions]:
Firefox 57.0a1 (Build Id:20170828100127)
[Affected platforms]:
Windows 10 64bit.
Mac 10.11.6.
Ubuntu 16.04 64bit.
[Steps to reproduce]:
1. Launch Firefox with a clean profile.
2. Access the https://www.youtube.com/ webpage.
3. Open 1 random video in a new tab without focusing it.
4. Hover over the tab.
[Expected result]:
The media is still blocked for that tab since it was not visited nor the Play Tab button was pressed.
[Actual result]:
The media gets unblocked and the Play Tab button dissapears.
[Regression range]:
This is a recent regression:
Pushlog: https://hg.mozilla.org/integration/autoland/pushloghtml?fromchange=2cbd34e59b97a522498c5b5dbd22677935a025e5&tochange=3bbd8e25df3d92464b97bcb42413febd9041af6f
[Additional Information]
Please observe the attached video for further information regarding this issue.
|
|
1,394,588 | 12,607,671 |
Created attachment 8902011
animation.png
User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:57.0) Gecko/20100101 Firefox/57.0
Build ID: 20170828100127
Steps to reproduce:
1. put Library button on the toolbar (if not yet)
2. open random web page
3. hit star button and click done
4. see broken Library button animation - it shifts down and then up on the navigation toolbar
Looks like it started with https://hg.mozilla.org/integration/autoland/rev/b5a5c38a0add
So regressed by Bug 1384953.
I didn't test animation when saving to Pocket.
|
|
1,386,224 | 12,525,467 |
Discuss & setup WNP for 55.0
|
|
1,386,558 | 12,528,462 |
A number of reports where sandboxing level 3 broke things said they needed to go back to level 1, not 2, to get things working again.
Manual verification shows that this is true: setting sandboxing to level 2 has no effect. That's likely because the preference is checked before actually being available.
|
|
1,387,079 | 12,533,269 |
[Affected versions]:
- Firefox 55.0-build 2
- Firefox 54.0.1
- Firefox 57.0a1
[Affected platforms]:
- Ubuntu 16.04
[Steps to reproduce]:
1. Start Firefox
2. Visit twitter and login using an account
3. Scroll the timeline until you hit a video (gif)
Alternative steps:
1. Start Firefox
2. Visit https://www.linkedin.com/in/bobhutchinsfacebookadexpert/detail/recent-activity/ and login
3. Start the video and scroll the page
1. Start Firefox
2. Visit pinterest and login
3. Click a long image and scroll the page
[Expected result]:
- User can scroll through the page nice and smooth.
[Actual result]:
- Noticeable lag can be noticed when scrolling, CPU usage jumps to 100% when scrolling.
[Regression range]:
- This is not a recent regression since it reproduces on 54.0.1 as well, will try and find a regression though ASAP.
[Additional notes]:
- Here is the Graphics section info from about:support https://pastebin.com/x2P3tkcf
|
|
1,387,718 | 12,539,254 |
User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:57.0) Gecko/20100101 Firefox/57.0
Build ID: 20170804193726
Steps to reproduce:
Open Firefox with many tabs. Pick a middle/center tab to drag so that there are many tabs to the right and left of the tab strip, tabs that are not visible on the screen.
Drag the tab to the left of the tabs strip, to make it the first tab.
Drag the tab to the right of the tab strip, to make it the last tab. When dragging right move your mouse to the right of the right arrow on the tab strip, to hover over where the new tab icon (+) is or where the tabs dropdown list arrow is.
Actual results:
When dragging right, moving the mouse to the right of the right arrow, the tab stops moving and you can't move it past that position or to be the last tab
When dragging right, moving the mouse so that it hovers over the right arrow, but not to the right of it, or when dragging to the left :
The tab doesn't stop moving, but moves very slowly, too slow. The tab strip keeps moving/scrolling very slowly, but the tab doesn't change its position relative to the other tabs . Reaching the first or last tab with that speed will take ages.
Expected results:
When the tab is dragged, tab movement should not stop, and should not move slowly, but move faster in a relatively fast constant speed, or accelerate with time.
|
|
1,381,085 | 12,474,444 |
STR:
- visit https://webvr.info/samples/03-vr-presentation.html
- Click the "Enter VR" button on the page (tested with an Oculus CV1 headset)
- Resize the 2d window to be more than 1/4 the area of the monitor (tested with a UHD resolution display)
- As each part of the window is repainted, including the browser chrome, it is filled with solid white.
- Resizing the window back to a smaller size results in the window being repainted normally again.
- After pressing "Exit VR", the effect is no longer seen.
A mozregression points to bug 1375743:
Enable Advanced Layers on Nightly Windows 8+ builds. (bug 1375743, r=milan)
|
|
1,383,138 | 12,494,371 |
User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:56.0) Gecko/20100101 Firefox/56.0
Build ID: 20170721030204
Steps to reproduce:
1. Download latest Nightly (20170721030204) with clean profile
2. Open it, should be 3 first-run tabs
3. Right-click and select "Bookmark all tabs"
4. Dialog has only "Add bookmark" button"
5. Click button
Actual results:
Folder named '[Folder name]' was created in Bookmarks Menu. No bookmarks present in folder.
Expected results:
Dialog should have folder selector and text box for folder name. Created folder should have the 3 bookmarked tabs.
|
|
1,384,094 | 12,504,083 |
Created attachment 8889875
hang.html
Reproducible: always
Steps To Reproduce:
0. Start with new profile
1. Open attached html and wait until the tab loading circle turns green
2. Exit browser and restart
3. Open Newtab Page (click [+] button)
---- observe CPU and Memory usage
Actual Results:
Even if there are about:home and Newtab page, browser consumes High CPU and memory.
Expected Results:
Very low CPU and memory usage.
WORKAROUND:
browser.newtabpage.enabled = false
and
browser.newtabpage.activity-stream.enabled = false (if activity-stream enabled)
|
|
1,391,227 | 12,574,802 |
Created attachment 8898261
new tab control completely blending out
Direct follow-up to bug 1387750 as requested.
Re-stating my comments for completeness:
I just tried nightly 2017-8-16 on Win 7 x64. The issue is NOT completely solved. Although the tab backgrounds are OK, the tab controls (new tab "+", scroll arrows, drop-down chevron) can still be completely invisible because the background is still non-fogged glass. At the very least it is extremely low contrast, at its worst it completely blends away.
attached screenshots:
1) Example of the new tab control completely blending out into the background.
2) The left window corner is fogged so it provides some contrast with the control (but vanishes completely if the arrow is in a disabled state which is also wrong). The right side of the tab bar isn't fogged (because the fog is behind the caption controls), which is still a problem.
If you want to use these kinds of controls on glass, maybe you need to outline them?
|
|
1,338,740 | 12,058,906 |
Oops.
|
|
1,329,616 | 11,966,642 |
STR:
1. Start Firefox in non-E10S mode, and have NVDA running.
2. Go to Gmail's inbox.
3. Arrow up and down.
Result: Every few lines, NVDA will not speak the newly highlighted conversation item, but speak the document or window name and then something about "unknown".
4. Press up or down arrow.
Result: NVDA may, or may not, revert to speaking the newly highlighted conversation, or continue to speak Unknown for at least one more arrow key press.
Notes:
1. This does not happen with E10S on.
2. It is not confined to Gmail, but happens on many web sites, including Bugmail, a directory listing of archive.mozilla.org when tabbing through, etc., etc.
Last good build is 2016-12-15, first bad nightly build is 2016-12-16. Regression window:
https://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=7652a58efa46f1c57c94bba26efc5d53b6184e83&tochange=63b447888a6469b9f6ae8f76ac5f0d7c6ea239da
Primary suspect is Bug 1323521: Prevent Windows touchscreen support from instantiating a11y. There is one more a11y-related patch in that period, but that's a patch by Alex for the a11y DOM work, which does not appear to be of impact here.
