Closed
Bug 273560
Opened 20 years ago
Closed 20 years ago
Crash when selecting inbox after selecting saved search folder before search complete [@ 0x00000000 - nsMsgSearchSession::NotifyListenersDone]
Categories
(Thunderbird :: General, defect)
Tracking
(Not tracked)
VERIFIED
FIXED
People
(Reporter: pdm9, Assigned: Bienvenu)
References
Details
(Keywords: crash)
Crash Data
Attachments
(2 files)
|
812 bytes,
patch
|
mscott
:
superreview+
asa
:
approval-aviary1.1a1+
|
Details | Diff | Splinter Review |
|
525 bytes,
patch
|
mscott
:
superreview+
chofmann
:
approval-aviary1.1a1+
|
Details | Diff | Splinter Review |
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.7.5) Gecko/20041107 Firefox/1.0 Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.7.5) Gecko/20041107 Firefox/1.0 Created a saved search folder searching for a text string in message body. After saving folder, clicked on saved search folder to start search. While search still in progress, clicked another folder in same account, and Thunderbird crashed. Problem seen since introduction of saved search folders Reproducible: Always Steps to Reproduce: 1. Create a saved search folder with a time-consuming search, ie. search message body 2. Click on the saved folder to start the search 3. Before the saved search is complete, click on another folder 4. Thunderbird crashes Actual Results: Thunderbird crashed Expected Results: saved search to be canceled and selected folder contents displayed
Comment 1•20 years ago
|
||
I can confirm this behaviour in Thunderbird version 0.9 (20041103). However, it seems to only happen the first time you create a saved search. After letting the first search complete, I could create new complex search-folders and change to other folders before the search-folder operation had completed. Can you confirm this, Paul?
| Reporter | ||
Comment 2•20 years ago
|
||
Yes, following the procedure listed below does *NOT* cause a crash: 1. Run a time-intensive saved-search folder to completion. 2. Click on another time-intensive saved search folder 3. Before 2nd search is done, click on any other folder or saved-search folder 4. Thunderbird does not crash This behavior validated on Thunderbird 1.0 (20041206)
Comment 3•20 years ago
|
||
I continued with Thunderbird version 1.0 (20041206), and the crash can still be reproduced. This time I used a slightly different approach than the test I did with 0.9: 1. Select a mail folder with ~3000 messages in 2. ctrl-shift-f, search for messages where body contains -- 3. Start search 4. Quickly press 'Save as Search folder', name search 'aa' 5. Exit 'Search Messages' dialog 6. Press the newly created "aa" search folder 7. Continue selecting other mail folders, including the "aa" search folder. 8. Thunderbird will crash after a few clicks.
Comment 4•20 years ago
|
||
The bug definitely exists in Thunderbird Version 1.0 (20041206). I use a saved search folder for "is Junk" on my inbox (~300 mails) and when I change quickly back and forth between the inbox and the saved search folder it crashes after some (from 5 to 25) clicks. When I use more than one saved search folder, the problem seems to happen more often.
Comment 5•20 years ago
|
||
The bug definitely exists in Thunderbird Version 1.0 (20041206). I use a saved search folder for "is Junk" on my inbox (~300 mails) and when I change quickly back and forth between the inbox and the saved search folder it crashes after some (from 5 to 25) clicks. When I use more than one saved search folder, the problem seems to happen more often.
I also repeatedly encounter this bug -- not just when selecting the Inbox, but when selecting any other folder. In particular, sometimes a new mail message seems to appear in the saved search folder after mail download. In which case, the following leads to a crash, always: 1. Read new mail in Inbox, press Space; 2. When asked "Advance to next unread message in [Saved Search Folder]", click Yes (specifically, by hitting the spacebar). 3. The search begins; before it ends, press space; 4. When asked "Advance to next unread message in [other folder]", click Yes (space). 5. Thunderbird crasehs.
Comment 7•20 years ago
|
||
Particularly since Thunderbird lacks auto-save of Compose-in-progress, I think this bug should be elevated to Blocking -- it's caused me to lose so much work (Compose-in-progress) that I'm giving up on Saved Searches (aka Virtual Folders) until it gets fixed.
