You can also try kicking off an full availability scan manually from the Agent prompt using the 'avail' command:
> help avail
Syntax: avail [--changed] [--verbose]
This will request an availability report that will tell you
which resources are up and which are down. By default, this will
show the availability of all inventoried resources. If --changed
was specified, only those resources whose availability status changed
from their last known state are shown. After displaying the contents
of the report, it will be sent up to the server.
-c, --changed : if specified, the report will contain availability
for only those resources whose status changed.
-v, --verbose : if true, additional information is displayed.
>
The plugin container should run avail scans periodically on its own. Nevertheless, when a resource is red I think should be green, that's usually the first thing I try, just in case there's some sort of glitch in the periodic scanning mechanism.
On 07/23/2010 12:02 PM, John Mazzitelli wrote:I would look at the agent's log file and see what kind of errors it might be logging. It might tell you why its failing to detect the server is back up. You can turn on debug on the agent (either via the debug agent prompt command or by restarting the agent with the env var "RHQ_AGENT_DEBUG" set to true). On 07/23/2010 12:01 PM, John Holland wrote:no, it's green initially, then when I try to trigger an event by shutting it down and starting it back up, it never goes green again. On 07/23/2010 11:54 AM, John Mazzitelli wrote:[diverting this thread to the rhq-users list, taking it off the devel list] Did it ever say it was green/up? If not, it sounds like you possibly have secured the remote JNP endpoint and the agent doesn't know about the credentials. Did you go to the JBossAS Server resource's Inventory tab, Connections subtab and confirm its all correct (including the JNP endpoint and credentials)? If that is wrong, the jboss plugin won't be able to connect to your JBossAs instance and will report it down. On 07/23/2010 11:47 AM, John Holland wrote:Sorry but I'm still not getting a reliable indication of the state of the server. Right now it's been up for 10 minutes, yet JON still says it is down. I've tried inventory discovery -f restarting the agent restarting the agent with --cleanconfig I can't seem to get it to tell JON the server is up!!!! I'm using JON 2.3.0 and rhq-agent 1.3.0.GA Any help would be great Thanks, John Holland On 07/23/2010 10:49 AM, John Mazzitelli wrote:If your alert is on the "Operation" success (and the operation is "restart" or "shutdown" - I think that's the op names) then yes. Alternatively, you can set alerts on "AVAILABILITY GOES UP" or "AVAILABILITY GOES DOWN". When the agent detects the JBossAS server has come up from being in a down state, the "Goes Up" alert is triggered. And when the agent detects the JBossAS server goes down (when it previously was seen as up), then the "Goes Down" alert is triggered. On 07/23/2010 10:37 AM, John Holland wrote:Hello, I have been experimenting with JON and RHQ and it seems to me that alerts set to go off when Jboss is restarted or shutdown only kick in when the action is triggered from JON. So if I do shutdown.sh -S on the box, the alert is not sent. Is that correct? Thanks, John Holland _______________________________________________ rhq-devel mailing list rhq-devel@lists.fedorahosted.org https://fedorahosted.org/mailman/listinfo/rhq-devel_______________________________________________ rhq-devel mailing list rhq-devel@lists.fedorahosted.org https://fedorahosted.org/mailman/listinfo/rhq-devel_______________________________________________ rhq-users mailing list rhq-users@lists.fedorahosted.org https://fedorahosted.org/mailman/listinfo/rhq-users
-- Ian Springer Sr. Software Developer JBoss Operations Network Red Hat ian.springer@redhat.com_______________________________________________ rhq-users mailing list rhq-users@lists.fedorahosted.org https://fedorahosted.org/mailman/listinfo/rhq-users