SERIOUS ERROR Converters (RTF to Text )

[sejek]

Hello,
Some weeks after upgrading to 11.0.1 (Enterprise Edition) everything has started to collapse every 3-5 days. The first symptom is that SWA complains about eval license has expired which is not true as it is a real license who is not expired.
All om-commands including omshut returns promt. Strace reveals omcommands complains about corrupted shared memory.
In the fatal logs the first sign of selfdestruction is:
SERIOUS ERROR Converters (RTF to Text ) Mon Feb 26 16:57:30 2007
[OM 10270] Process about to terminate due to error.
Signal (Segmentation Violation) trapped by process 14214
Procedure trace follows:
<- cvc_enhCnvString
-> cvc_enhCnvString
<- cvc_CnvStringTryIconv
<- cvc_enhCnvString
-> cvc_enhCnvString
<- cvc_CnvStringTryIconv
<- cvc_enhCnvString
-> cvc_enhCnvString
<- cvc_CnvStringTryIconv
<- cvc_enhCnvString
-> cvc_enhCnvString
<- cvc_CnvStringTryIconv
<- cvc_enhCnvString
-> cvc_enhCnvString
<- cvc_CnvStringTryIconv
<- cvc_enhCnvString
Pid of logging process: 14214


SERIOUS ERROR Converters (RTF to Text ) Mon Feb 26 16:57:30 2007
[OM 10272] BACKTRACE:
/opt/scalix/lib/libom_er.so(er_add_backtrace+0xc6)[0x4001c366]
/opt/scalix/lib/libom_er.so[0x4001c665]
/opt/scalix/lib/libom_er.so(er_DumpProcAndExit+0x1f)[0x4001c7ef]
[0xffffe420]
/opt/scalix/lib/libom_rtfl.so(rtfl_BuildLine+0x127a)[0x4003f90f]
/opt/scalix/lib/libom_rtfl.so[0x40040cdc]
/opt/scalix/lib/libom_rtfl.so(rtfl_Parse+0x280)[0x40042729]
/opt/scalix/lib/libom_rtfl.so(rtfl_convert+0xff)[0x4003fdcb]
/opt/scalix/bin/rtf.browse[0x8048ccb]
/lib/tls/libc.so.6(__libc_start_main+0xd0)[0x40082210]
/opt/scalix/bin/rtf.browse[0x8048891]
Pid of logging process: 14214

SERIOUS ERROR PC Monitor (Socket Monitor) Mon Feb 26 16:58:15 2007
[OM 10270] Process about to terminate due to error.
Signal (Segmentation Violation) trapped by process 14316
Procedure trace follows:
-> uakd:negotiateParameters
<- uakd:negotiateParameters
-> uakd:readLogicalBlock
<- uakd:readLogicalBlock
-> uakd:startUalRemote
-> uakd_lookupUser
-> ul_OpenUL
-> dr_ACISetDefaultContext
-> dr_ACIModContextFlags
<- dr_ACIModContextFlags
<- dr_ACISetDefaultContext
<- ul_OpenUL
-> uald_FindPUorAlias
-> ul_FindAuthId
-> dr_ACIModContextFlags
<- dr_ACIModContextFlags
Pid of logging process: 14316

The only way to startup Scalix again is to kill all processes manually and run omrc.
Platform is SLES9 patchlevel3 with 2.6.5-7.257-smp
My feeling is that some kind of attachment is the root cause, but only guessing.
Any hints ?

[kanderson]

You aren't alone. Worse, It won't be fixed in 11.0.2

http://bugzilla.scalix.com/show_bug.cgi?id=14709

So far this bug remains open. As the bug only exists on SLES9, you could change platforms. That's obviously an ugly solution for many users, especially in a large install. Support would likely benefit from your assistance on this, as it seems to be a hard bug to replicate. Some sites have it, some don't. There are instructions in bugzilla showing how to turn on logging. If you are able to generate a log for them, send it to support@sclaix.com and with that bug ID as the subject, or better yet, dump it straight into bugzilla.

Thanks
Kev.

[sejek]

You aren't alone. Worse, It won't be fixed in 11.0.2

http://bugzilla.scalix.com/show_bug.cgi?id=14709

So far this bug remains open. As the bug only exists on SLES9, you could change platforms. That's obviously an ugly solution for many users, especially in a large install. Support would likely benefit from your assistance on this, as it seems to be a hard bug to replicate. Some sites have it, some don't. There are instructions in bugzilla showing how to turn on logging. If you are able to generate a log for them, send it to support@sclaix.com and with that bug ID as the subject, or better yet, dump it straight into bugzilla.

Thanks
Kev.

Thanx for the hint, have enabled the glibc trap
//Johan

[fkienker]

The same problem exists with Centos 4.4 (all current updates) and Scalix. 11.0.2.1. Our current work-around is to restart Scalix during Daily Maintenance.

Apparently there is a hotfix. Does anyone know when this will make it into the Community Edition?

[chris]

The same problem exists with Centos 4.4 (all current updates) and Scalix. 11.0.2.1. Our current work-around is to restart Scalix during Daily Maintenance.

Apparently there is a hotfix. Does anyone know when this will make it into the Community Edition?

The fix, which was just checked in today, will be in 11.0.3 for community users.

Please contact your Scalix representative regarding the hotfix.

Chris

[enneris]

I have the same problem (CentOS 4.4, 11.0.2.1). Worse, the server is not usable at all, errors show up on the first connection attempt. Is there a workaround untill the fix is avaiable?

Thank you

Edit : French userbase, lots of non-ASCII characters involved.

[chris]

CentOS is obviously not a supported operating system.

I suspect, however, that you're seeing a different error if it's happening at the first connection attempt.

Have you read the bug?

[swordfish]

When is 11.0.3 due to be released?

[chris]

The current target is early April - stay tuned to the Announcements forum for details.