Updated•20 years ago
|
Flags: blocking-aviary1.1+
| Assignee | ||
Comment 8•20 years ago
|
||
trunk builds have auto save of composed messages. In any case, you're not allowed to set the blocking + flag unless you are a driver... I have not been able to reproduce this crash, but I'll try some more.
Flags: blocking-aviary1.1+ → blocking-aviary1.1?
Comment 9•20 years ago
|
||
Reproduced again in Thunderbird version 1.0.2 (20050317) on Windows XP SP2 running on a 2 GHz Mobile Pentium 4 notebook (IBM ThinkPad T30). What I've got is quite a few different Saved Searches (aka Virtual Folders), all of which search multiple real folders, including a (real) folder with over 1300 messages, so each Saved Search takes several seconds to complete. I normally have 10 Extensions installed, and with all of them enabled I'll get a crash every time when clicking on a new folder (real or virtual) before a Saved Search has completed. When I disable all 10 Extentions, the frequency of crash goes way down, but I can still get a crash when quickly clicking on several Saved Searches in sequence (Incident ID TB5176714K) or when quickly clicking back and forth between Saved Searches and real folders (Incident ID TB5177201E). I've checked to see if any particular Extension has any bearing, but it seems just a matter of frequency of crashing going up as more Extensions are enabled. I speculate that this may be related to the new bug 290851 I just entered, since both are crashes that seem to be caused by Saved Search not being allowed to complete. Thanks for the explanation on the Blocking flag -- I'm new to this, and the on-line help wasn't terribly helpful. ;) I'm reluctant to use a Trunk Build just to get Compose Auto-Save -- I think that might be jumping from the frying pan into the fire. :)
| Assignee | ||
Comment 10•20 years ago
|
||
OK, I've reproduced this, and I believe I have a fix.
Assignee: mscott → bienvenu
Comment 11•20 years ago
|
||
Great! Terrific! Thank you!
| Assignee | ||
Comment 12•20 years ago
|
||
Attachment #181299 -
Flags: superreview?(mscott)
Updated•20 years ago
|
Attachment #181299 -
Flags: superreview?(mscott) → superreview+
| Assignee | ||
Comment 13•20 years ago
|
||
Comment on attachment 181299 [details] [diff] [review] proposed fix simple null check to prevent frequent crasher.
Attachment #181299 -
Flags: approval-aviary1.1a?
Comment 14•20 years ago
|
||
Comment on attachment 181299 [details] [diff] [review] proposed fix a=asa
Attachment #181299 -
Flags: approval-aviary1.1a? → approval-aviary1.1a+
| Assignee | ||
Comment 15•20 years ago
|
||
fixed on trunk.
Status: NEW → RESOLVED
Closed: 20 years ago
Resolution: --- → FIXED
Comment 16•20 years ago
|
||
If the fix made it into the Thunderbird Trunk Build 20-Apr-2005, then the problem seems better, but not completely fixed -- see Incident ID TB5241578K.
Comment 17•20 years ago
|
||
To amplify my prior post, I ran another test, clicking back and forth between long and complex Saved Searches (of multiple real folders, including a real folder with over 1300 messages) and real folders. 10 extensions are installed, of which 8 are enabled (2 not playing nice with this build). After about a half dozen such clicks, I got another crash -- see Incident ID TB5249131K.