Aaron, could you take a look to see if there's a chance that this could be happening? The symptoms look like a wm_getobject failing even though a11y has already be instanciated.
I am on Windows 10 on a Microsoft Surface Book, and it does have a touch screen. No other Windows machine without a touch screen currently available, so I cannot test whether this behaves differently on a non-touchscreen device.
|
|
1,330,257 | 11,973,404 |
In Firefox 49, LastPass worked with Firefox for Android (albeit in a limited fashion, requiring you to copy/paste usernames/passwords, as opposed to them actually being filled in like with Chrome and other Android applications).
In Firefox 50, LastPass doesn't work at all. If you tap the LastPass notification, it just displays a toast that reads 'LastPass was unable to auto-fill this app'.
Putting this in Keyboards/IME but uncertain of the mechanism LastPass actually uses to auto-fill applications.
|
|
1,136,707 | 9,956,440 |
Steps to reproduce:
1) Use Nightly on Windows Vista or later or Mac (I tried Windows 8.1)
2) Open a new non-e10s window.
3) Load http://media.junglecode.net/test/dash-eme/sintel.html
4) Observe the video play.
5) Flip use browser.eme.ui.enabled to true in about:config
6) Open Preferences
7) Open the Content pane
8) Uncheck Play DRM Content
9) Reload http://media.junglecode.net/test/dash-eme/sintel.html
Actual results:
An infobar says I must enable DRM for the video to play and the video does not play.
Expected results:
Expected the Play DRM Content pref not to affect the Clear Key key system, since Clear Key is mere encryption and not DRM. (There's no Robustness&Compliance regime for Clear Key.)
|
|
1,120,810 | 9,785,960 |
Created attachment 8547978
Debugger.png
User Agent: Mozilla/5.0 (Windows NT 6.1; rv:36.0) Gecko/20100101 Firefox/36.0
Build ID: 20150112004004
Steps to reproduce:
I've added breakpoints in js file for an "input" event of an input element
Actual results:
When above event fires and debugger starts , it is getting struck and doesn't allow me to click on anywhere , even the next tab or resume button unless I minimize the firefox (36.0a2 Developer Edition) and maximize it again
Expected results:
It should not get struck and all debugging buttons should be available
|
|
1,121,748 | 9,796,338 |
This has originally been found while investigating on bug 1119420 and bug 1119381. Hereβs a way to reproduce it:
β’ Settings > Language > Arabic
β’ long-press the power button
Expected: the power menu appears.
Actual: the menu is blank.
It seems that in some cases, Gecko does not render elements any more in RTL blocks (everything is fine in LTR). It looks like static elements are displayed properly, but dynamically created/modified elements can be completely blank.
When inspecting such elements in the WebIDE, everything gets back to normal.
Hereβs a rough regression window, using the latest Gaia on my Keon:
β’ Gecko-7b34913: works (GeeksPhone 2015-01-07 build)
β’ Gecko-58c53ce: buggy (GeeksPhone 2015-01-09 build)
This is a blocker for many RTL-related items.
|
|
1,097,928 | 9,581,846 |
Connectivity: wifi
SIM used:
Gaia/device: flame/ central (2.2.0.0-prerelease)
Build number: eng.cltbld.2014112.073256
Platform version: 36.0a1
Build identifier: 2014112040208Git commit info: 2014-11-11 18:27:28 5ae28ff1
steps to reproduce:
1. Launch marketplace-dev on your flame
2. Sign in
3. Search for :paid
4. Start the purchase of a paid app
expected behavior:
Enter PIN loads
actual behavior:
trusted UI loads and is stuck at the "Loading" screen due to [JavaScript Error: "Error: Exposing privileged accessor properties is prohibited" {file: "https://marketplace-dev.mozflare.net/mozpay/spa/js/main.min.js?bust=6777619d400789e8143c8a6258d90eb404e35f06" line: 64}]
This is currently blocking our testsuite on fxos to be turned on.
NOTE: purchase works fine on 2.1
|
|
1,103,637 | 9,626,411 |
User Agent: Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:33.0) Gecko/20100101 Firefox/33.0
Build ID: 20141013200257
Steps to reproduce:
High
My googles agendas disappear if I don't have any internet connexion at thunderbird startup :
linux ubuntu 14.04
thunderbird 31.2.0
lightning 3.3.1
provider for google agenda 1.0.2
messaging menu and unity launcher integration 1.3.1
thunderbird message filter import/export enhanced 1.0.3
Actual results:
I start th, my google agendas are shown
I close the connection, I stop th
I re-start th, my google agendas are not shown
Expected results:
At re-start, my google agendas use to be shown even when connexion was down. Apparently the option display even out connexion does'nt work anymore.
Thank you for helping
|
|
1,107,702 | 9,661,806 |
We have a confirmed case of GetAdaptersInfo reliably failing for one of our users, which is something we've seen telltale signs of in telemetry and statistics from TokBox. We need to determine why, and whether there is something we can improve here.
|
|
1,132,034 | 9,907,805 |
User Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:37.0) Gecko/20100101 Firefox/37.0
Build ID: 20150211004833
Steps to reproduce:
On a fresh profile play any youtube video more than a minute long (preferably one with strong audio/video cues) using the html5 player. Quality does not seem to matter. Be sure to have all media source extension options enabled (https://www.youtube.com/html5 should show that everything is supported).
Actual results:
After about 30sec to a minute audio and video are clearly out of sync. Seeking fixes temporarily effectively resetting de-syncing process but not eliminating it.
Expected results:
Normal video/audio playback.
|
|
1,213,775 | 10,828,820 |
User Agent: Mozilla/5.0 (Windows NT 6.3; WOW64; rv:41.0) Gecko/20100101 Firefox/41.0
Build ID: 20150929144111
Steps to reproduce:
Use MediaRecorder to record a video (larger than 2 second). Then create local url by URL.createObjectURL on this blob object.
Actual results:
Warning message was issued onto brower console when I set this local url into src abribute of <video>, it shows this local url could not be decode. But it working fine when the recoded video less then or equal 2 second.
Expected results:
It always working fine before the Latest update of FireFox browser.
|
|
1,196,398 | 10,636,746 |
Created attachment 8650054
ff_video_corrupt.png
User Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9_5) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/44.0.2403.155 Safari/537.36
Steps to reproduce:
1. open firefox
2. navigate to local file (file://) or remote url (http://) where you can open (stream) a video mp4 file which is > 3Gb.
In my case I have a mp4 video file = 3.9 Gb.
I've tested this on FF version 40 on Mac (2 machines) and virtual Windows 7.
Result is the same and doesn't matter if file is opened via http or locally.
File is valid - it plays in Chrome, Safary, IE 9.
Other files recorded on the same device which are < 1.9Gb works in FF without a problem.
Actual results:
Firefox shows a black page with the "Video can't be played because the file is corrupt" message. See attached file.
Expected results:
Video should be played
|
|
1,197,152 | 10,643,850 |
User Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.10; rv:41.0) Gecko/20100101 Firefox/41.0
Build ID: 20150817163452
Steps to reproduce:
- start Firefox with a new profile
- ensure the Firefox is the active application!!
- put Mac to sleep (Apple-menu -> Sleep)
- wait for 10-40 seconds to ensure the Mac is in sleep mode
- bring back Mac from sleep (press shift key)
- click + to open a new tab several times (2-5 times), because nothing happens
Actual results:
Nothing happens for up to 10-15 seconds, after that the new tabs are opened
Expected results:
A new tab should have opened immediately
|
|
1,217,677 | 10,865,948 |
Created attachment 8677808
Wireshark
User Agent: Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/46.0.2490.71 Safari/537.36
Steps to reproduce:
Receive video at 1920 Kbps
Actual results:
Video has low frame rate and video lag. We can see packet drop in about:webrtc while wireshark trace shows no packet drop. Screenshots are attached. This problem is observed only when PC connected to WiFi
Expected results:
video is smooth
|
|
1,233,005 | 11,010,127 |
User Agent: Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20100101 Firefox/38.0
Build ID: 20150828185855
Steps to reproduce:
Since bug #672509 the Gecko SDK is not built anymore as part of the xulrunner build (which has been stopped as of Firefox 42), but as part of Firefox's build.
