72492 - Mail is unstable on the trunk

Status: VERIFIED WORKSFORME

(SeaMonkey :: MailNews: Message Display, defect, P1)

Description

[Phil Peterson]

Using trunk builds from late last week and Saturday, I'm crashing all over the
place. Usually a second or so after deleting a message, or sending one.

I've filed a bunch of Talkback reports, but the query tool just generates error
messages

Comment 1

[chris hofmann]

here is a list of the 'phil' crashes to go hunting for when the 
talkback report server comes back up...

build id, incident id, data of submission, platform, phil, stack sig, comment

2001031515 27876132 2001-03-16 Windows NT  4.0 build 1381 [e-mail] 
gklayout.dll + 0x73529 (0x60353529) 8ce19988 ?
2001031515 27876130 2001-03-16 Windows NT  4.0 build 1381 [e-mail] 
gklayout.dll + 0x73529 (0x60353529) 8ce19988 composing AIM
2001031515 27875738 2001-03-16 Windows NT  4.0 build 1381 [e-mail] 
nspr4.dll + 0x39ba (0x60c839ba) 196ea378 ?
2001031515 27861857 2001-03-16 Windows NT  4.0 build 1381 [e-mail] 
gklayout.dll + 0x73567 (0x60353567) cd73e7a3 aim buddy list
2001031515 27860142 2001-03-16 Windows NT  4.0 build 1381 [e-mail] 
gklayout.dll + 0x73567 (0x60353567) cd73e7a3 ?

Comment 2

[Phil Peterson]

I thought I'd crashed in Necko.dll, but talkback probably remembers better than
I do...

Comment 3

[scottputterman]

nominating for nsbeta1.  I've also crashed randomly in the last couple of days.
 I'll see what my talkback reports say when talkback comes back up.

Comment 4

[scottputterman]

I finally got to see my talkback reports and it looks like I've had this one
happen to me a number of times as well.

Comment 5

[scottputterman]

reassigning to sspitzer.  I'm reassigning to Seth because I think we had a bunch 
of outliner crashes that he fixed last week which would be in gklayout and I 
haven't been crashing as much as I was before some of these went in.

Phil, are you still getting this crash?

Comment 6

[lchiang]

I am crashing just as much using a build Win32 2001-03-27-12-trunk doing various
things like Phil is.  Unfortunately, due to a bug in Talkback, the stack traces
don't provide useful stack traces.  Shiva tells me that this will be fixed in
tomorrow's builds so hopefully, the crash stack trace can prove more useful. 
Once we have better stack trace, we can probably narrow exactly which steps are
to reproduce this so that we can tie all the crashes together.

Comment 7

[(not reading, please use [e-mail] instead)]

this could be related to #73840

I'd need to see some stack traces to confirm.

Comment 8

[lchiang]

There may be many reports here.  The sending crashes and viewing crashes could
be part of bugs:
http://bugzilla.mozilla.org/show_bug.cgi?id=73742 (which was duped and then
marked fixed very recently)
or
http://bugzilla.mozilla.org/show_bug.cgi?id=74035

Comment 9

[scottputterman]

I haven't been noticing this general problem anymore. I've seen some crashes 
with messages with images and I've seen the crash on exit, but not the random 
crashes I was seeing when this bug is filed? Does anyone have any recent stack 
traces to add to this to support keeping this open?  

73840, 73742 are fixed and and 74035 is assigned to pavlov.

Comment 10

[scottputterman]

I'm going to mark WFM because I haven't noticed any of the crashes outside the
ones I mentioned above which already have bugs.

Comment 11

[esther]

verifying as worksforme, I'm not crashing using mail sending, receiving, 
deleting, saving, copying and moving.  All the bugs mentioned in this bug and 
their duplicates are verified.