Comment 18•20 years ago
|
||
Incident ID: 5176714 Stack Signature 0x00000000 60c6166f Product ID Thunderbird10 Build ID 2005031711 Trigger Time 2005-04-18 11:31:10.0 Platform Win32 Operating System Windows NT 5.1 build 2600 Module URL visited User Comments Clicked on several different long Saved Searches without waiting for each one to finish, with a crash on about the 8th click. Since Last Crash 11 sec Total Uptime 1324993 sec Trigger Reason Access violation Source File, Line No. N/A Stack Trace 0x00000000 nsMsgSearchSession::NotifyListenersDone [e:/builds/tinderbox/Tb-Aviary1.0.1/WINNT_5.0_Depend/mozilla/mailnews/base/search/src/nsMsgSearchSession.cpp, line 586] nsMsgSearchSession::TimerCallback [e:/builds/tinderbox/Tb-Aviary1.0.1/WINNT_5.0_Depend/mozilla/mailnews/base/search/src/nsMsgSearchSession.cpp, line 525] nsTimerImpl::Fire [e:/builds/tinderbox/Tb-Aviary1.0.1/WINNT_5.0_Depend/mozilla/xpcom/threads/nsTimerImpl.cpp, line 382] nsAppShellService::Run [e:/builds/tinderbox/Tb-Aviary1.0.1/WINNT_5.0_Depend/mozilla/xpfe/appshell/src/nsAppShellService.cpp, line 495] main [e:/builds/tinderbox/Tb-Aviary1.0.1/WINNT_5.0_Depend/mozilla/mail/app/nsMailApp.cpp, line 58] kernel32.dll + 0x16d4f (0x7c816d4f) 5177201 was the same as 5176714 Incident ID: 5241578 Stack Signature nsMsgSearchSession::GetRunningScope 12dab36d Product ID ThunderbirdTrunk Build ID 2005042006 Trigger Time 2005-04-20 23:05:31.0 Platform Win32 Operating System Windows NT 5.1 build 2600 Module thunderbird.exe + (0042abbd) URL visited User Comments Testing fix to Bug 273560 Since Last Crash 894 sec Total Uptime 1021 sec Trigger Reason Access violation Source File, Line No. e:/builds/tinderbox/thunderbird-trunk/WINNT_5.0_Depend/mozilla/mailnews/base/search/src/nsMsgSearchSession.cpp, line 645 Stack Trace nsMsgSearchSession::GetRunningScope [e:/builds/tinderbox/thunderbird-trunk/WINNT_5.0_Depend/mozilla/mailnews/base/search/src/nsMsgSearchSession.cpp, line 645] nsTimerImpl::Fire [e:/builds/tinderbox/thunderbird-trunk/WINNT_5.0_Depend/mozilla/xpcom/threads/nsTimerImpl.cpp, line 383] nsAppStartup::Run [e:/builds/tinderbox/thunderbird-trunk/WINNT_5.0_Depend/mozilla/toolkit/components/startup/src/nsAppStartup.cpp, line 145] main [e:/builds/tinderbox/thunderbird-trunk/WINNT_5.0_Depend/mozilla/mail/app/nsMailApp.cpp, line 61] kernel32.dll + 0x16d4f (0x7c816d4f) Incident ID: 5249131 Stack Signature nsMsgSearchScopeTerm::TimeSlice e1f2e065 Product ID ThunderbirdTrunk Build ID 2005042006 Trigger Time 2005-04-21 08:47:02.0 Platform Win32 Operating System Windows NT 5.1 build 2600 Module thunderbird.exe + (0042cb2c) URL visited User Comments Testing fix for Bug 273560 in Thunderbird Trunk Build 20-Apr-2005. Since Last Crash 33526 sec Total Uptime 34547 sec Trigger Reason Access violation Source File, Line No. e:/builds/tinderbox/thunderbird-trunk/WINNT_5.0_Depend/mozilla/mailnews/base/search/src/nsMsgSearchTerm.cpp, line 1494 Stack Trace nsMsgSearchScopeTerm::TimeSlice [e:/builds/tinderbox/thunderbird-trunk/WINNT_5.0_Depend/mozilla/mailnews/base/search/src/nsMsgSearchTerm.cpp, line 1494] nsTimerImpl::Fire [e:/builds/tinderbox/thunderbird-trunk/WINNT_5.0_Depend/mozilla/xpcom/threads/nsTimerImpl.cpp, line 383] nsAppStartup::Run [e:/builds/tinderbox/thunderbird-trunk/WINNT_5.0_Depend/mozilla/toolkit/components/startup/src/nsAppStartup.cpp, line 145] main [e:/builds/tinderbox/thunderbird-trunk/WINNT_5.0_Depend/mozilla/mail/app/nsMailApp.cpp, line 61] kernel32.dll + 0x16d4f (0x7c816d4f)
Summary: Crash when selecting inbox after selecting saved search folder before search complete → Crash when selecting inbox after selecting saved search folder before search complete [@ 0x00000000 - nsMsgSearchSession::NotifyListenersDone]
Updated•20 years ago
|
Flags: blocking-aviary1.1?