However, only the nightly Aurora and Mozilla Central builds happen to produce the SDK. It is not made available for Beta releases nor Stable releases of Firefox.
If my analysis is correct, this means that mozilla does not produce anymore builds of the SDK for beta and stable.
Expected results:
I have seen various posts of people looking for the SDK and I would like mozilla to consider releasing it again for beta and stable releases of Firefox. It is currently missing for Firefox 42 and 43.
|
|
1,235,395 | 11,033,205 |
User Agent: Mozilla/5.0 (X11; FreeBSD amd64; rv:43.0) Gecko/20100101 Firefox/43.0
Build ID: 20151225125557
Steps to reproduce:
Execute the whine.pl cronjob according to http://bugzilla.readthedocs.org/en/5.0/installing/optional-post-install-config.html?highlight=whine
Actual results:
the script broke due to "Undefined subroutine &main::generate_email called at ./whine.pl line 355"
Expected results:
Working whine mails should have been sent
It looks like there is no generate_email() routine defined in Bugzilla/Mailer.pm in 5.0.2., while there is one in the current head/master
|
|
1,245,813 | 11,137,109 |
In fact, https://hg.mozilla.org/integration/fx-team/annotate/tip/toolkit/content/widgets/browser.xml#l1210 throws a βNOT_IMPLEMENTEDβ error.
Right now Iβve put it in a tryβ¦catch block to catch this, but it blocks correct functioning of detaching chat windows, for example.
|
|
1,254,694 | 11,229,374 |
After upgrading from Firefox 44.0.2 64 bit to 45.0 64 bit on a Windows 8.1 machine, my search engines have all lost their keywords and their order (it's alphabetic now). I wasn't able to reproduce with a new profile created in 44.0.2 and launching that in 45.0 but will test with a copy of my backup tomorrow.
Further information:
- The profile is very old. But this shouldn't matter since the search engines got migrated from files to a database like structure.
- Some of the search engines got added with the extension 'Add To Search Bar': https://addons.mozilla.org/firefox/addon/add-to-search-bar/
- amazon.de is twice in my list of search engines (might have been done by myself): Once as 'amazon.de', once as 'Amazon.de'
|
|
1,269,713 | 11,377,233 |
User Agent: Mozilla/5.0 (Windows NT 6.1; Win64; x64; rv:49.0) Gecko/20100101 Firefox/49.0
Build ID: 20160502030207
Steps to reproduce:
Address Book Contact Drag & Drop not Working from One Address Book to Another Book
Actual results:
Address Book Contact Drag & Drop not Working from One Address Book to Another Book
Expected results:
Address Book Contact Drag & Drop not Working from One Address Book to Another Book
|
|
1,270,096 | 11,381,140 |
User Agent: Mozilla/5.0 (X11; Linux x86_64; rv:46.0) Gecko/20100101 Firefox/46.0
Build ID: 20160502172042
Steps to reproduce:
- Execute some crossdomain and GET XHR (it will execute an OPTION request followed by a GET request)
- Go to the debugger and try to replay the same XHR
Actual results:
The OPTION XHR actually returns 200, so that's pretty good, i'm allowed to do some great stuffs.. but, the GET request NEVER launch...
Expected results:
The options returns a 200, then the GET will be launched and executed as the first GET xhr was....
|
|
1,260,850 | 11,288,883 |
1. Click "Browse"
2. Look to select a file,
3. Switch to Finder
4. Switch back to FF
Result: File dialog is stuck open. Cancel doesn't work, Open doesn't work. I have to quit the browser.
Step 3 might or might not be necessary.
Screen recording:
https://www.youtube.com/edit?video_id=J6tP_b9GcdI&video_referrer=watch
|
|
1,278,262 | 11,466,107 |
Created attachment 8760287
Client 06-06-16 16-52.crash
Build: 5407821
Device: IPone 6 Plus
Steps to reproduce:
1. Sign in to FxA on mobile device and bookmark a few pages
2. Sign in to FxA on desktop, add several bookmarks in the mobile folder and run sync
3. Go to Firfox for IOS and delete some bookmarks while sync is running
Actual results:
- The application will crash.
Note:
- The crash will occur only when there are new bookmarks in the 'Mobile Bookmarks' folder imported from desktop to the device.
|
|
1,278,530 | 11,468,930 |
[Affected versions]:
Firefox 50.0a1 (2016-06-06)
Firefox 49.0a2 (2016-06-07)
Firefox 48.0a2 (2016-06-06)
[Affected platforms]:
Windows 10 64-bit
Ubuntu 14.04 32-bit
Mac OS X 10.11
[Steps to reproduce]:
1.Launch Firefox with clean profile.
2.Create the xpinstall.signatures.dev-root pref in about:config and set it to true.
3.Install the following webextension: https://addons-dev.allizom.org/en-US/firefox/addon/kb-sticky-notes/
4.Restart the browser.
5.Check the about:addons.
[Expected Results]:
Webextensions without id are not erased after a browser restart.
[Actual Results]:
The WebExtensions are removed after a browser restart if the Gecko ID is not specified in manifest.json.
[Additional notes]:
- This issue reproduces for both xpi and zip files.
- It is not reproducible for webextensions with gecko ID in manifest.json (https://addons-dev.allizom.org/en-US/firefox/addon/kb-sticky-notes2/?src=ss)
|
|
1,304,394 | 11,725,802 |
Created attachment 8793330
IMG_0050.PNG
Build: 5.3(1) RC
Device: iPad Air 2
iOS: 10.0.1
Steps to reproduce:
1. Launch Firefox in portrait view
2. Open the menu
3. Background Firefox by pressing the home button
4. Restore Firefox
Actual results;
- The menu layout will be broken and all the icons will overlap
Expected:
- The menu should be correctly displayed
Note:
- This issue was encountered on iPad Air 2 and iPad mini 4 running iOS 10.
- This issue was not reproduced on 5.2(1) released version.
|
|
1,296,505 | 11,638,304 |
User Agent: Mozilla/5.0 (Windows NT 10.0; WOW64; rv:48.0) Gecko/20100101 Firefox/48.0
Build ID: 20160817112116
Steps to reproduce:
Installed firefox 32 bit
(20160817112116
User Agent Mozilla/5.0 (Windows NT 10.0; WOW64; rv:48.0) Gecko/20100101 Firefox/48.0)
Installed Media Feature Pack for N and KN versions of Windows 10 (both 64 and 32 bit)
browsed this page: http://www.quirksmode.org/html5/tests/video.html
Actual results:
no h.264 in Firefox
Expected results:
passed the http://www.quirksmode.org/html5/tests/video.html video test for h.264
|
|
1,299,499 | 11,670,098 |
[Affected versions]:
- Firefox 51.0a1 (2016-08-31)
[Affected platforms]:
- Windows 10x64, Mac OS X 10.10.5, Ubuntu 16.04x64
[Steps to reproduce]:
1. Launch Firefox.
2. From about:config, enable the devtools.responsive.html.enabled pref.
3. Enable RDM.
4. Open non-remote(Eg. about:config) page the opened tab.
5. Select another tab.
6. Click on the previous tab.
[Expected result]:
- Firefox works correctly.
[Actual result]:
- Firefox becomes unresponsive and in the Browser Console, the Error: Assertion failuretabbrowser.xml:3545:23 is thrown.
[Regression range]:
- I will search for regression ASAP.
