335945 - Crash when cancelling copy of sent mail (large attachment) to Send folder

Status: RESOLVED FIXED

(Thunderbird :: Message Compose Window, defect)

Description

[alta88]

User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.0.2) Gecko/20060308 Firefox/1.5.0.2
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.0.2) Gecko/20060308 Firefox/1.5.0.2


In Tb v1.5.02 20060308 winxp.

Occured 3 times now, twice sending added attachment and once forwarding with large attachment (2x 3M, 1x 6M).  Account is Imap, configured to place copy in Sent folder on server.  The message is received normally. Never had this problem with NS 7.2 mail.

I'm mostly just testing the email part for suitability right now.  My general feeling is there are some big Imap problems, already lost a remote folder that lived happily in NS for years, but have to do some searches before commenting more, elsewhere.

Reproducible: Always

Comment 1

[Magnus Melin [:mkmelin]]

Did you get a talkback id? (If talkback is not installed, you can re-install and opt in for it in a custom install).

Comment 2

[alta88]

(In reply to comment #1)
> Did you get a talkback id? (If talkback is not installed, you can re-install
> and opt in for it in a custom install).
> 

I didn't make a note, sorry.  But for one of the crashes I did send the Talkback, no email tho; it would have been within an hour or so before the post, the comment was similar to this title.

Comment 3

[alta88]

hmm.  i just now recreated a crash, sent along a Talkback, but saw no reference # in the incident anywhere..

Comment 4

[Magnus Melin [:mkmelin]]

The id should be displayed in a list once the report has been sent and recieved.

I guess it's Incident ID: 18113678, which unfortunately does not tell alot.

Comment 5

[alta88]

(In reply to comment #4)
> The id should be displayed in a list once the report has been sent and
> recieved.
> 
> I guess it's Incident ID: 18113678, which unfortunately does not tell alot.
> 

although i could recreate it at will yesterday, i have since rebooted winxp.  it does not happen now - worst kind of problem solving case..

i had been doing a lot of imap testing/configuring etc over the weekend.  who knows what routine flipped what bit where.  i know that the Tb 'Error copying message' appeared, then the crash.  Normally that message appears (hit cancel), then the 'Message Sent successfully but error copying' appears, (hit ok).

Comment 6

[alta88]

just to let you know, this again is happening consistently.  if there's something specific i can do that might help figure it out, let me know.

Comment 7

[David :Bienvenu]

I believe I have a fix for this.

Comment 8

[David :Bienvenu]

Attachment: protect output stream from getting closed from an other thread

this is a variation of a patch sent to me by a contributor.

You always have to worry about deadlock when adding monitors, so I'll run with this for a bit to see if it's ok. It seems to fix the crash.

Comment 9

[David :Bienvenu]

putting on tb 2.0 radar - I'd like to let this patch bake on the trunk a little bit.

Comment 10

[alta88]

(In reply to comment #9)
> putting on tb 2.0 radar - I'd like to let this patch bake on the trunk a little
> bit.
> 

this has not reoccurred for me so far.

perhaps this is a separate bug, but the message handling is wrong when cancelling intentionally a (large) 'copy to sent'.  it should not be 

'Error copying to the Sent folder. Retry?' followed by a second
'Message was sent.... Return to Compose?'

but rather a

'Confirm cancel' Yes or No.

and then close Compose on Yes and finish, or continue the Copy on No.

Comment 11

[David :Bienvenu]

ok, thx, fixed on branch as well.

The incorrect error message is a known issue.

Comment 12

[Scott MacGregor]

clearing blocking flag on a fix which has already landed on the branch.

Comment 13

[Carsten Book [:Tomcat]]

verified fixed 1.8.1.3 with verified fixed 1.8.1.3 with Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US; rv:1.8.1.3) Gecko/20070326 Thunderbird/2.0.0.0 Mnenhy/0.7.5.0 ID:2007032620 no crash on the steps to reproduce - is working as expected - no crash -> adding verified keyword