Comment 19•20 years ago
|
||
Is this fix in Mozilla Thunderbird trunk build 28-Apr-2005 1014? If so, then the problem is *not* completely fixed, because it easily crashed in my tests of clicking on Inbox before a long complex Saved Search had completed -- see Incidents TB5451173Y, TB5451511Z, and TB5451569X.
| Assignee | ||
Comment 20•20 years ago
|
||
those talkback id's all show a seemingly un-related stack trace: 0x0267cb7e nsMessengerMigrator::SetMailCopiesAndFolders [e:/builds/tinderbox/Tb-Aviary1.0.1/WINNT_5.0_Depend/mozilla/mailnews/base/src/nsMessengerMigrator.cpp, line 919] xpcom_core.dll + 0x2b44b (0x6035b44b) nsNSSComponent::StopCRLUpdateTimer [e:/builds/tinderbox/Tb-Aviary1.0.1/WINNT_5.0_Depend/mozilla/security/manager/ssl/src/nsNSSComponent.cpp, line 882] and the build id is 2005031711, which is from before the fix was checked in, if that build id is being correctly reported by talkback.
Comment 21•20 years ago
|
||
On the issue of the Build ID in my last series of Incidents, that may be because I installed "version 1.0+ (20050428)" over the top of the 1.0.2 Release. To check, I uninstalled all Extensions and Thunderbird, deleted the Program Files directory, and then did a fresh install of "version 1.0+ (20050428)". When I triggered the crash, Talkback did *not* come up. Even when I tried to launch Talkback.exe directly, it would *not* start up. Reinstalling Thunderbird did *not* help. Something is wrong with Talkback in that build. So I uninstalled Thunderbird again, and this time did a fresh install of 2005-04-27-05-trunk (the prior nightly trunk build), verified after install as "version 1.0+ (20050427)". I easily forced the crash again and sent it in as Incident TB5457628Q, which correctly reports: Build Identifier 2005042704 Deployment Identifier MozillaOrgThunderbirdTrunkWin322005042704 Please check this incident. Thank you.
| Assignee | ||
Comment 22•20 years ago
|
||
ah, much better: 0x00000000 nsMsgSearchSession::NotifyListenersDone [e:/builds/tinderbox/thunderbird-trunk/WINNT_5.0_Depend/mozilla/mailnews/base/search/src/nsMsgSearchSession.cpp, line 586] nsMsgSearchSession::TimerCallback [e:/builds/tinderbox/thunderbird-trunk/WINNT_5.0_Depend/mozilla/mailnews/base/search/src/nsMsgSearchSession.cpp, line 525] nsTimerImpl::Fire [e:/builds/tinderbox/thunderbird-trunk/WINNT_5.0_Depend/mozilla/xpcom/threads/nsTimerImpl.cpp, line 383]
Status: RESOLVED → REOPENED
Resolution: FIXED → ---
| Assignee | ||
Comment 23•20 years ago
|
||
we should be interrupting the search before deleting the search session...I've had this patch in my tree for a while - it might be why I haven't been able to reproduce this crash.
Attachment #182201 -
Flags: superreview?(mscott)
Updated•20 years ago
|
Attachment #182201 -
Flags: superreview?(mscott) → superreview+
| Assignee | ||
Updated•20 years ago
|
Attachment #182201 -
Flags: approval-aviary1.1a?
Comment 24•20 years ago
|
||
Comment on attachment 182201 [details] [diff] [review] an other fix... a=chofmann
Attachment #182201 -
Flags: approval-aviary1.1a? → approval-aviary1.1a+
| Assignee | ||
Comment 25•20 years ago
|
||
please try tomorrow's trunk build - I don't know for sure if this will fix it or not...
