LDAP Daemon Aborts

[patrickmoore]

Hi all

I am running 11.0.1 on Centos 4.3, and have been running without any major problems since my version 10 install late last year. (minor problems relating to outlook connector 11, but that's another story). I am collecting mails from a few POP accounts using fetchmail. My problem: twice in the last month, mails stop coming in, and the cause appears to be that the LDAP Daemon has aborted, as when I restart LDAP, the mails come in no problem. Has anyone an idea as to why the LDAP Daemon might abort ?

See extract from Service Event Log below:

SERIOUS ERROR LDAP Daemon (LDAP Listener ) 02.13.07 14:08:20
[OM 10270] Process about to terminate due to error.
Signal (Segmentation Violation) trapped by process 10411
Procedure trace follows:


SERIOUS ERROR LDAP Daemon (LDAP Listener ) 02.13.07 14:08:20
[OM 10272] BACKTRACE:
/opt/scalix/lib/libom_er.so(er_add_backtrace+0xc6)[0xd8fee6]
/opt/scalix/lib/libom_er.so[0xd901e6]
/opt/scalix/lib/libom_er.so(er_DumpProcAndExit+0x1f)[0xd9038f]
/lib/tls/libpthread.so.0[0x4d2888]
/lib/tls/libpthread.so.0[0x4cc371]
/lib/tls/libc.so.6(__clone+0x5e)[0x4159be]

[dkelly]

Do you use the Scalix Management Console at all ? There have been times, which we are unable to reproduce reliably, where trying to sign on to the console using an internet address causes the LDAP service to abort.

Is this the case for you ?

Cheers

Dave

[patrickmoore]

Hi Dave

Don't think that's the case here, I did not sign into SAC on the day in question. Could it be Webmail related ? I signed in to another user's account via Webmail, and they were logged in to their Outlook at the same time.

Rgds
Patrick

[heupink]

Hi!

Any progression on this? I saw it as well, this morning:

Code: Select all

SERIOUS ERROR           LDAP Daemon   (LDAP Listener ) Tue Feb 27 10:17:57 2007
[OM 10270] Process about to terminate due to error.
Signal (Segmentation Violation) trapped by process 11321
Procedure trace follows:
Pid of logging process: 11321


SERIOUS ERROR           LDAP Daemon   (LDAP Listener ) Tue Feb 27 10:17:58 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]
/lib/tls/libpthread.so.0[0x400abcf7]
/lib/tls/libc.so.6(__clone+0x5e)[0x4017121e]
Pid of logging process: 11321

I did not try to login sac at that time, I always login sac with my username ONLY, excluding @domain.com)

[eyalm]

I got the same error today.
LDAP Daemon Aborted.

SERIOUS ERROR LDAP Daemon (LDAP Listener ) Mon Mar 5 14:57:17 2007
[OM 10270] Process about to terminate due to error.
Signal (Segmentation Violation) trapped by process 11133
Procedure trace follows:
Pid of logging process: 11133


SERIOUS ERROR LDAP Daemon (LDAP Listener ) Mon Mar 5 14:57:17 2007
[OM 10272] BACKTRACE:
/opt/scalix/lib/libom_er.so(er_add_backtrace+0xc6)[0xf7ff5ee6]
/opt/scalix/lib/libom_er.so[0xf7ff61e6]
/opt/scalix/lib/libom_er.so(er_DumpProcAndExit+0x1f)[0xf7ff638f]
[0xffffe500]
/lib/tls/libpthread.so.0[0xb20371]
/lib/tls/libc.so.6(__clone+0x5e)[0xa799be]
Pid of logging process: 11133

Scalix 11.0.1 RHEL 4

[heupink]

Meanwhile I have upgraded to scalix 11.0.2 on sles9, sp3, fully patched, and I still see the error:

Code: Select all

SERIOUS ERROR           LDAP Daemon   (LDAP Listener ) Sun Mar  4 07:50:06 2007
[OM 10270] Process about to terminate due to error.
Signal (Segmentation Violation) trapped by process 6465
Procedure trace follows:
Pid of logging process: 6465


SERIOUS ERROR           LDAP Daemon   (LDAP Listener ) Sun Mar  4 07:50:06 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]
/lib/tls/libpthread.so.0[0x400abcf7]
/lib/tls/libc.so.6(__clone+0x5e)[0x4017121e]
Pid of logging process: 6465

I also noticed that outgoing mail is no longer delivered when the ldap interface has crashed, so this problem is more serious than I actually thought:

Code: Select all

Mar  4 07:57:20 intech007 sendmail[15285]: l246vKKR015285: SYSERR(root): Error getting LDAP results in map ldapsx: Unknown error 325
Mar  4 08:02:37 intech007 sendmail[15323]: l2472bsq015323: SYSERR(root): Error getting LDAP results in map ldapsx: Unknown error 325
Mar  4 08:03:20 intech007 sendmail[15324]: l2473Kfe015324: SYSERR(root): Error getting LDAP results in map ldapsx: Unknown error 325
Mar  4 08:08:37 intech007 sendmail[15328]: l2478b3b015328: SYSERR(root): Error getting LDAP results in map ldapsx: Unknown error 325


Is anyone from scalix looking into this issue, or does anyone here have an idea?