[Additional notes]:
- Also, after performing step 4 from STR, the RDM tool is displayed as being enabled in the Developer menu.
|
|
1,292,071 | 11,598,243 |
2016-08-03 01:00 +0000
Bug 1180944 landed as https://hg.mozilla.org/integration/fx-team/rev/474f38fc48be9f97b49b084ebf0b59293b81cf16
Eslint passes: https://treeherder.mozilla.org/#/jobs?repo=fx-team&revision=474f38fc48be&filter-searchStr=eslint
2016-08-03 01:53 +0000
Bug 1285863 landed as https://hg.mozilla.org/integration/fx-team/rev/4bd1976fbb4a
Eslint passes: https://treeherder.mozilla.org/#/jobs?repo=fx-team&revision=555caa0082c3&filter-searchStr=eslint
With the next merge 13 hours later, eslint returns errors for these changes: https://treeherder.mozilla.org/#/jobs?repo=fx-team&revision=4f072e4cc25ac1b0d68887e58f3ee2d5912a8173&filter-searchStr=eslint
So does ESlint lint outdated files? Tomcat mentioned he observed this behavior already one day earlier.
|
|
1,292,288 | 11,599,917 |
Steps to reproduce (I've only done this once, not sure if it reproduces 100%, and this is on 10.12):
1. Watch the video on https://ahal.ca/blog/2016/taskcluster-interactive-loaner/ in fullscreen mode.
2. Click the button to exit fullscreen mode.
3. The screen goes black and Firefox hangs.
Main thread:
> __psynch_mutexwait (in libsystem_kernel.dylib) + 10 [0x7fffb6abacba]
> _pthread_mutex_lock_wait (in libsystem_pthread.dylib) + 100 [0x7fffb6ba2f8a]
> _pthread_mutex_lock_slow (in libsystem_pthread.dylib) + 285 [0x7fffb6ba06cd]
> CGLLockContext (in OpenGL) + 41 [0x7fffa67986bf]
> -[ChildView setFullscreen:] (in XUL) + 59 [0x10fc8af2b]
> nsChildView::PrepareWindowEffects() (in XUL) + 369 [0x10fc83771]
> mozilla::layers::ClientLayerManager::EndTransaction(void (*)(mozilla::layers::PaintedLayer*, gfxContext*, mozilla::gfx::IntRegionTyped<mozilla::gfx::UnknownUnits> const&, mozilla::gfx::IntRegionTyped<mozilla::gfx::UnknownUnits> const&, mozilla::layers::DrawRegionClip, mozilla::gfx::IntRegionTyped<mozilla::gfx::UnknownUnits> const&, void*), void*, mozilla::layers::LayerManager::EndTransactionFlags) (in XUL) + 44 [0x10e6f9b6c]
> nsDisplayList::PaintRoot(nsDisplayListBuilder*, nsRenderingContext*, unsigned int) (in XUL) + 3958 [0x10ff5a916]
> nsLayoutUtils::PaintFrame(nsRenderingContext*, nsIFrame*, nsRegion const&, unsigned int, nsDisplayListBuilderMode, nsLayoutUtils::PaintFrameFlags) (in XUL) + 5550 [0x10ff8dcfe]
> PresShell::Paint(nsView*, nsRegion const&, unsigned int) (in XUL) + 1626 [0x10ffb517a]
> nsViewManager::ProcessPendingUpdatesPaint(nsIWidget*) (in XUL) + 496 [0x10fc41f30]
> nsViewManager::ProcessPendingUpdatesForView(nsView*, bool) (in XUL) + 368 [0x10fc41b30]
> nsViewManager::ProcessPendingUpdates() (in XUL) + 195 [0x10fc42983]
> nsViewManager::WillPaintWindow(nsIWidget*) (in XUL) + 98 [0x10fc42862]
> nsView::WillPaintWindow(nsIWidget*) (in XUL) + 26 [0x10fc4073a]
> -[ChildView viewWillDraw] (in XUL) + 521 [0x10fc8ba89]
> [...]
> -[NSWindow makeKeyAndOrderFront:] (in AppKit) + 111 [0x7fff9f64bbf7]
> nsCocoaWindow::Show(bool) (in XUL) + 1431 [0x10fcc7b17]
> nsCocoaWindow::HideWindowChrome(bool) (in XUL) + 614 [0x10fcc8bf6]
> nsBaseWidget::MakeFullScreen(bool, nsIScreen*) (in XUL) + 36 [0x10fc47174]
> nsCocoaWindow::DoMakeFullScreen(bool, bool) (in XUL) + 176 [0x10fcc9330]
> FullscreenTransitionTask::Run() (in XUL) + 303 [0x10e8e37df]
Compositor thread:
> __psynch_mutexwait (in libsystem_kernel.dylib) + 10 [0x7fffb6abacba]
> _pthread_mutex_lock_wait (in libsystem_pthread.dylib) + 100 [0x7fffb6ba2f8a]
> PR_Lock (in libnss3.dylib) + 14 [0x105fc42de]
> nsChildView::MaybeDrawTitlebar(mozilla::layers::GLManager*) (in XUL) + 31 [0x10fc84cbf]
> nsChildView::DrawWindowOverlay(mozilla::layers::GLManager*, mozilla::gfx::IntRectTyped<mozilla::LayoutDevicePixel>) (in XUL) + 64 [0x10fc84c30]
> nsChildView::DrawWindowOverlay(mozilla::layers::LayerManagerComposite*, mozilla::gfx::IntRectTyped<mozilla::LayoutDevicePixel>) (in XUL) + 53 [0x10fc84bb5]
> mozilla::layers::LayerManagerComposite::Render(mozilla::gfx::IntRegionTyped<mozilla::gfx::UnknownUnits> const&, mozilla::gfx::IntRegionTyped<mozilla::gfx::UnknownUnits> const&) (in XUL) + 3322 [0x10e738a0a]
> mozilla::layers::LayerManagerComposite::UpdateAndRender() (in XUL) + 717 [0x10e737a5d]
> mozilla::layers::LayerManagerComposite::EndTransaction(mozilla::TimeStamp const&, mozilla::layers::LayerManager::EndTransactionFlags) (in XUL) + 183 [0x10e7376c7]
> mozilla::layers::CompositorBridgeParent::CompositeToTarget(mozilla::gfx::DrawTarget*, mozilla::gfx::IntRectTyped<mozilla::gfx::UnknownUnits> const*) (in XUL) + 669 [0x10e74a15d]
> mozilla::layers::CompositorVsyncScheduler::Composite(mozilla::TimeStamp) (in XUL) + 139 [0x10e749a8b]
|
|
1,292,534 | 11,602,039 |
Created attachment 8778238
mozilla-flex_buffer_overrun.patch
http://seclists.org/oss-sec/2016/q3/163
CVE-2016-6354
Some versions of The Fast Lexical Analyzer contain a bug which causes it to produce code potentially vulnerable to a buffer overrun.
While the fix should be made upstream (CMU Sphinx and ANGLE) in the first place, it might be a good idea to apply this hotfix for now.
|
|
1,306,825 | 11,752,004 |
User Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.12; rv:43.0) Gecko/20100101 SeaMonkey/2.40
Build ID: 20160120191716
Steps to reproduce:
Holding Mouse button down to move a page away to see the desktop.
Actual results:
When clicking om the top Page bar with the Apple USB mouse, I can no longer move the pages around on the desk top, the will not move any direction.
The mouse works perfectly on Firefox, Chrome and Safari, but not Seamonkey.
Expected results:
The web page should be able to be moved around on the desk top, it will not move!
|
|
1,147,271 | 10,074,211 |
+++ This bug was initially created as a clone of Bug #1145507 +++
|
|
1,150,371 | 10,109,902 |
Steps to reproduce:
1) Use Nightly on Windows Vista or later (without e10s for the time being).
2) Navigate to http://media.junglecode.net/test/dash-eme/sintel.html
3) Look at the space between the back button and the location in the location bar.
4) Click the icon shown there.
5) Right-click the video.