Status: REOPENED → RESOLVED
Closed: 20 years ago → 20 years ago
Resolution: --- → FIXED
Comment 26•20 years ago
|
||
I confirm this problem as RESOLVED-FIXED in Thunderbird nightly trunk 03-May-2005 0850. Thank you!
Comment 28•20 years ago
|
||
CAVEAT: While the crash has been fixed, there remains a behavior problem. If a long and complex Saved Search is selected, a regular folder is then selected, and a long and complex Saved Search is then immediately selected (without any delay in the regular folder), that Saved Search will not be performed, staying with an empty result set forever. Work-around: When selecting a regular folder after having selected a Saved Search, wait a few seconds in the regular folder before selecting a Saved Search.
Comment 29•20 years ago
|
||
*** Bug 290851 has been marked as a duplicate of this bug. ***
Comment 30•20 years ago
|
||
Hello! I just installed this build: version 1.0+ (20050504) ... because I wanted to verify wether the bug still exists or not. At first view it seemed to be fixed. I moved quick between the saved search folders (partyally verly large) and some previewd mails. No crash anymore ... ... until I closed thunderbird afterwards. Then the talkback agent cames up. I startet Thunderbird again and used it without saved search. No crash. Closed Thunderbird. No crash. Then I started Thunderbird again. Moved quick between saved search folders. Closed Thunderbird and the Talkback agent came up. I do not know whether it is the same bug-reason or it is a new bug.
Comment 31•20 years ago
|
||
Yesterday I could easyly reproduce the incedent. Today not. Here are the incedent numbers: TB5615081W, TB5594351Y, TB5594323X
Comment 32•20 years ago
|
||
With Thunderbird nightly trunk 06-May-2005 0755, I cannot produce the problem described in Additional Comment #30 From delcour@libertus.de 2005-05-05 05:04 PDT
Comment 33•19 years ago
|
||
Seems like the bug is still there... (TB 1.0.6, installed over 1.0.2). I have just sent Talkback details (including this bug number). Brief description of incident: I pressed Space (for "Next unread"); a dialog popped up, "Advance to next unread message in [search folder]?"; I hit Space for "Yes" (the default). TB moved to the search folder; I *very quickly* pressed Space again (to move to the next folder) and Space (for "Yes"), and... boom.
Comment 34•19 years ago
|
||
(In reply to comment #33) > Seems like the bug is still there... (TB 1.0.6, installed over 1.0.2). I have > just sent Talkback details (including this bug number). That is because it is fixed in the trunk and not the branch. The fix will be included in Thunderbrid 1.5 when it is released.
Comment 35•19 years ago
|
||
*** Bug 277418 has been marked as a duplicate of this bug. ***
Comment 36•19 years ago
|
||
Seems to be happening again (TB1.5). I've sent several reports via Talkback; any need to upload a Talkback dump as an attachment here?
Comment 37•19 years ago
|
||
(In reply to comment #36) > Seems to be happening again (TB1.5). I've sent several reports via Talkback; > any need to upload a Talkback dump as an attachment here? Confirmed. See Incident ID TB14614299X. It can be hard to reproduce -- only happens some of the time -- but there's definitely still a bug -- it needs to be reopened!
| Assignee | ||
Comment 38•19 years ago
|
||
Tal, yes, incident id's would be helpful. John, that looks like a completely different stack trace, so I suspect that's a different bug, probably with similar steps to reproduce. It would be useful to know if Tal's stack trace is the same as yours.
Comment 39•19 years ago
|
||
so, my quick read of branches doesn't show this patch made the 18 branch or the tb 15 release which means I wouldn't expect the crashes not to happen there. jaime was slightly misinformed. currently it looks like TB2 would be the first release unless this makes a tb1.5.x.y release.
