Closed
Bug 1176073
Opened 9 years ago
Closed 9 years ago
Crash: AssertWorkerThread() on main thread under netwerk
Categories
(Core :: Networking, defect)
Tracking
()
RESOLVED
FIXED
mozilla42
People
(Reporter: luke, Assigned: jduell.mcbugs)
References
Details
(Keywords: crash)
Crash Data
Attachments
(1 file)
1.14 KB,
patch
|
mayhemer
:
review+
ritu
:
approval-mozilla-aurora+
Sylvestre
:
approval-mozilla-beta+
Sylvestre
:
approval-mozilla-release+
|
Details | Diff | Splinter Review |
Just crashed with this stack: https://crash-stats.mozilla.com/report/index/98b96d47-4e95-4256-bd17-c73122150618 in current Nightly.
Comment 1•9 years ago
|
||
jason - this is deep in the bowels of ipc. Any idea where to go with it?
Flags: needinfo?(jduell.mcbugs)
Assignee | ||
Comment 2•9 years ago
|
||
Bill, :bent told me you're our new IPDL expert now that he's leaving. Got any thoughts about how to fix this (especially w/o STR?). Yuk.
Flags: needinfo?(jduell.mcbugs) → needinfo?(wmccloskey)
Well, PHttpChannel is part of PContent, which runs on the main thread. So we're clearly running on the correct thread. It seems like the MessageChannel is probably corrupted somehow, probably from being freed. It looks like there isn't the usual mIPCClosed check in ReportSecurityMessage. Could that be the problem?
Flags: needinfo?(wmccloskey)
Assignee | ||
Comment 4•9 years ago
|
||
> It looks like there isn't the usual mIPCClosed check in
> ReportSecurityMessage. Could that be the problem?
Well, it's sure worth a try.
Assignee: nobody → jduell.mcbugs
Status: NEW → ASSIGNED
Attachment #8633725 -
Flags: review?(honzab.moz)
Updated•9 years ago
|
Attachment #8633725 -
Flags: review?(honzab.moz) → review+
Assignee | ||
Updated•9 years ago
|
Keywords: checkin-needed
Assignee | ||
Comment 7•9 years ago
|
||
https://treeherder.mozilla.org/#/jobs?repo=try&revision=419cce3eab69
Comment 9•9 years ago
|
||
https://hg.mozilla.org/mozilla-central/rev/bcc208c22c9e
Status: ASSIGNED → RESOLVED
Closed: 9 years ago
status-firefox42:
--- → fixed
Resolution: --- → FIXED
Target Milestone: --- → mozilla42
Comment 10•9 years ago
|
||
Jason, could you request uplift to Aurora please? I suspect that we won't bother for Beta, but I'm requesting tracking anyway to let the release managers make the call. [Tracking Requested - why for this release]: Moderate topcrash, 0.53% of crashes on 40b8
status-firefox40:
--- → affected
status-firefox41:
--- → affected
tracking-firefox41:
--- → ?
tracking-firefox42:
--- → ?
Flags: needinfo?(jduell.mcbugs)
Assignee | ||
Comment 11•9 years ago
|
||
Comment on attachment 8633725 [details] [diff] [review] v1 Approval Request Comment [Feature/regressing bug #]: unknown [User impact if declined]: crashes [Describe test coverage new/current, TreeHerder]: usual mochitests/xpcshell [Risks and why]: None [String/UUID change made/needed]: none This is a risk-free patch. It adds a check that we're always supposed to do before doing IPC communication, the absence of which may be causing a crash. We should uplift it.
Flags: needinfo?(jduell.mcbugs)
Attachment #8633725 -
Flags: approval-mozilla-beta?
Attachment #8633725 -
Flags: approval-mozilla-aurora?
Tracked as it's a crash.
Comment on attachment 8633725 [details] [diff] [review] v1 This has been in m-c for a few days, let's uplift to Aurora and see if the crash goes away.
Attachment #8633725 -
Flags: approval-mozilla-aurora? → approval-mozilla-aurora+
Comment 14•9 years ago
|
||
Comment on attachment 8633725 [details] [diff] [review] v1 [Triage Comment] Adding to approval-mozilla-release to make sure it lands in the 40 branch as we are going to do the m-b => m-r merge today. Taking it as it this should improve the quality of the release. However, Jason, I don't agree with your evaluation. Every patch introduced a risk... (I have many examples of bug with Risk: None which will break the tree)
Attachment #8633725 -
Flags: approval-mozilla-release+
Attachment #8633725 -
Flags: approval-mozilla-beta?
Attachment #8633725 -
Flags: approval-mozilla-beta+
You need to log in
before you can comment on or make changes to this bug.
Description
•