Actual results:
There is an icon of links of a chain on a gray circle between the back button and the location in the location bar. Clicking this icon drops doorhanger that claims that some video or audio on this site use DRM software. The last item of the context menu is "Learn more about DRMβ¦"
Expected results:
Since Clear Key is not DRM (obtaining the keys is not conditional on demonstrating adherence to a DRM robustness and compliance regime), the DRM UI should not be shown when the Clear Key CDM is the only CDM in use.
Additional info:
Filing as a Core bug, because I'm assuming the Gecko is responsible for informing the Firefox UI about the presence of DRM.
Marking as "major", because this UI showing up for Clear Key substantially misrepresents the situation on a controversial topic.
|
|
1,172,301 | 10,354,347 |
User Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.10; rv:41.0) Gecko/20100101 Firefox/41.0 SeaMonkey/2.38a1
Build ID: 20150606003002
Steps to reproduce:
In latest trunk, SeaMonkey loads pages only if selected from bookmarks menu. If I try to load pages collected in location bar, this doesn't works. If I try to write an URL in location bar and then hold down Enter, page doesn't load.
|
|
1,176,672 | 10,413,842 |
User Agent: Mozilla/5.0 (Windows NT 5.1; rv:38.0) Gecko/20100101 Firefox/38.0
Build ID: 20150525141253
Steps to reproduce:
Open this link in Firefox 38.x on Windows XP SP3 AND Intel Haswell CPU:
http://base-n.de/webm/VP9%20Sample.html
It does not matter if Firefox runs in "safe mode" or not.
Actual results:
Instant EXCEPTION_ILLEGAL_INSTRUCTION crash such as this:
https://crash-stats.mozilla.com/report/index/5e1ec086-ec17-47dd-bda4-9df582150622
because libvpx attempted to execute an AVX2 instruction. AVX and AVX2 instructions are not supported on Windows XP SP3.
Expected results:
The video should have started to play.
|
|
1,180,080 | 10,448,969 |
+++ This bug was initially created as a clone of Bug #924069 +++
Steps to reproduce:
Type the first bit of an URL in the urlbar for which a completion is available. Press RET.
E.g. say bugzilla.mozilla.org is completable, type "bug"
Actual results:
The completed bit of the URL ("zilla.mozilla.org") is now in the X clipboard.
Expected results:
The X clipboard's contents should not have been tampered with.
|
|
426,271 | 3,555,475 |
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9pre) Gecko/2008033105 Minefield/3.0pre
Build Identifier:
Auto-detect of Japanese Character Encoding does not work on Trunk.
Reproducible: Always
Steps to Reproduce:
1.Set preference as
user_pref("intl.accept_languages", "ja,en-us,en");
user_pref("intl.charset.default", "Shift_JIS");
user_pref("intl.charset.detector", "ja_parallel_state_machine");
user_pref("intl.charsetmenu.browser.static", "Shift_JIS, EUC-JP, ISO-2022-JP,
ISO-8859-1, UTF-8");
2.Acccess
http://space.geocities.jp/alice0775/STORE/EUC-JP.html (for EUC-JP)
http://space.geocities.jp/alice0775/STORE/UTF-8.html (UTF-8)
3.
Actual Results:
Auto-detect is failed
Expected Results:
Auto-detect should success.
I think this issue is jp-critical and blocking1.9.
|
|
429,118 | 3,580,556 |
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US; rv:1.9pre) Gecko/2008041406 Minefield/3.0pre
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US; rv:1.9pre) Gecko/2008041403 Thunderbird/3.0a1pre ID:2008041403
When composing messages in Thunderbird the inline spell checker only checks the first word of the message but ignores the rest.
So for example, if my message is:
wnorg worng gnorw
Spell checker should highlight all 3 words but instead only the first word is highlighted.
Reproducible: Always
Steps to Reproduce:
1.Open Thunderbird
2.Create a new message
3.Enter message content
4.Let the inline spell checker does its job
Actual Results:
Only the first word is highlighted (if it is spelt wrong)
Expected Results:
All spelling errors should be highlighted.
|
|
430,363 | 3,591,818 |
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US; rv:1.9b5) Gecko/2008032620 Firefox/3.0b5
Build Identifier: version 0.1.29
After Weave is left turned on for a few hours on at least 2 machines (which are perfectly synchronized in the beginning), the deltas file on the server becomes enormous, i.e. much larger than a few hours of browsing would warrant.
Reproducible: Always
Steps to Reproduce:
Steps to reproduce the problem (using Weave v0.1.29, but earlier versions behaved the same way):
1. Install Weave on two machines (A and B), each running Firefox 3 beta 5. One machine is Linux, the other is Vista, if that matters (e.g. if the reconciler gets confused by the different newline conventions).
2. Sync machine A, then machine B, then again machine A. This ensures that both machines now share the same history and bookmarks.
3. Delete all the data on the server.
4. Again, sync machine A, then machine B, then again machine A. Both machines are synced again, and the deltas size should be near-zero.
5. Note the sizes of the files in /user-data/history. In my case, snapshot.json is 332 KB, and deltas.json is very small.
6. Browse casually for several hours on one or both machines.
Actual Results:
Deltas.json in /user-data/history/ has grown enormously and is now 1400 KB! This is not normal behaviour, the deltas accumulated for several hours should not be 4 times larger than the snapshot of several months activity. This also leads to poor performance because syncing starts to break often (failed download/upload), and the reconciler becomes too CPU-hungry.
Expected Results:
Deltas.json should have grown by several KB at most.
Link to forum post regarding this problem: https://labs.mozilla.com/forum/index.php/topic,689.0.html
|
|
438,241 | 3,653,858 |
User-Agent: Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.5; en-US; rv:1.9) Gecko/2008053008 Firefox/3.0
Build Identifier: Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.5; en-US; rv:1.9) Gecko/2008053008 Firefox/3.0
Try this example that demonstrates the problem:
http://eyestreet.com/~rmcgeary/test-hover-ff3.html
Reproducible: Always
Steps to Reproduce:
1. Try hovering over the input elements in the example.
2. Continue to move the mouse over the input elements until more than one element is left red.
Actual Results:
In Firefox 3 RC2, some elements are left red even though the mouse has left those elements.
Expected Results:
Upon hover (mouseover) of any element, the background color should be made red. Mouseout should return the background color to white.
|
|
438,252 | 3,653,987 |
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9) Gecko/2008052906 Firefox/3.0
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9) Gecko/2008052906 Firefox/3.0
When organizing bookmarks if you edit the address of a bookmark, then edit the tag information the address change will be updated, but not the tag info.
Editing the bookmark title and tag at the same time works fine.
Reproducible: Always
Steps to Reproduce:
1. Create a bookmark
2. Go into organize bookmarks and select the bookmark
3. edit the URL of the bookmark
4. without changing your selection on the bookmark list, edit the tags of the bookmark
5. Changing your selection or closing the bookmark window will save the updated address info, but not the updated tag info.
Actual Results:
The address was saved, but not the tag.
Expected Results:
Saved the tag information as well.
|
|
440,366 | 3,668,465 |
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; de; rv:1.8.1.14) Gecko/20080404 Firefox/2.0.0.14
Build Identifier: DE Thunderbird Version 2.0.0.14 (20080421)
all keywords of a mail are no longer present after I cut off (not delete) attachments
Reproducible: Always
Steps to Reproduce:
1. add keyword to mail
2. cut off attachment of this mail
Actual Results:
added keyword is no longer present
Expected Results:
added keyword is still present
|
|
436,722 | 3,643,408 |
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9) Gecko/2008051206 Firefox/3.0
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9) Gecko/2008051206 Firefox/3.0
On IE7 and Firefox 2, I see a lot of marketing information. I see barely anything with Firefox 3 RC1.
Reproducible: Always
Steps to Reproduce:
1. http://www.bankofamerica.com/onlinebanking/
2.
3.