Comment 40•19 years ago
|
||
All of tal's crashes look like this one by (john) Incident ID: 14614299 Stack Signature 0x0331ff51 3088fab0 Product ID Thunderbird15 Build ID 2005120115 Trigger Time 2006-02-01 00:58:13.0 Platform Win32 Operating System Windows NT 5.1 build 2600 Module URL visited User Comments Reproduced Bugzilla Bug 273560 Since Last Crash 1022090 sec Total Uptime 1716145 sec Trigger Reason Access violation Source File, Line No. N/A Stack Trace 0x0331ff51 nsMsgLocalMailFolder::OnStopRunningUrl [e:/builds/tinderbox/Tb-Mozilla1.8/WINNT_5.0_Depend/mozilla/mailnews/local/src/nsLocalMailFolder.cpp, line 3340] nsUrlListenerManager::BroadcastChange [e:/builds/tinderbox/Tb-Mozilla1.8/WINNT_5.0_Depend/mozilla/mailnews/base/src/nsUrlListenerManager.cpp, line 97] nsUrlListenerManager::OnStopRunningUrl [e:/builds/tinderbox/Tb-Mozilla1.8/WINNT_5.0_Depend/mozilla/mailnews/base/src/nsUrlListenerManager.cpp, line 123] nsMsgProtocol::OnStopRequest [e:/builds/tinderbox/Tb-Mozilla1.8/WINNT_5.0_Depend/mozilla/mailnews/base/util/nsMsgProtocol.cpp, line 397] nsMailboxProtocol::OnStopRequest [e:/builds/tinderbox/Tb-Mozilla1.8/WINNT_5.0_Depend/mozilla/mailnews/local/src/nsMailboxProtocol.cpp, line 395] nsInputStreamPump::OnStateStop [e:/builds/tinderbox/Tb-Mozilla1.8/WINNT_5.0_Depend/mozilla/netwerk/base/src/nsInputStreamPump.cpp, line 507] 0x0350a230 nsMailboxProtocol::`vftable' nsGopherChannel::AddRef [e:/builds/tinderbox/Tb-Mozilla1.8/WINNT_5.0_Depend/mozilla/netwerk/protocol/gopher/src/nsGopherChannel.cpp, line 98] 0x8b560c45 They are: Incident ID: 14597311 Incident ID: 14583543 Incident ID: 14441697 Incident ID: 14436762 Except for tal's last one, which is: Incident ID: 14346249 Stack Signature 0x00000000 b11039ed Product ID Thunderbird15 Build ID 2005120115 Trigger Time 2006-01-24 00:23:01.0 Platform Win32 Operating System Windows NT 5.1 build 2600 Module URL visited User Comments Undo-delete did not work; I opened the Trash folder, and.... crash. Since Last Crash 148235 sec Total Uptime 148235 sec Trigger Reason Access violation Source File, Line No. N/A Stack Trace 0x00000000 nsMsgMailSession::OnItemEvent [e:/builds/tinderbox/Tb-Mozilla1.8/WINNT_5.0_Depend/mozilla/mailnews/base/src/nsMsgMailSession.cpp, line 262] nsMsgDBFolder::NotifyFolderEvent [e:/builds/tinderbox/Tb-Mozilla1.8/WINNT_5.0_Depend/mozilla/mailnews/base/util/nsMsgDBFolder.cpp, line 4644] nsMsgDBFolder::OnStopRunningUrl [e:/builds/tinderbox/Tb-Mozilla1.8/WINNT_5.0_Depend/mozilla/mailnews/base/util/nsMsgDBFolder.cpp, line 1306] nsMsgLocalMailFolder::OnStopRunningUrl [e:/builds/tinderbox/Tb-Mozilla1.8/WINNT_5.0_Depend/mozilla/mailnews/local/src/nsLocalMailFolder.cpp, line 3355] nsUrlListenerManager::BroadcastChange [e:/builds/tinderbox/Tb-Mozilla1.8/WINNT_5.0_Depend/mozilla/mailnews/base/src/nsUrlListenerManager.cpp, line 97] nsUrlListenerManager::OnStopRunningUrl [e:/builds/tinderbox/Tb-Mozilla1.8/WINNT_5.0_Depend/mozilla/mailnews/base/src/nsUrlListenerManager.cpp, line 123] nsMsgProtocol::OnStopRequest [e:/builds/tinderbox/Tb-Mozilla1.8/WINNT_5.0_Depend/mozilla/mailnews/base/util/nsMsgProtocol.cpp, line 397] nsMailboxProtocol::OnStopRequest [e:/builds/tinderbox/Tb-Mozilla1.8/WINNT_5.0_Depend/mozilla/mailnews/local/src/nsMailboxProtocol.cpp, line 395] nsInputStreamPump::OnStateStop [e:/builds/tinderbox/Tb-Mozilla1.8/WINNT_5.0_Depend/mozilla/netwerk/base/src/nsInputStreamPump.cpp, line 507] 0x0356b460 nsMailboxProtocol::`vftable' nsGopherChannel::AddRef [e:/builds/tinderbox/Tb-Mozilla1.8/WINNT_5.0_Depend/mozilla/netwerk/protocol/gopher/src/nsGopherChannel.cpp, line 98] 0x8b560c45
Comment 41•19 years ago
|
||
timeless, thanks for digging out my incidents. Incident 14346249 seems related to a different bug; I was not dealing with a search folder when it happened.
