Open Bug 101715 Opened 24 years ago Updated 4 years ago

mozilla doesn't handle SIGXCPU graacefully

Categories

(SeaMonkey :: UI Design, defect)

All
Linux
defect
Not set
critical

Tracking

(Not tracked)

People

(Reporter: brianr, Unassigned)

References

Details

(Whiteboard: [2012 Fall Equinox])

From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:0.9.4) Gecko/20010913 BuildID: 2001091311 When mozilla receives SIGXCPU, it should clean up and then exit. For example, the mail client should save messages currently being composed in the drafts folder. Reproducible: Always Steps to Reproduce: 1. ulimit -S -t 30 ; ulimit -H -t 60 2. compose a message, but don't send yet. 3. use mozilla for a while 4. mozilla catches SIGXCPU, and exits without cleaning up or saving. Your message is lost. Actual Results: Mozilla quit abruptly when it reached the soft CPU time limit. Expected Results: Mozilla should have cleaned up as quickly as possible to avoid data loss and hitting the hard CPU limit.
ccing blizzard because he may know something about this...
Interesting. I don't know how to fix that. It's an interesting issue, though.
confirming
Status: UNCONFIRMED → NEW
Ever confirmed: true
not really my domain! Seems more a XPApp generic problem than just a msg compose one to me. Reassign...
Assignee: ducarroz → pchen
Component: Composition → XP Apps
Product: MailNews → Browser
QA Contact: sheelar → sairuh
->bryner
Assignee: pchen → bryner
-> future. I don't think this is something the vast majority of users will encounter.
Status: NEW → ASSIGNED
Target Milestone: --- → Future
Product: Core → Mozilla Application Suite
It might make sense to lump this bug into a more generic "clean up before crashing" bug if one exists. For example, there are other signals which indicate impending disaster but would still allow Mozilla to at least attempt some basic cleanup before exiting.
Assignee: bryner → jag
Status: ASSIGNED → NEW
QA Contact: bugzilla
Target Milestone: Future → ---
Assignee: jag → nobody
QA Contact: ui-design
How does SeaMonkey (how do the Mozilla apps) handle SIGTERM? Shouldn't the cases be similar?
Hardware: x86 → All
Whiteboard: [2012 Fall Equinox]
See Also: → 1641761
You need to log in before you can comment on or make changes to this bug.