106479 - Check "Empty Trash on Exit" from account setting freeze an IMAP account without Trash folder

Status: VERIFIED FIXED

(MailNews Core :: Networking: IMAP, defect)

Description

[Terence Dewaele]

Hello,

I have 3 mail account in IMAP 4 and only one have a Trash folder.
If I check Empty Trash (In account preference) on exit on account with haven't a
Trash Folder, on exit Mozilla is frozen !!!

Sorry for my english i'am french :)

Thx for mozilla it's good browser !!!

Comment 1

[Karen Huang]

Multiple accounts -> Ninoschka.
Updating the summary a little bit to specify this problem.
Can you describe what kind of the multiple IMAP accounts that you have?

Comment 2

[Karen Huang]

adding keywords: hang

Comment 3

[Karen Huang]

All of our internal IMAP interop server have trash folder unless "AOL" or 
"unspecify trash on other IMAP delete modes". There are existing bugs for 
disable empty trash on exit for other IMAP delete modes and AOL. The important 
thing is: How Netscape handle "Empty Trash on Exit" associate with different 
accounts w/ trash & wo/ trash folder......

Comment 4

[Terence Dewaele]

3 accounts in IMAP.
My provider is not AOL.
But if i uncheck checkbox "Empty trash on exit" mozilla work.
Because 2/3 account haven't Trash Folder.

Resume:
If i check "Empty trash on exit" and i haven't Trash Folder, mozilla freeze !

Comment 5

[Chris Chernesky]

I can reproduce this with one or multiple IMAP accounts. I am currently using
build 2002022308.

It is as described. If there is not a trash folder on the IMAP server, and
Mozilla is set to empty trash folders on exit, then Mozilla will hang and will
need to be terminated forcefully..

If that option (empty trash on exit) is unchecked, Mozilla exits correctly..

Aside from this small problem, Mozilla's mail support is the best that I've
found. Thanks for a great product!

Comment 6

[Vadim Berezniker]

*** Bug 132645 has been marked as a duplicate of this bug. ***

Comment 7

[Vadim Berezniker]

confirming based on comment #5 and based on dupe.

os -> all

Comment 8

[Vadim Berezniker]

*** Bug 108847 has been marked as a duplicate of this bug. ***

Comment 9

[Henry Jia]

Attachment: patch for review

The reason for this bug is because the return value is ignored and mozilla
continue to work on a wrong status.

Comment 10

[Henry Jia]

Can anyone take a look at this bug and review, superreview the patch(attachment
78158 [details] [diff] [review])? Thanks.

r= & sr= wanted

Comment 11

[Scott MacGregor]

Comment on attachment 78158 [details] [diff] [review]
patch for review

sr=mscott

Although in the future, please get a first level review before requesting a
super review. There are more reviewers than sr's out there so it cuts down on
the time we have to look at a bug if the first level review catches anything.

thanks for the patch.

Comment 12

[Henry Jia]

Thanks, mscott.
There is another bug 120485 that needs your superreview. One of the bug's patch
has passed r=mscott and sr=bienvenu, and you suggested me to give a check for
memory allocation, then I gave the required patch again. It has passed
r=bienvenu. I hope you can sr= it. The bug is about a basic function of IMAP.
Thanks.

Henry

Comment 13

[timeless]

+        return NS_OK;
+    }
+    else
+    {
+        return rv;
     }
-    return NS_OK;
your patch is creating an else after return.

change it so that you have:
+        return NS_OK;
     }
-    return NS_OK;
+    return rv;

instead. with that, r=timeless

Comment 14

[Henry Jia]

Attachment: revised patch

According to timeless' suggestion, I give the new patch.
mscott, can you sr= it again? Thanks.

Comment 15

[timeless]

Comment on attachment 79565 [details] [diff] [review]
revised patch

carrying over sr

Comment 16

[Pete Zha]

Fix checked in

Comment 17

[Vadim Berezniker]

*** Bug 125355 has been marked as a duplicate of this bug. ***

Comment 18

[Scott MacGregor]

Should we carry this over into the 1.0 branch?

Comment 19

[Mike Shaver (:shaver -- probably not reading bugmail closely)]

Comment on attachment 79565 [details] [diff] [review]
revised patch

a=shaver for 1.0 branch.

Comment 20

[Karen Huang]

*** Bug 98328 has been marked as a duplicate of this bug. ***

Comment 21

[Karen Huang]

Removing multiple accounts in the summary since problem can be reproduced in a 
single account as bug 98328 description.

Copy/Paste comments from bug 98328

---------------------------------------------------
Comment #11 From [e-mail] 2001-10-02 10:39:

OK, I can reproduce it now in these steps:
1) Create new profile
2) Go to mail/news, create IMAP account.
3) Change preferences to "mark as deleted"
4) Unsubscribe "Trash" folder. I disappears from view.
5) Change preferences to "move to trash" and check "delete trash on exit"
6) Exit mail/news
--------------------------------------------------------------------------

Comment 22

[Christopher Blizzard (:blizzard)]

Comment on attachment 79565 [details] [diff] [review]
revised patch

a=blizzard on behalf of drivers for 1.0

Comment 23

[Scott MacGregor]

checked into the 1.0 branch.

Comment 24

[David :Bienvenu]

*** Bug 144355 has been marked as a duplicate of this bug. ***

Comment 25

[Dawn Endico]

this was backed out of the branch last night in an attempt to fix orangeness.
Please check back in asap.

Comment 26

[Scott MacGregor]

as soon as the branch opens for checkins again, I'll land this back on the
branch. I backed it out to see if it was causing the mail smoketests on
tinderbox to go orange. It was not the cause. removing fixed1.0.0

Comment 27

[Scott MacGregor]

adding fixed1.0.0 keyword again. this is back in the branch.

Comment 28

[Karen Huang]

Verified on all the platforms, no hang or freeze anymore for comment #21 
scenario.
06-13-07-1.0.0
06-13-07-1.0.0
06-13-05-1.0.0
Marking as verified and changing keywords from fixed1.0.0 (Scott, should befixed 
1.0.1, right?) to verified1.0.1.

Comment 29

[Karen Huang]

Oops! I mean:
Windows 06-13-07-1.0.0
Linux 06-13-07-1.0.0
Mac 06-13-05-1.0.0

Comment 30

[Aleksey Nogin]

If the Trash folder does exist, but the server can not be reached, will Mozilla
still hang? Please see bug 184576.