1833948 - Crash in [@ Watchdog::Init] with PR_CreateThread failed!

Status: NEW

(Core :: XPCOM, defect)

Description

[Andrew McCreight [:mccr8]]

Crash report: https://crash-stats.mozilla.org/report/index/194e72e6-65aa-47f5-a7d9-1c2330230517

MOZ_CRASH Reason: MOZ_CRASH(PR_CreateThread failed!)

Top 10 frames of crashing thread:

0  libxul.so  Watchdog::Init  js/xpconnect/src/XPCJSContext.cpp:175
0  libxul.so  WatchdogManager::StartWatchdog  js/xpconnect/src/XPCJSContext.cpp:413
0  libxul.so  WatchdogManager::RefreshWatchdog  js/xpconnect/src/XPCJSContext.cpp:386
1  libxul.so  WatchdogManager::RegisterContext  js/xpconnect/src/XPCJSContext.cpp:305
1  libxul.so  XPCJSContext::XPCJSContext  js/xpconnect/src/XPCJSContext.cpp:1118
1  libxul.so  XPCJSContext::NewXPCJSContext  js/xpconnect/src/XPCJSContext.cpp:1427
2  libxul.so  nsXPConnect::InitJSContext  js/xpconnect/src/nsXPConnect.cpp:92
2  libxul.so  xpc::InitializeJSContext  js/xpconnect/src/nsXPConnect.cpp:107
3  libxul.so  NS_InitXPCOM  xpcom/build/XPCOMInit.cpp:466
4  libxul.so  mozilla::dom::ContentProcess::Init  dom/ipc/ContentProcess.cpp:153

Comment 1

[BugBot [:suhaib / :marco/ :calixte]]

The severity field is not set for this bug.
:nika, could you have a look please?

For more information, please visit BugBot documentation.

Comment 2

[Nika Layzell [:nika] (ni? for response)]

There's not much context here - it seems like this could potentially be a system resource exhaustion issue (not having a pid to give to the new thread?), though I suppose it could also be an OOM. The stack is only 32kb, though, so an OOM seems more unlikely (https://searchfox.org/mozilla-central/rev/b91e9bef5a6d6f7b549fbc9cba70cb4665ed3866/js/xpconnect/src/XPCJSContext.cpp#166). It appears to be happening during a content process startup. It is interesting that this crash appears to have also happened on Windows a couple of times.

I don't see anything about the way Watchdog::Init uses the PR_CreateThread API which would make it more prone to this kind of issue, and I'm not sure how actionable this could be. Marking as S3 for now unless the rate picks up, in which case we may need to add more diagnostics or similar.

Comment 3

[BugBot [:suhaib / :marco/ :calixte]]

The bug is linked to a topcrash signature, which matches the following criterion:

• Top 10 desktop browser crashes on nightly

:nika, could you consider increasing the severity of this top-crash bug?

For more information, please visit BugBot documentation.

Comment 4

[Nika Layzell [:nika] (ni? for response)]

Looks like this crash may have been a fluke. The spike which put it into top-10 appears to be 67 crashes on April 2, and looking at crash-stats all but 59/67 of those crashes have the same install time, implying that one person probably got very unlucky.

Perhaps they were in a resource exhaustion situation and tried to reload all of their tabs, leading to a cascade of content process startup crashes?

There doesn't seem to be anything actionable here right now.

Comment 5

[BugBot [:suhaib / :marco/ :calixte]]

Based on the topcrash criteria, the crash signature linked to this bug is not a topcrash signature anymore.

For more information, please visit BugBot documentation.