589123 - thunderbird trunk will completely freeze up within 4 to 6 minutes from starting up

Status: RESOLVED WORKSFORME

(Thunderbird :: General, defect)

Description

[Karanbir Singh]

User-Agent:       Mozilla/5.0 (X11; Linux x86_64; rv:2.0b5pre) Gecko/20100818 Minefield/4.0b5pre
Build Identifier: Mozilla/5.0 (X11; Linux i686; rv:2.0b5pre) Gecko/20100818 Shredder/3.2a1pre

I've been running thunderbird nightly / shredder builds, updating with every build release. From 20100818 I've started seeing that the entire app will freeze after a few minutes and needs a sigkill to terminate the app.

I've got 5 IMAP accounts setup, gloda search is disabled. The remote mail servers are faily close in network terms and available on a 100mbps lan connection so I don't think its a network latency issue.


Reproducible: Always

Actual Results:  
strace'ing thunderbird results in something like this towards the end. Note
that once it hits the FUTEX_WAIT_PRIVATE, the app no longer responds to anything and must be killed to terminate ( in this case, I left it in the frozen state for 12 minutes ).




gettimeofday({1282300218, 287388}, NULL) = 0

read(15, "\372", 1)                     = 1

fstat64(78, {st_mode=S_IFREG|0664, st_size=41661, ...}) = 0

_llseek(78, 40960, [40960], SEEK_SET)   = 0

read(78, "}2F}@\n\n@$${31{@\n< <(a=c)> // (f="..., 701) = 701

_llseek(78, 40960, [40960], SEEK_SET)   = 0

read(78, "}2F}@\n\n@$${31{@\n< <(a=c)> // (f="..., 4096) = 701

fstat64(78, {st_mode=S_IFREG|0664, st_size=41661, ...}) = 0

_llseek(78, 40960, [40960], SEEK_SET)   = 0

read(78, "}2F}@\n\n@$${31{@\n< <(a=c)> // (f="..., 4096) = 701

_llseek(78, 40960, [40960], SEEK_SET)   = 0

read(78, "}2F}@\n\n@$${31{@\n< <(a=c)> // (f="..., 4096) = 701

_llseek(26, 0, [1348450], SEEK_CUR)     = 0

fstat64(26, {st_mode=S_IFREG|0664, st_size=1348450, ...}) = 0

_llseek(26, 1347584, [1347584], SEEK_SET) = 0

read(26, "5B4F{@\n@$$}5B4F}@\n\n@$${5B50{@\n@$"..., 866) = 866

_llseek(26, 1347584, [1347584], SEEK_SET) = 0

read(26, "5B4F{@\n@$$}5B4F}@\n\n@$${5B50{@\n@$"..., 4096) = 866

fstat64(26, {st_mode=S_IFREG|0664, st_size=1348450, ...}) = 0

_llseek(26, 1347584, [1347584], SEEK_SET) = 0

read(26, "5B4F{@\n@$$}5B4F}@\n\n@$${5B50{@\n@$"..., 4096) = 866

_llseek(26, 1347584, [1347584], SEEK_SET) = 0

read(26, "5B4F{@\n@$$}5B4F}@\n\n@$${5B50{@\n@$"..., 4096) = 866

fstat64(26, {st_mode=S_IFREG|0664, st_size=1348450, ...}) = 0

_llseek(26, 1347584, [1347584], SEEK_SET) = 0

read(26, "5B4F{@\n@$$}5B4F}@\n\n@$${5B50{@\n@$"..., 4096) = 866

_llseek(26, 1347584, [1347584], SEEK_SET) = 0

read(26, "5B4F{@\n@$$}5B4F}@\n\n@$${5B50{@\n@$"..., 4096) = 866

_llseek(26, 1347584, [1347584], SEEK_SET) = 0

read(26, "5B4F{@\n@$$}5B4F}@\n\n@$${5B50{@\n@$"..., 4096) = 866

write(26, "\n@$${5B6D{@\n<(2FC85E=4c6e577f)>["..., 63) = 63

close(78)                               = 0

munmap(0xb7e12000, 4096)                = 0

gettimeofday({1282300218, 289508}, NULL) = 0

gettimeofday({1282300218, 289598}, NULL) = 0

recv(52, "\27\3\1\0 ", 5, 0)            = 5

recv(52, "\266\"\3323\347\332\316\225A\213\6Ij\230\1772AdDUd*\230\245\30\365\372\272\355~\363:", 32, 0) = 32

