The Berkley stuff was so fragile in out environment, was almost walked away from Spacewalk. Switching to SQLite fixed those.
Sent from my iPhone
spacewalk-list-bounces redhat com
Am 30. November 2017 21:15:22 MEZ schrieb "oogiejonz yahoo com
" <oogiejonz yahoo com
I am looking at using spacewalk for orchestration - for instance, I
want to be able to upgrade an app on a group of servers every month
without any downtime. To do this, I would need to ensure that the app
installs successfully on every server within a specific time frame. I
have osad running on my servers so that in theory as soon as I send a
remote command from spacewalk it should run on the server, but keep
running into servers where commands don't run until I manually login
and run rhn_check. So I'm thinking if I have rhn_check run from cron
every 15 minutes then the remote commands will always be picked up. Is
there any downside to running rhn_check this frequently?
Please try to stop osad on the problem systems, remove the /etc/sysconfig/rhn/osad-auth file and restart osad. This might fix your problem that jobs are not picked up instantly.
Alternatively, see if switching jabberd on your Spacewalk server from BerkeleyDB to SQLite makes it more stable. Instructions are here:
Avi Miller | Product Management Director | +61 (3) 8616 3496
Oracle Linux and Virtualization
417 St Kilda Road, Melbourne, Victoria 3004 Australia