[Pulp-list] Pulp 2.8.0 stops running on CentOS 6.7

[Sparks, Alan]

The only thing I've found in the syslogs, before all goes south is:
Apr  1 18:28:19 oa-ftc-repo0001 pulp: pulp.server.async.scheduler:ERROR: Worker 'reserved_resource_worker-8 at oa-ftc-repo0001.oa.ftc.hpelabs.net' has gone missing, removing from list of workers
Apr  1 18:28:19 oa-ftc-repo0001 pulp: pulp.server.async.tasks:ERROR: The worker named reserved_resource_worker-8 at oa-ftc-repo0001.oa.ftc.hpelabs.net is missing. Canceling the tasks in its queue.
Apr  1 18:28:19 oa-ftc-repo0001 pulp: pulp.server.async.scheduler:ERROR: Worker 'reserved_resource_worker-6 at oa-ftc-repo0001.oa.ftc.hpelabs.net' has gone missing, removing from list of workers
Apr  1 18:28:19 oa-ftc-repo0001 pulp: pulp.server.async.tasks:ERROR: The worker named reserved_resource_worker-6 at oa-ftc-repo0001.oa.ftc.hpelabs.net is missing. Canceling the tasks in its queue.
Apr  1 18:28:19 oa-ftc-repo0001 pulp: pulp.server.async.scheduler:ERROR: Worker 'reserved_resource_worker-4 at oa-ftc-repo0001.oa.ftc.hpelabs.net' has gone missing, removing from list of workers
Apr  1 18:28:19 oa-ftc-repo0001 pulp: pulp.server.async.tasks:ERROR: The worker named reserved_resource_worker-4 at oa-ftc-repo0001.oa.ftc.hpelabs.net is missing. Canceling the tasks in its queue.
Apr  1 18:28:19 oa-ftc-repo0001 pulp: pulp.server.async.scheduler:ERROR: Worker 'reserved_resource_worker-5 at oa-ftc-repo0001.oa.ftc.hpelabs.net' has gone missing, removing from list of workers

No idea what is happening at that time, that would cause all the processes to "disconnect".   They seem to still be running... qpid or mongo going bad somehow?  Just vanilla stock installs of those.
-Alan


From: pulp-list-bounces at redhat.com [mailto:pulp-list-bounces at redhat.com] On Behalf Of Sparks, Alan
Sent: Tuesday, April 05, 2016 10:19 AM
To: pulp-list at redhat.com
Subject: [Pulp-list] Pulp 2.8.0 stops running on CentOS 6.7

I have a 2.8.0 install running on a patched-up CentOS 6.7 machine, running all from the pulp-stable distribution.   QPID as MQ, and Mongo server mongodb-server-2.4.14-1.el6.x86_64.

I can't seem to keep it running more than a week before it falls over, tasks stop running, and following repeated in the syslog:
Apr  5 16:09:22 oa-ftc-repo0001 pulp: pulp.server.async.scheduler:ERROR: There are 0 pulp_resource_manager processes running. Pulp will not operate correctly without at least one pulp_resource_mananger process running.
Apr  5 16:09:22 oa-ftc-repo0001 pulp: pulp.server.async.scheduler:ERROR: There are 0 pulp_celerybeat processes running. Pulp will not operate correctly without at least one pulp_celerybeat process running.


I do see this running in process list:
10606 ?        Sl     7:57 /usr/bin/python /usr/bin/celery beat --app=pulp.server.async.celery_instance.celery --scheduler=pulp.server.async.scheduler.Scheduler --workdir=/var/run/pulp/ -f /var/log/pulp/celerybeat.log -l INFO --detach --pidfile=/var/run/pulp/celerybeat.pid

If I attempt to stop celerybeat:
# service pulp_celerybeat stop
celery init v10.0.
Using configuration: /etc/default/pulp_workers, /etc/default/pulp_celerybeat
Stopping pulp_celerybeat... ERROR
Timed out while stopping (30s)

I'm not sure how to determine what it dying... If I hard stop and start everything, or if I reboot, I can get out of the issue for a few days, before it recurs.
Does anyone have advice on what to look for?   The Pulp logs basically says everything is logging to syslog, but have not found a smoking gun to indicate what fell over.
-Alan

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/pulp-list/attachments/20160405/f335c38b/attachment.htm>