Project

General

Profile

Bug #60234

FBS crashing after RHEL7 update

Added by Dirk Ahrnke over 3 years ago. Updated almost 3 years ago.

Status:
Resolved
Priority:
Normal
Assignee:
QAGroup
Category:
Scalix Server
Target version:
Start date:
01/29/2016
Due date:
05/20/2016
% Done:

100%

Estimated time:
Operation System:
Rhel 7

Description

customer reported FBS-crashing on 12.5.0 after applying latest OS-updates.

SERIOUS ERROR                  Free Busy Serv(Service 115   ) 01.28.16 20:40:10
[OM 10272] BACKTRACE:
/opt/scalix/lib/libom_er.so(er_add_backtrace+0xa1)[0xf7698fb1]
/opt/scalix/lib/libom_er.so(+0x428a)[0xf769928a]
/opt/scalix/lib/libom_er.so(er_DumpProcAndExit+0x1d)[0xf769944d]
[0xf7731400]
[0xf7731430]
/usr/lib/libc.so.6(gsignal+0x47)[0xf7233667]
/usr/lib/libc.so.6(abort+0x143)[0xf7234ea3]
/usr/lib/libglib-2.0.so.0(+0x1630f)[0xf743830f]
/usr/lib/libglib-2.0.so.0(g_cond_wait_until+0xbb)[0xf74b076b]
/usr/lib/libglib-2.0.so.0(g_cond_timed_wait+0x73)[0xf743a293]
sxfbs[0x805693a]
/usr/lib/libglib-2.0.so.0(+0x6efaa)[0xf7490faa]
/usr/lib/libglib-2.0.so.0(+0x1801d)[0xf743a01d]
/usr/lib/libpthread.so.0(+0x6b2c)[0xf7563b2c]
/usr/lib/libc.so.6(clone+0x5e)[0xf72fc71e] 

according to the customer the items in question showed errors before the RHEL update, but FBS was not crashing.

errors:
3x Reason: List MAPI props extension length = 0:
129x Reason: Cannot find appointment start date in List MAPI properties


Files

Free Busy Server_crash.txt Free Busy Server_crash.txt 115 KB Richard Hall, 01/29/2016 06:28 AM
installedrpms.log installedrpms.log 14.5 KB Kevin Anderson, 04/07/2016 03:44 PM

History

#1

Updated by Richard Hall over 3 years ago

Hi Dirk,
Need some more info if we are to make any progress on this:
  1. Does FBS crash immediately on start up?
  2. If not, is there a particular Calendar item that causes the problem?
  3. Anything more in the logs that might identify a calendar item (or user)?

Thanks.

#2

Updated by Richard Hall over 3 years ago

  • Status changed from New to Answered
  • Assignee changed from ServerDevsGroup to Richard Hall

If a particular user's calendar is suspected of causing the problem can we get an sxmboxexp archive of that folder?
- Thx

#3

Updated by Richard Hall over 3 years ago

From email...

some informations regarding #60234

-------- Weitergeleitete Nachricht --------
Betreff: Re: Problem mit dem Free Busy Server nach dem RedHat Update
Datum: Fri, 29 Jan 2016 12:17:46 +0100
Von:
An: Dirk Ahrnke <>

Hallo,

>>  1. Does FBS crash immediately on start up?
No, see attached log.
>>  2. If not, is there a particular Calendar item that causes the problem?
Possible, the crash occurs every time after appointment 0001f785ed75c420
see attached log
>>  3. Anything more in the logs that might identify a calendar item (or
>> user)?
I have to check with the user if its ok to forward appointment
0001f785ed75c420.

The log contains three startups.
The first 2 were with updated RedHat 7.
The 3th was with the restored vm snapshot (original RedHat 7)
I noted that all appointments are from 2009 (130x) and 2010 (2x).
The appointments are only from 2 mail accounts.

Mit freundlichen Grüßen
...

#4

Updated by Richard Hall over 3 years ago

Can I see exactly what's installed.
Can you post the output of "rpm -qa"
- Thanks

#5

Updated by Kevin Anderson over 3 years ago

Hi Richard.

I have multiple servers exhibiting this error as well. I can easily provide you with full root SSH access to one of them belonging to one of our techs if that would help.

I've asked him for the rpm -qa output as well.

Thanks
Kev.

#6

Updated by Kevin Anderson over 3 years ago

Attached is the rpm -qa.

#7

Updated by Richard Hall over 3 years ago

  • Assignee changed from Richard Hall to ServerDevsGroup

Richard is no longer working at Scalix.
Assigning this bug report back to ServerDevsGroup.

#8

Updated by Danny T over 3 years ago

Sorry for the delay, I am working on a patch for sxfbs so that the customer can try out very soon.

#9

Updated by Danny T over 3 years ago

Please supply the following patch of /opt/scalix/bin/sxfbs for customer to try:
-back up the original version
-unzip this patch version
-replace original with patch, make sure permission correct
-restart scalix completely (omshut, omrc)

Please feedback to this ticket if the patch has worked or not.
Thanks,
Danny

#10

Updated by Danny T over 3 years ago

Had trouble uploading patch due to SSL issue, patch emailed to support to customer.
Initial signs are good, await further feedback from customer to confirm.

#11

Updated by Danny T over 3 years ago

  • Due date set to 05/20/2016
  • Status changed from Answered to Resolved
  • Assignee changed from ServerDevsGroup to Danny T
  • % Done changed from 0 to 100

Fixed in latest version of 12.6.0 and merged to trunk r14883:

Changed paths:
M /trunk/src/bin/fbs/logInterface.c
M /trunk/src/bin/fbs/searchInterface.c
M /trunk/src/bin/fbs/updatesInterface.c

merge r14859 from Bluenose (12.6.0):
Bug #60234 FBS crashing after RHEL7 update

#12

Updated by Danny T about 3 years ago

  • Assignee changed from Danny T to QAGroup
#14

Updated by Danny T almost 3 years ago

  • Target version set to Scalix 12.6

Also available in: Atom PDF