OD is requires frequent reboot

OpenDeploy is frequently (sometimes twice a day), crashing, and requiring a reboot. What could be causing it to crash? Is the common for OpenDeploy?

Additional info:
The deployments are running DNR scripts that execute datadeploy. If the script fails, I send back a <response code='-2' />. This is working correctly as far as I can tell.

Find more posts tagged with

Comments

Migrateduser

Which version of OpenDeploy, which service packs and patches applied, which platform?

Todd Scallan
Group Product Manager
Interwoven
t: 408-530-7167
e: tscallan@interwoven.com

MoReese

Platform: Solaris
Version: 5.5.1 Server
Service Pack 2 is installed as well.

Thanks for your help.

WebTop.PNG

Adam Stoller

To try to get a better idea of the circumstances...

Are you kickiing off a lot of deployments from workflows, using the '-inst' parameter?

Are you kicking off a lot of deployments from the OpenDeploy Scheduler?

Is it the Base Server which needs to be restarted or the Receiver(s)?

Can you determine any kind of pattern to when it needs to be restarted (after approximately N deployments? At certain times of the day?, etc.)?

--fish
(Interwoven Senior Technical Consultant)

MoReese

Deployments are all being called via workflows. The workflows are calling perl scripts, and the perl scripts fire off iwodstart <deployment name>. Most deployments are filelist deployments that also call a datadeploy perl script, which runs the iwdd.ipl cmd.

Originally it seemed to be the base server that always needed rebooted, but now, a rcvr is needing rebooted frequently.

I haven't been able to determine a pattern yet. One theory was that it was failing when my perl script was returning "<response code='-2' />". But this seems to be working most of the time.

MoReese

Anther note: We are using 40 bit encryption.

Migrateduser

No sure if this is your problem but there was a known problem with TeamSite workflows consuming file descriptors over time when kicking off external tasks. This would often show itself when the iwodstart process (not the actual Base Server) would crash.

This problem has been discussed in a previous thread, and is supposed to have been fixed in TeamSite 5.5.2 SP3.

Which version of TeamSite and SP do you have?

Todd Scallan
Group Product Manager
Interwoven
t: 408-530-7167
e: tscallan@interwoven.com

nipper

Are we certain this was fixed in SP3 ? What level of OD is required ?

This bug has bitten us a bunch of times.

Andy

Migrateduser

The bug that is believed to be the root cause is reported as fixed in the Fixed Bugs list for TS 5.5.2 SP3:

11839 iwserver runs out of file descriptors with rogue iwat script

OD 5.6 and OD 5.5.1 (with SP2 recommended) are supported with TS 5.5.2.

Todd Scallan
Group Product Manager
Interwoven
t: 408-530-7167
e: tscallan@interwoven.com