[eyalm]

Happend again last night.
Also, ldapmapper died, and I was getting the same error:

SYSERR(root): Error getting LDAP results in map ldapsx: Unknown error 325

[kanderson]

Sounds like (for now), the ldap mapper isn't running.

/etc/init.d/ldapmapper restart

The crashing smtpd daemon was a bug in 11.0.0GA when running on SLES9. It was resolved in 11.0.1.

You MAY have some messages remaining in your error queue. You can check with "omstat -q error" and resubmit them with "omresub -q error". This can also be done via the management console.

[eyalm]

I restarted ldapmapper and it's working fine now.
I just want to know why it died but I can't find anything but that LDAP Daemon error.

[kanderson]

I have not seen problems with the ldapmapper. I would assume it will be a one time fluke.

In answer to your curiosity about why something happened, I believe it logs to the sendmail logs. /var/log/mail*

In terms of the Scalix LDAP service crashing, you might find something in /var/opt/scalix/??/s/logs/fatal. Or omshowlog -s ldap (to show logs from the LDAP service only).

Kev.

[pgsousa]

Hi,

Any update on this? I'm having the same problem with Scalix 11.0.2 and Centos 4.4:

SERIOUS ERROR LDAP Daemon (LDAP Listener ) 03.12.07 07:22:39
[OM 10270] Process about to terminate due to error.
Signal (Segmentation Violation) trapped by process 17704
Procedure trace follows:


SERIOUS ERROR LDAP Daemon (LDAP Listener ) 03.12.07 07:22:39
[OM 10272] BACKTRACE:
/opt/scalix/lib/libom_er.so(er_add_backtrace+0xc6)[0xc06ee6]
/opt/scalix/lib/libom_er.so[0xc071e6]
/opt/scalix/lib/libom_er.so(er_DumpProcAndExit+0x1f)[0xc0738f]
/lib/tls/libpthread.so.0[0x138898]
/lib/tls/libpthread.so.0[0x132371]
/lib/tls/libc.so.6(__clone+0x5e)[0x4ddffe]

Pgsousa

[patrickmoore]

I am also still having this problem... can anyone shed light on this ?

[Alik]

We are running Scalix 11.02 on SLES9 and have had this problem twice, today and yesterday.

After the update to the version 11 we can not suggest to anybody to install the version 11. My opinion is, that Scalix is not really stable enough for the enterprise use.

Version 11 brought some nice and welcome features, but since we updated to this version we were spending all our time to make some workarounds, tune our monitoring system and others to be sure Scalix 11 is running (I am not talking about all the Outlook crashes). The first think I do in the morning is checking Scalix running and sending and receiving mails.

Now I know, that I never will run any software with a x.0 Release.

Sorry to the guys from Scalix, but I am just really tired from all the problems with scalix.

I can not tell my CIO that the ldap deamon is crashed again and I have to restart it and the next bug fix may should solve this problem. My boss did not care about stuff like this, he just can not send e-mails and realize, that the Mail server is not running again and he is thinking what do they do in the IT department.

;-(

[heupink]

Alik, I really agree with you, I regret having installed version 11 as well. We had SO many issues. (some of them were resolved through scalix support, others not)

Anyway, we are also seeing the ldap crashes. What I did now:
I have created a cron job to start the ldap deamon every half hour.
If it happens to be running already, this only gives an error message.

But again: I agree: i will not recommend scalix 11 the way it is now to anyone. You really need to have a sys admin on site to continuously be there to monitor and correct issues, whereas before (scalix 9 and 10) it would run rock solid.

Some of the issues we saw were specific to sles9. Maybe sles9 is not so well supported as other linuxes..?

[heupink]

patrickmoore and Alik: Are you still seeing these problems?

Are you both running sles9?

I'm talking with scalix support, and they thought these issues were related to:
http://bugzilla.scalix.com/show_bug.cgi?id=14709

But I doubt this. I am running latest scalix, WITH a hotfix for bug 14709, and I had three or four ldap crashes "Error getting LDAP results in map ldapsx: Unknown error 325" over the last few weeks.

So, what's the situation on your machines?