[Spacewalk-list] Scheduled actions slow to be acted upon

[Mullis, Josh (CCI - Atlanta)]

Been having an issue where if the spacewalk server is rebooted or a
"rhn-satellite restart" is issued, scheduled actions DO NOT happen until
an rhn_check is run.

By rhn_check I mean either manually running rhn_check, or restarting the
osad daemon on the client (which essentially does an rhn_check).
After restarting osad, if I immediately schedule another action, it
still does not get pushed to client unless another osad restart or
rhn_check is done.




If I do the following steps, my scheduled actions are taken care of
immediately....

-------------------
1.  rm -Rf /var/lib/jabberd/db/*
2.  service jabberd restart
3.  service osa-dispatcher restart
4.  service osad restart  (On spacewalk clients)
----------------------



I have tried just restarting jabberd and osa-dispatcher on the server,
and then osad on clients, it makes no difference.
However, if I re-do the above steps, it starts to handle actions very
quickly again.



>From what I understand, the above commands re-build the jabber auth
database.

So why after a database rebuild does it work properly, but breaks after
a reboot or "rhn-satellite restart"???
What is changing after a reboot or "rhn-satellite restart" ?

I still see clients successfully logging into jabber via the logs.


Using spacewalk v0.6 on RHEL5


Thanks for reading through this!
-Josh
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/spacewalk-list/attachments/20100115/54301b69/attachment.htm>