Availability of the platform is tied to the availability of the RHQ
Agent. If the agent is not running then the platform will be down (and
all of the resources under it will be set to unknown for that period).
The agent, when running, pings the server frequently with a heartbeat,
if this ping is not received for several minutes then the agent, even if
running, will be considered down by the server, and again availability
will go down. That shouldn't happen unless communication with the
server is interrupted due to connectivity issues, there is perhaps
extreme load on the agent, or for some reason the agent's plugin
container is not actually running (in which case the agent process is
running, but no actual monitoring is taking place. This typically only
happens during a plugin update and does not stay that way very long).
That's all I can think of, you can always take a look at the agent log
to see if there is anything in there that jumps out at you.
On 9/17/2013 11:20 AM, Attila Heidrich wrote:
Thanks a lot for the very quick answers I got recently!
I still have a single problem. Why is it so, that a Platform (the only
Windows XP we have got) is reported to be available only some 60% of
the time, while all the measurements on the same machine are OK?
Also the accessibility of the services on this Platform (high level
network accessibility tested by port services from outside the virtual
machine which runs the Platform) are OK, but the availability of the
Platform itself isn't!
Regards,
Attila
_______________________________________________
rhq-users mailing list
rhq-users(a)lists.fedorahosted.org
https://lists.fedorahosted.org/mailman/listinfo/rhq-users