298276 - Program freezes when sending mail.

Status: RESOLVED WORKSFORME

(Camino Graveyard :: General, defect)

Description

[Gregg]

User-Agent:       Mozilla/4.0 (compatible; MSIE 6.0b; Windows 98)
Build Identifier: Camino Version 2005061608 (0.9a1)

Once an e-mail is sent... I get nothing but beach ball on Camino. It fails to
respond, and has to be force quit. It is repeatable

Reproducible: Always

Steps to Reproduce:
1.Have MAil.app open
2.Open Camino
3.go back to email, send mail
4. back to camino and beach ball... needs a force quit.

Actual Results:  
Program froze

Expected Results:  
Gone on to the next task, just ignoring the sent mail.

Using Tiger 10.4.1, MAil Version 2.0.1 (730), G5, 2.7 dual, 2 gig ram,

Comment 1

[Smokey Ardisson (offline for a while; not following bugs - do not email)]

[e-mail]: Can you cause this to happen again and, before force quitting
Camino, open Activity Monitor (in the Utilities folder within the Applications
folder), select Camino in the list of processes and click the "Inspect" button
on the toolbar.

In the new window that appears, select the "Sample" button, and when the sample
is finished, save it to disk.  Then *attach* the text file to this bug (using
the "Create New Attachment" link above (do not paste into the comments box).  

This will help the developers figure out what is happening.  Thanks!

Comment 2

[Gregg]

Attachment: The result of Camino frozen after mail send

here is the state of Camino file...

Comment 3

[Mike Pinkerton (not reading bugmail)]

ew, the address book change causes us to lock up. i wonder why.

Comment 4

[Josh Aas]

Sort of looks like we should be hanging up in fillAddressBook, but the
notification isn't even getting there. I have no idea how that sample is
happening. The notification has been sent and is making its way to our specified
method, but Camino code isn't even getting called.

Potential problems...
1) we have a problem with our event handling
2) we're initializing or leaving the address book in a bad state
3) its Apple's bug
4) its a problem with the reporter's system

I'm going to guess #4. I recommend that if we do not have anybody that can
duplicate this within the next week we bump it from the 0.9 list.

Comment 5

[Jasper]

bumping of unconfirmed so we can more easily track any progress.

Comment 6

[Josh Aas]

For the record, I could not reproduce this. My earlier comment was just an
analysis of the description and the stack trace. So it seems that only the
reporter repro, which means this is definitely not confirmed.

Comment 7

[Josh Aas]

-> 1.0 for reasons in previous comments. we'll see if we have any new info them.

Comment 8

[Simon Fraser [no longer active]]

Works for me. Please reopen if you can reproduce.