futex(0xa1a1ebe4, FUTEX_WAIT_PRIVATE, 2, NULL <unfinished ...>

+++ killed by SIGKILL +++

Comment 1

[Karanbir Singh]

The only addon I have installed is sieve-0.1.10 and the British English dictionary

Comment 2

[Wayne Mery (:wsmwk)]

does it reproduce in safe mode?
if you are in unified view, does it happen in "All Folders" view?

Comment 3

[Karanbir Singh]

(In reply to comment #2)
> does it reproduce in safe mode?

yes

> if you are in unified view, does it happen in "All Folders" view?

I can reproduce this in the unified view, the all-folders view and the unread folders view

Comment 4

[Karanbir Singh]

Build: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.2.9pre) Gecko/20100824 Lanikai/3.1.3pre does not have this issue

Comment 5

[Wayne Mery (:wsmwk)]

Karanbir, so it fails in 20100818 but works in 20100817? 

The approximate list of comm-central checkins is http://hg.mozilla.org/comm-central/pushloghtml?startdate=2010-08-17+03%3A00%3A00&enddate=2010-08-18+06%3A00%3A00

Comment 6

[Karanbir Singh]

I've now gone back to running Shredder, and no longer see the issue in atleast the last 2 weeks of builds. Currently running :

Mozilla/5.0 (X11; Linux i686; rv:2.0b8pre) Gecko/20101014 Thunderbird/3.3a1pre

Comment 7

[Wayne Mery (:wsmwk)]

hmm, that's odd. 

no hang bugs got fixed during this period - https://bugzilla.mozilla.org/buglist.cgi?type1-0-0=substring&keywords=hang&keywords_type=allwords&field0-0-0=short_desc&type0-0-1=substring&field0-0-1=keywords&type1-0-1=allwordssubstr&resolution=---&resolution=FIXED&classification=Client%20Software&classification=Components&chfieldto=2010-10-01&query_format=advanced&chfieldfrom=2010-08-24&type0-0-0=anywordssubstr&field1-0-0=short_desc&product=MailNews%20Core&product=Thunderbird&field1-0-1=short_desc

And I don't see perf bugs fixed in thunderbird components since 2010-08-24 that would explain the problem going away.  And the list of core bugs fixed is a bit obscure https://bugzilla.mozilla.org/buglist.cgi?type1-0-0=substring&field0-0-0=short_desc&bug_severity=major&bug_severity=normal&bug_severity=minor&type0-0-1=substring&field0-0-1=keywords&type1-0-1=allwordssubstr&resolution=FIXED&classification=Client%20Software&classification=Components&chfieldto=2010-10-01&chfield=resolution&query_format=advanced&chfieldfrom=2010-08-24&chfieldvalue=fixed&value0-0-1=perf&type0-0-0=anywordssubstr&value0-0-0=slow&field1-0-0=short_desc&product=Core&field1-0-1=short_desc

Nothing in core bugs stands out either https://bugzilla.mozilla.org/buglist.cgi?type1-0-0=substring&keywords=hang&keywords_type=allwords&field0-0-0=short_desc&type0-0-1=substring&field0-0-1=keywords&type1-0-1=allwordssubstr&resolution=---&resolution=FIXED&classification=Client%20Software&classification=Components&chfieldto=2010-10-01&query_format=advanced&chfieldfrom=2010-08-24&bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&bug_status=RESOLVED&bug_status=VERIFIED&bug_status=CLOSED&type0-0-0=anywordssubstr&field1-0-0=short_desc&product=Core&field1-0-1=short_desc

Comment 8

[Wayne Mery (:wsmwk)]

WFM per comment 6