1865305 - Crash in [@ nsPrintSettingsService::MaybeSavePrintSettingsToPrefs]

Status: RESOLVED WORKSFORME

(Core :: Printing: Setup, defect)

Description

[Gabriele Svelto [:gsvelto]]

Crash report: https://crash-stats.mozilla.org/report/index/486627c7-dcdf-4b88-95dc-ed8010231116

MOZ_CRASH Reason: MOZ_DIAGNOSTIC_ASSERT(false) (Print settings must be saved with a prefix)

Top 10 frames of crashing thread:

0  libxul.so  nsPrintSettingsService::MaybeSavePrintSettingsToPrefs  widget/nsPrintSettingsService.cpp:961
1  libxul.so  NS_InvokeByIndex  
2  libxul.so  CallMethodHelper::Invoke  js/xpconnect/src/XPCWrappedNative.cpp:1627
2  libxul.so  CallMethodHelper::Call  js/xpconnect/src/XPCWrappedNative.cpp:1180
2  libxul.so  XPCWrappedNative::CallMethod  js/xpconnect/src/XPCWrappedNative.cpp:1126
3  libxul.so  XPC_WN_CallMethod  js/xpconnect/src/XPCWrappedNativeJSOps.cpp:966
4  libxul.so  CallJSNative  js/src/vm/Interpreter.cpp:472
4  libxul.so  js::InternalCallOrConstruct  js/src/vm/Interpreter.cpp:566
4  libxul.so  InternalCall  js/src/vm/Interpreter.cpp:633
4  libxul.so  js::CallFromStack  js/src/vm/Interpreter.cpp:638

We have a single crash on file for this, but it's a diagnostic assert and a printing bug on top of it, so it's bound to be extremely low volume. I'm filing it so we can track it and eventually address it. If some information is needed from within the minidump to diagnose this please let me know and I'll dig it out.

Comment 1

[Daniel Holbert [:dholbert]]

Thanks. Looks like probably indicates that we're working with a nsPrintSettings object with an empty-string as the printer name, which I think is unexpected?

Tentatively triaging as S3 since it's just a single instance and we seem to handle the failure gracefully (returning NS_ERROR_FAILURE, refusing to write to an unprefixed pref value) in official release where diagnostic asserts aren't enabled. Also, let's add a dependency on bug 1668210 since that's where the assert was introduced.

The crash is on Ubuntu 23.10 (in Firefox 121 nightly), FWIW, so if/when we have cycles to investigate, we should be looking at the linux/gtk-specific print settings code.

Comment 2

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

Closing because no crashes reported for 12 weeks.