Actual Results:
See a lot of white space
Expected Results:
See some information
Uninstalled RC1, installed 2.0, site worked. Installed 3.0 again, it doesn't work. No addons.
|
|
436,930 | 3,645,370 |
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9) Gecko/2008051206 Firefox/3.0
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9) Gecko/2008051206 Firefox/3.0
This is unique to Firefox 3 -- also tested in the June 2nd nightly -- same result. Also tested in the the latest versions of IE7, Safari, Opera -- this bug does not occur in those browsers.
When hovering over either of the Share/Save or Subscribe buttons on http://www.addtoany.com/ , a menu appears. The menu is designed to disappear onmouseout (after a short delay).
The problem is: After hovering over the button to show the drop-down menu, if you mouse-over the menu too rapidly, the menu disappears. However, if you slowly move the mouse from the button and onto the menu... the menu will stay, like it is designed to do.
I tried to pinpoint the bug, but haven't had much success yet. My guess is that this occurs due to the floating elements inside of the menu, but I haven't been able to verify that yet.
Reproducible: Always
Steps to Reproduce:
1. Go to http://www.addtoany.com/ and hover over either of their buttons (Share/Save or Subscribe).
Actual Results:
A menu appears.
If you move the mouse too quickly over the menu, the menu inappropriately disappears.
If you move the mouse SLOWLY over the menu, the menu successfully stays in place.
Expected Results:
The menu should stay in place like it does in all other browsers (FF2, IE6+, Safari, Opera)
I've had problems with other drop-down menus around the web that are similar in functionality. Unfortunately, I do not recall the instances.
|
|
430,552 | 3,593,335 |
Created attachment 317384
Screenshot after tab was closed
Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.5; en-US; rv:1.9pre) Gecko/2008042304 Minefield/3.0pre ID:2008042304
There is a major issue with the tabbed browser which happens in conjunction with a probably redirection while browsing on the given site.
Steps to reproduce:
1. Create a fresh profile
2. Open Firefox and create a new tab by hitting Cmd+T
3. Load following URL into this tab: http://www.mobunity.de
4. See that you are redirected to http://www.mobunity.de/mobunity/index.html
5. Click on Messaging within the main menu
6. Hit Cmd+W to close this tab
As result you see something like the following attachment:
* No tab is connected to the Bookmarks Toolbar anymore
* All keyboard shortcuts (e.g. Cmd+N, Cmd+T, Cmd+W...) don't work anymore until you focus a tab or the window again
* The location bar still has the last URL inside
I'm currently not able to test this on Windows or Linux. So if anyone could have a look it would be great.
|
|
430,948 | 3,597,128 |
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-us; rv:1.9pre) Gecko/2008042603 Minefield/3.0pre Firefox/2.0.0.14
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-us; rv:1.9pre) Gecko/2008042603 Minefield/3.0pre
Places:Drag & Drop a Favicon or a Link to Bookmarks Toolbar, Bookmarks Menu or Sidebar Bookmarks does not work.
Reproducible: Sometimes
Steps to Reproduce:
1.Start Firefox with New Profile.
2.Drag & Drop a Favicon or a Link to Bookmarks Toolbar, Bookmarks Menu or Sidebar Bookmarks.
3.
Actual Results:
Nothing happen
Expected Results:
Should add a bookmark item to the dropped zone.
In Console2 window.
Error: uncaught exception: [Exception... "Component returned failure code: 0x80070057 (NS_ERROR_ILLEGAL_VALUE) [nsINavBookmarksService.getFolderIdForItem]" nsresult: "0x80070057 (NS_ERROR_ILLEGAL_VALUE)" location: "JS frame :: chrome://browser/content/places/controller.js :: PCDH_canMoveContainer :: line 1387" data: no]
This may be Regression.
|
|
431,615 | 3,603,779 |
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-GB; rv:1.8.1.14) Gecko/20080404 Firefox/2.0.0.14
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-GB; rv:1.8.1.14) Gecko/20080404 Firefox/2.0.0.14
I've been using the "HTML 4.01 MAP element to group links, then identifying the group with the "title" attribute". This is a suggested AAA W3C WCAG 1.0 accessibility technique under checkpoint 13.6
www.w3.org/TR/WCAG10-HTML-TECHS/#group-bypass
WCAG 2.0, released today as a Candidate Recommendation, also suggests the MAP element as one technique that can be used to group links.
www.w3.org/TR/2004/WD-WCAG20-HTML-TECHS-20041119/#linkgroups
However, the map element seems to be affected when a screen reader virtualises the page in Firefox.
Using Firefox 2 with Jaws 9 on Windows XP, all the navigation links were missing from the virtual buffer. Turning the virtual cursor off, the user could tab between all the links (including the missing ones). A similar problem was encountered with Window Eyes.
After deleting the map element from the markup all the links were visible (audible/feelable) with Firefox/Jaws.
A test page demonstrating the bug is available at:
www.e-bility.com/braillespace/bwp/test.php
and the same page (now without map element) is at:
www.braillespace.com
The problem has also been reported to Freedom Scientific (via the JAWS Beta List) and GW Micro by email.
Thanks for your help and the great work you are doing with Firefox.
Reproducible: Always
Steps to Reproduce:
1. Go to test page above using JAWS or Window Eyes
2.
3.
Actual Results:
Using the map element to group related links results in none of the links being available to a screen reader. They are not even aware that there are links on the page. However the links are available if the web page is viewed in Internet Explorer. Removing the map element from the mark-up fixes the problem.
Expected Results:
The links should be visible and the map element title should indicate that they are related, e.g. that all links belong to the "Main navigation" or "Section navigation" etc.
Because it is a recommended W3C technique used to enhance accessibility it is likely to be most prevalent on websites for this audience. The technique is also most likely to be used for important link groups, such as navigation areas. However, any site using the map element will mean it cannot be navigated by a screen reader in Firefox.
|
|
432,225 | 3,609,233 |
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9pre) Gecko/2008050504 Minefield/3.0pre
Build Identifier: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9pre) Gecko/2008050504 Minefield/3.0pre
I'm finding that text fields in Minefield are not always spell checked, despite having the Check Spelling box checked. Some times it works but much of the time it doesn't. I'm especially noticing this behavior on my WordPress.com blog but the behavior is certainly not restricted to that site.
Reproducible: Always
Steps to Reproduce:
1. Enter some text in a text box that includes misspelled words
2. Right click inside the text box and verify that the Check Spelling box is checked
3. Look for indications that words are identified as misspelled
Actual Results:
No words are shown as misspelled and right clicking on a known misspelled word does not offer any spelling alternatives.
Expected Results:
As words are typed with misspellings the built-ion spell checker should flag them as being misspelled and a right click on a misspelled word should reveal other possible spellings as well as an option to add the spelling to the dictionary
I am running Minefield 3.0pre nightly from Mozilla, not from Ubuntu's repositories.
|
|
441,359 | 3,675,363 |
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/525.13 (KHTML, like Gecko) Version/3.1 Safari/525.13
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; de; rv:1.9) Gecko/2008052906 Firefox/3.0
Currently, it appears that only the URL of the top-level page, and perhaps iframes(?) are checked against the SafeBrowsing database. To get the full utility of the service, all resources (e.g. especially scripts on a page, flash files, etc) should be checked against the SafeBrowsing lists. This will generate more lookup requests, so it might be necessary to implement something in-memory (e.g. a bloom filter to catch the first level requests, going to the database only if necessary), but we see many many new compromises where someone injects a bunch of script tags that point to things already on the malware list. If all the resources on the page were checked, a user would get much better protection than they're currently getting.
Reproducible: Always
Steps to Reproduce:
Go to http://www.ianfette.com/jstest.html
(This pulls in a harmless file from ianfette.org that just throws up a javascript alert. However, ianfette.org is on the blacklist, were this a real attack the user could be in trouble.)
Actual Results:
Get an alert box
Expected Results:
Should get an interstitial telling you that the page contains resources from ianfette.org which is a known bad site, and that the page is not safe to view.
|
|
441,473 | 3,676,223 |
Implement user font set object, serves as container for use with @font-face rules.
- create by style system code, attached to PresShell(?)