Comment 42•19 years ago
|
||
(In reply to comment #38) > ... John, that looks like a completely > different stack trace, so I suspect that's a different bug, probably with > similar steps to reproduce. ... I think it probably is a different bug because I haven't been able to reproduce it in many tries. When that one crash occurred, TB had been running for quite some time. I clicked a complex saved search, then Local Folders (by accident), then Inbox, whereupon it crashed. Since then I haven't been able to make it crash.
Comment 43•19 years ago
|
||
(In reply to comment #38) > ... John, that looks like a completely > different stack trace, so I suspect that's a different bug, probably with > similar steps to reproduce. ... Happened again: TB14660524G
Comment 44•19 years ago
|
||
I suspect (no confirmation yet) that this happens if TB is building the summary file for some folder or another in the background (perhaps the folder we left before entering the search, or we select after leaving the search -- no idea). Can this be checked from the incident data?
Comment 45•19 years ago
|
||
(In reply to comment #44) > I suspect (no confirmation yet) that this happens if TB is building the summary > file for some folder or another in the background (perhaps the folder we left > before entering the search, or we select after leaving the search -- no idea). > Can this be checked from the incident data? I seem to remember that TB was building a summary file for the Trash folder when TB14614299X happened. However, I'm unable to reproduce the problem by forcing a summary file rebuild before clicking on the saved search and then inbox.
Comment 46•19 years ago
|
||
john: you need a newer build, try trunk in a few weeks (currently it's very unstable, so now isn't quite the time...) i'm not sure when / if bienvenu plans to commit this to some branch you could use.
Comment 47•19 years ago
|
||
(In reply to comment #46) > john: you need a newer build, try trunk in a few weeks (currently it's very > unstable, so now isn't quite the time...) > > i'm not sure when / if bienvenu plans to commit this to some branch you could > use. Thanks for the suggestion, but the problem is rare enough that I'm not interested in running non-production code.
| Assignee | ||
Comment 48•19 years ago
|
||
both these fixes are in 1.5 - timeless must have misread the branches. The crash reported in 1.5 has a different stack trace and I don't know that there's a fix on the trunk for that or not (but I suspect not).
Comment 49•19 years ago
|
||
(In reply to comment #48) > both these fixes are in 1.5 - timeless must have misread the branches. The > crash reported in 1.5 has a different stack trace and I don't know that there's > a fix on the trunk for that or not (but I suspect not). Then we need to either get this bug reopened or a new bug entered!
| Assignee | ||
Comment 50•19 years ago
|
||
yes, a new bug would be appropriate.
Comment 51•19 years ago
|
||
(In reply to comment #50) > yes, a new bug would be appropriate. Done. Bug 325768.
Updated•13 years ago
|
Crash Signature: [@ 0x00000000 - nsMsgSearchSession::NotifyListenersDone]
You need to log in
before you can comment on or make changes to this bug.
Description
•