Scalix 11.3 on Centos 5.1 Wierd Crashes and instability

[sbrath]

I have migrated to a 10.x installs mailboxes via an Scalix Enterprise 11.3 install, that pulled in my old mailboxes.

My current environment is a P4 2.6G, 2G ram, Centos 5.1, Scalix Enterprise 11.3.0-I1-GA, Clients using Outlook Connector from 10.x

I'm having several problems.

First, users who while composing a message in Outlook, select "TO" to look up in the address book, get a screen full of garbage. Not sure why.

Second I have a large number of fatal errors in the fatal log, and I'm wondering if someone can make heads or tails of them:
--------------------

ERROR Local Delivery(Local Delivery) Thu Mar 6 08:53:40 2008
[OM 7720] An error has occurred while performing an AutoFile operation,
the file has been aborted.
User Name: Shane Brath / xxxxxx- DELETED ---
Pid of logging process: 29746
Current errno value: 2
Last Msg Subject: [SPAM] Software
Last Msg Sender: heqcskleincif/internet
Last Msg Id: 678213548.70551400284759(a)csklein.com


ERROR Local Delivery(Local Delivery) Thu Mar 6 08:53:40 2008
[OM 28870] To many serialised items open.
User Name: XXXX
Pid of logging process: 29746
Current errno value: 2
Last Msg Subject: [SPAM] Software
Last Msg Sender: heqcskleincif/internet
Last Msg Id: 678213548.70551400284759(a)csklein.com




SERIOUS ERROR CDA Server (CDA Server ) Thu Mar 6 08:54:38 2008
[OM 10270] Process about to terminate due to error.
Signal (Segmentation Violation) trapped by process 31822
Procedure trace follows:
<- uni_GetCharset
<- cvc_IsMappingAvail2
-> cvc_IsMappingAvail2
-> uni_GetCharset
-> cvc_iconvGetCharSetName
<- cvc_iconvGetCharSetName
-> cvc_iconvGetCharSetName
<- cvc_iconvGetCharSetName
-> cvc_IsMappingAvail3
-> cvc_iconvGetCharSetName
<- cvc_iconvGetCharSetName
-> cvc_iconvGetCharSetName
<- cvc_iconvGetCharSetName
<- cvc_IsMappingAvail3
<- uni_GetCharset
<- cvc_IsMappingAvail2
Pid of logging process: 31822


SERIOUS ERROR CDA Server (CDA Server ) Thu Mar 6 08:54:38 2008
[OM 10272] BACKTRACE:
/opt/scalix/lib/libom_er.so(er_add_backtrace+0xb3)[0x123b03]
/opt/scalix/lib/libom_er.so[0x123df3]
/opt/scalix/lib/libom_er.so(er_DumpProcAndExit+0x1f)[0x123f7f]
[0x14d420]
/lib/libc.so.6(__strtol_internal+0x3f)[0x328d8f]
/lib/libc.so.6(atoi+0x29)[0x3263b9]
cda.server[0x804db48]
/opt/scalix/lib/libom_cl.so(cl_CallProc+0x86)[0x114d4a]
cda.server[0x8051afa]
/lib/libc.so.6(__libc_start_main+0xdc)[0x311dec]
cda.server[0x804a411]
Pid of logging process: 31822




Any help on these would be great, otherwise I'll be calling in a support call.

This is causing me concern about Scalix 11.3 as my 10.x install was quite stable. Now I'm seeing issues like email just stops being delivered, and is lost?

Shane.

[kanderson]

Hi Shane.

This will be a problem with one of the auto actions (rules) created for your user.

Can you use:

sxaa --user "shane brath"

to look at the rules that are in place. You can use sxaa alone on a line to find more details about the command itself.

What you'll see is a list of the rules you have in place, showing a name and a action number. For more details on any of those rules, just use the same command and follow it with --info #### where the #### is the rule number.

Your rule says it's an autofile action, so it will likely be something like moving spam into a spam folder that doesn't exist, or something similar.

For the garbage you're seeing in the TO field, Can you please post more details about that and we'll work on that next.

Thanks
Kev.

[sbrath]

I will look into the Auto File, that makes sense.

What about the CDA server not staying running? Every time I try to start it I get the crash as listed in the post. This is not good.

DO you have any sugestions on fixing that issue?

[kanderson]

A "stuck" CDA usually happens after a power failure or similar outage. The fix is fairly easy.

Delete /var/opt/scalix/??/s/sys/omcda.lck

Then

Restart it. omon cda

It should be fine after that.

Kev.

[sbrath]

That worked for the CDA server.
Maybe as a feature request:
Put in the log file " Oops, please delete the old lock file" instead of
"
SERIOUS ERROR CDA Server (CDA Server ) Sun Mar 9 20:42:21 2008
[OM 10272] BACKTRACE:
/opt/scalix/lib/libom_er.so(er_add_backtrace+0xb3)[0xe07b03]
/opt/scalix/lib/libom_er.so[0xe07df3]
/opt/scalix/lib/libom_er.so(er_DumpProcAndExit+0x1f)[0xe07f7f]
[0xb5e420]
/lib/libc.so.6(__strtol_internal+0x3f)[0x313d8f]
/lib/libc.so.6(atoi+0x29)[0x3113b9]
cda.server[0x804db48]
/opt/scalix/lib/libom_cl.so(cl_CallProc+0x86)[0x294d4a]
cda.server[0x8051afa]
/lib/libc.so.6(__libc_start_main+0xdc)[0x2fcdec]
cda.server[0x804a411]
Pid of logging process: 30964

"

I will have more on the TO: Corruption thing this weekend.