- passed down to text run, font group
- resolves <font family> ==> <font entry object>
- initial lookup sparks the download of font data
- handles download of font data
- download completion causes a reflow(how? hmm...)
- also handles lookup using full font names (e.g. local(xxx))
|
|
441,592 | 3,677,091 |
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.8.1.13) Gecko/20080316 SUSE/2.0.0.13-0.2 Firefox/2.0.0.13
Build Identifier: R3.0.4
As discussed on mozilla.support.bugzilla, the Bugzilla 3.0 release/stable stream is missing this line in the Bugzilla::ModPerl::CleanupHandler
Bugzilla::_cleanup();
In CVS, mod_perl.pl with the rtag Bugzilla_Stable, BUGZILLA-3_0_4, BUGZILLA-3_0_3 and BUGZILLA-3_0-STABLE is 1.5.2.2, and that version does not call Bugzilla::_cleanup();
Anyone installing BZ from CVS, using any of those rtags and using mod_perl will likely have a problem with too many DB connections to the master.
The call to Bugzilla::_cleanup() appears in mod_perl.pl 1.6, which is rtagged BUGZILLA-3_1_1 (and BUGZILLA-3_1_2).
Reproducible: Always
Steps to Reproduce:
1.
2.
3.
|
|
441,751 | 3,678,256 |
User-Agent: Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; .NET CLR 1.1.4322; .NET CLR 2.0.50727; .NET CLR 3.0.04506.648; .NET CLR 3.5.21022)
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-GB; rv:1.9) Gecko/2008052906 Firefox/3.0
From your description, you're encountering some problem with disable
client-side browser cache in firefox browser.
Regarding on this, I've also performed some test on my side. Generally, the
following code should be necessary to disable client-side cache:
=======================
Protected Sub Page_Load(ByVal sender As Object, ByVal e As
System.EventArgs) Handles Me.Load
Response.Cache.SetAllowResponseInBrowserHistory(False)
Response.Cache.SetCacheability(HttpCacheability.NoCache)
Response.Cache.SetNoStore()
Response.Expires = 0
Response.Write("<br/>" & DateTime.Now.ToLongTimeString())
End Sub
==============================
And my local tests showed that it works corretly for IE7 and firefox 2.0.
However, for firefox 3.0, it does not work. I think it is due to the
implemenation of the firefox 3. BTW, the cache control http header is a
advisory info and the actual implemation is not mandatory and depend on the
browser, so I suggest you also try submit this issue to firefox community
to see whether this is an by design behavior.
Reproducible: Always
Steps to Reproduce:
1. Load page with "no-cache" or "no-store" directive
2. Go forward to next page
3.Go back button on browser, page loads from cache.
Actual Results:
Page loads from cache, but should re-load from server or show message that page has expired.
|
|
346,984 | 2,931,309 |
xbl and event handlers cause trouble in mailnews and editor
i am not sure how many bugs are in this report, so help split them if needed.
when an image is included via xbl in an email message, its javascript event
handlers (onload,onmouseover) fire in both preview pane and editor (when
forward inline or reply).
so this allows:
1. javascript exploits in email
2. when forwarding or replying, it is possible to steal user's files via
dhtml - img.src='file:///etc/passwd';document.body.appendChild(img);
3. privacy implications because of the image fetching.
affected are thunderbird 1.5.0.5, 2.0a1 and seamonkey trunk at least.
local mailbox with info and xbl file to be attached.
|
|
348,341 | 2,941,787 |
In Firefox and Thunderbird, there are no problems.
As for Camino alone, this function doesn't work.
Firefox and Thunderbird will be influenced in the future if it is a problem of Cocoa widget.
Reproducible: Always
Steps to Reproduce:
1. Open URL
2. Click search(input) field
3. IME is turned on.
4. Input japanese
e.g. kotoeri (It is displayed by the hiragana as "γγ¨γγ".)
5. The input is fixed pressing return.
6. Input "γγ¨γγ" is selected.
7. "Reverse conversion" is selected from the menu of IME.
( or push Contol+Shift+r)
It is explanation Movie of this function.
http://www.apple.com/jp/articles/panther/20031107/function04.html
Mac OS X 10.3.9
2006081107 (v1.2+)
|
|
349,852 | 2,954,239 |
The βI'm added to or removed from this capacityβ user pref doesn't work for new bugs, at least not for Assignee (didn't test for others). This affects at least 2.22 and tip. I suspect earlier releases to be affected, too.
Steps to reproduce:
o Mangle a test user account userA:
- Switch off all mail notifications
- Switch on βI'm added to or removed from this capacityβ for Assignee
o Log on as somebody else (userB)
o File a bug, assigning it to userA
Actual results:
userA is not notified.
Expected results:
userA is notified.
|
|
360,511 | 3,036,288 |
It's unclear where the bug is here, but I can't find an existing one on the problem, and it appears to be at a lower level than the URLBar, so I'm going to go for docshell for now.
Steps:
1. Load a page that has hash links, like the one in the URL, or a Bugzilla bug.
2. Click a hash link (e.g. "link here" on my page).
3. Click a normal link ("CVS log", Bugzilla attachment, etc.).
4. Click back once.
Actual results:
URLBar (and document.location) show the URL *without* the hash part.
Expeted results:
URLBar (and document.location) show the URL *with* the hash part.
Note that clicking back a 2nd time will actually reload the page (seems wrong), and show the URL without the hash (correct).
This occurs for me with Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US; rv:1.9a1) Gecko/20061110 Firefox/3.0a1 ID:2006111004, and Hannibal on IRC confirmed it does not happen in his Firefox 2.
|
|
360,572 | 3,036,703 |
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; de; rv:1.8.1) Gecko/20061010 Firefox/2.0
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; de; rv:1.8.1) Gecko/20061010 Firefox/2.0
The deletion of previous search entries with shift+del doesn't work if search suggest shows some options. Using shift+del deletes the current entry, but the next time the entry is back again.
Reproducible: Always
Steps to Reproduce:
1. Clear your search history.
2. Search for "test" using the search bar.
3. Enter "test" again in the search bar, select the previous entry "test" from the drop-down and hit shift+del in order to delete it from the search history.
4. The entry "test" is apparently deleted.
5. Enter "test" again in the search bar, notice that "test" is still shown in the drop-down box.
Actual Results:
The search entry "test" ist not deleted.
Expected Results:
The search entry should be deleted after the first time I hit shift+delete.
I don't know whether this bug should be in the component "Form Manager" instead.
Using a search engine with no auto-suggest feature, deleting old search entries works without problems.
Workaround: Klick in the search bar and don't enter any text, use the cursor (down) to see the previous searches and delete them with shift+delete.
|
|
354,541 | 2,995,956 |
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.0.7) Gecko/20060909 Firefox/1.5.0.7
Build Identifier: FireFox 2.0 rc1
www.xcellery.com uses AJAX.NET from Michael Schwartz which uses the trim function from javascript. Sign in as demo user on www.xcellery.com brings "responseText.trim is not a function" when loading the My Documents page. This doesn't happen with any of the previous versions of FireFox.
Reproducible: Always
Steps to Reproduce:
1. go to www.xcellery.com
2. click on "Sign in as Demo user"
Actual Results:
you see the error "responseText.trim is not a function" in a message box.
Expected Results:
it should load the workbooks in the "My Documents" page.
This only happens with FireFox 2 rc1.
|
|
355,304 | 3,001,715 |
When exporting a storage calendar to iCalendar format recurring events loose any exceptions.
Steps to reproduce:
1. create daily recurring event (5 times)
2. drag first occurrence one week ahead
3. export whole calendar, choose iCalendar format
Result: no exceptions in ics file. After importing the data again, the exeptions are not shown.
found using Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9a1) Gecko/20061002 Sunbird/0.3
|
|
356,474 | 3,009,627 |
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.0.7) Gecko/20060909 Firefox/1.5.0.7
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.0.7) Gecko/20060909 Firefox/1.5.0.7
There are sites that try to scam our company (http://www.sandhills-publishings.com/). What they are doing is passing our site (http://www.sandhills.com/) in source of the frames.
We are running asp.net and are checking URLs of scam sites. When we find one, we add it to our web.config file, then check the http reference against it. That way, if they are running our web site through a frame, our site will catch it and redirect theirs to our scam site.
Reproducible: Always
Steps to Reproduce:
1.Go to http://www.sandhills-publishings.com/ with v1.5.0.7
2.Go to http://www.sandhills-publishings.com/ with v1.0.7
3.Go to http://www.sandhills-publishings.com/ with IE (sorry guys, have to put this here to show the results)
Actual Results:
1. Stays at http://www.sandhills-publishings.com/
2. Redirects to http://www.sandhills.com/scam.aspx
3. Redirects to http://www.sandhills.com/scam.aspx
Expected Results:
v1.5.0.7 should redirect to http://www.sandhills.com/scam.aspx
You can contact me at this email for further information.
|
|
358,248 | 3,021,592 |
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.8.1b2) Gecko/20060910 SeaMonkey/1.1b
Build Identifier: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.8.1b2) Gecko/20060910 SeaMonkey/1.1b
Tested on Fedora Core 4 running NSPR 4.6.3, NSS 3.11.3 and mod_nss 1.0.5. It crashes in a similar way on Solaris 9 and 10.
A standard multi-process NSS server program starts up something like this:
PR_Init();
SSL_ConfigMPServerSIDCache();
NSS_Initialize();
(and setup policy, ciphers, etc).
And then at some point it might shut down with:
SSL_ShutdownServerSessionIDCache();
NSS_Shutdown();
The call to SSL_ConfigMPServerSIDCache() launches a new thread, LockPoller, that wakes up every 30 seconds and checks the value of cache->sharedCache->stopPolling to see if it should exit. If it is ever PR_TRUE then the thread will exit gracefully.
This condition will NEVER be true. Nowhere in the NSS code is this set to PR_TRUE.
What I see with mod_nss is memory corruption and eventually a core when I attempt to restart apache with httpd -k restart.
What happens with a restart in Apache is all modules are unloaded and reloaded. In the case of mod_nss this causes the shutdown routines to run (as above) and then another initialization of NSS>
On a few occasions Apache crashed in LockPoller testing the value of cache->stopPolling. As I looked at this more I determined that even with a graceful shutdown this thread will never go away.
To test this I set the value of the timeout in LockPoller to 1 second and then add this to CacheClose() at the very top:
cache->sharedCache->stopPolling = PR_TRUE;
PR_Sleep(PR_SecondsToInterval(2));
So obviously this is not graceful code but it does make mod_nss stop crashing on restarts by giving LockPoller a chance to quit before continuing the cache cleanup.
In summary, the problem is that the LockPoller continues to run after shutting down the cache. So the next time it comes out of the PR_Sleep() it craps out because its memory has been freed.
Reproducible: Always
Steps to Reproduce:
1. Build and install mod_nss
2. Start Apache
3. Restart Apache with httpd -k restart
Actual Results:
core dump. The parent httpd process will die but its children will live.
Expected Results:
No crash.
|
|
343,801 | 2,903,703 |
Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.8.1a3) Gecko/20060706 BonEcho/2.0a3
Steps to reproduce:
1. Open at least two tabs and close one of them.
2. Go to History and move your mouse down till you reach the "Recently Closed Tabs" menu item.
3. Move your mouse back up over the listed History items.
Results:
The list of History items disappears and the menu "jumps" around.
Definitely a regression from bug 342700.
|
|
344,711 | 2,910,925 |
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.0.4) Gecko/20060620 Firefox/1.5.0.4 Flock/0.7.1
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.0.4) Gecko/20060620 Firefox/1.5.0.4 Flock/0.7.1
% cat test.js
var a1 = {abc2 : 1, abc3 : 3};
var j = a1\
.abc2;
print(j);
% xpcshell test.js
Type Manifest File: D:\mozilla\mozilla\ffd\dist\bin\components\xpti.dat
nsNativeComponentLoader: autoregistering begins.
nsNativeComponentLoader: autoregistering succeeded
nsNativeComponentLoader: registering deferred (0)
test.js:2: SyntaxError: illegal character:
test.js:2: .abc2;
test.js:2: ..............................................................
................................................................................
[about 22000 dots printed]
StackTrace:
> xpcshell.exe!my_ErrorReporter(JSContext * cx=0x00c54448, const char * message=0x00cd4f10, JSErrorReport * report=0x00cd69e0) Line 156 + 0x9 C++
js3250.dll!js_ReportErrorAgain(JSContext * cx=0x00c54448, const char * message=0x00cd4e80, JSErrorReport * reportp=0x00cd69e0) Line 1034 + 0x15 C
js3250.dll!js_ReportUncaughtException(JSContext * cx=0x00c54448) Line 1162 + 0x11 C
js3250.dll!JS_CompileFileHandleForPrincipals(JSContext * cx=0x00c54448, JSObject * obj=0x00cb3fd0, const char * filename=0x0039711d, _iobuf * file=0x1027c898, JSPrincipals * principals=0x00ca886c) Line 3789 + 0x25 C
xpcshell.exe!ProcessFile(JSContext * cx=0x00c54448, JSObject * obj=0x00cb3fd0, const char * filename=0x0039711d, _iobuf * file=0x1027c898) Line 584 + 0x1c C++
xpcshell.exe!Process(JSContext * cx=0x00c54448, JSObject * obj=0x00cb3fd0, const char * filename=0x0039711d) Line 681 + 0x15 C++
xpcshell.exe!ProcessArgs(JSContext * cx=0x00c54448, JSObject * obj=0x00cb3fd0, char * * argv=0x0039710c, int argc=1) Line 833 + 0x11 C++
xpcshell.exe!main(int argc=1, char * * argv=0x0039710c, char * * envp=0x00392f88) Line 1100 + 0x15 C++
xpcshell.exe!mainCRTStartup() Line 398 + 0x11 C
kernel32.dll!7c816d4f()
kernel32.dll!7c8399f3()
The crash occurs in the third line in this code:
n = report->tokenptr - report->linebuf;
for (i = j = 0; i < n; i++) {
if (report->linebuf[i] == '\t') {
for (k = (j + 8) & ~7; j < k; j++) {
fputc('.', gErrFile);
}
continue;
}
fputc('.', gErrFile);
j++;
}
The problem is that report->tokenptr points to garbage (0x00ce4ad6).
report->linebug points to 0x00cd4ad8
The code that tries to align the '^' under the bad token ends up
triggering a fault in the IO mechanism.
Reproducible: Always
Steps to Reproduce:
1. Start up js shell (or xpcshell)
2. At prompt 1, type "a1 = {abc2 : 1, abc3 : 3};"
3. At prompt 2, type this (with newlines, without quotes):
js> "j = a1\
.abc2;"
Actual Results:
Many dots and a crash. See above.
|
|
390,716 | 3,268,522 |
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US; rv:1.9a7pre) Gecko/2007080214 Minefield/3.0a7pre
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US; rv:1.9a7pre) Gecko/2007080214 Minefield/3.0a7pre
Regressed within the last 24hrs
Reproducible: Always
Steps to Reproduce:
1. make sure your bookmark toolbar (or in my case, i drug the "Bookmark Toolbar Items" to the Menubar to save space) has a folder in it
2. close firefox; open firefox
3. left-click the folder
Actual Results:
tiny box several pixels wide/heigh displays (a very very tiny menu)
Expected Results:
entire menu (in my case over 10 bookmarks and 2 other folders inside) should drop down.
I can reproduce it every time, but I can't always reproduce how to 'fix' it. Sometimes double-clicking the titlebar to mazimize/unmazimise works, other times not. Usually after opening some other bookmarks, tabs, etc., gets it to expand eventually.
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.