how to setup alert only after the issue last more than certain time period

[erichuang82412668]

e.g. the high CPU last for more than half hour, then  trigger the alert.  this remind me another question, how the count works,  how long it will add up 1 count.

[Kathy Maguire]

Hi -

Alarm publication after a metric exceeds configured thresholds for a specific amount of time is handled by configuring Time Over Threshold alarms for the metric.  See the The Time Over Threshold Event Rule techdoc for more information.

The Message Counter field in an Auto-Operator Profile allows you to use the alarm count in the filter you define in the profilee. You specify a Message Counter operand (Less than, Equals to, or Greater than or equal) and the message count that will be used for the execution of the Action type that you define in the profile.

------------------------------
Kathy Maguire
Technical Support Engineer 4
Broadcom
------------------------------

[David Michel]

That is a ToT Time over Threshold alarm. 
For anything beyond that more info is needed.

------------------------------
Support Engineer
Broadcom
------------------------------

[erichuang82412668]

Is this new feature in UIM 9?  what about 8.51?

[Kathy Maguire]

ToT alarm configuration is available in UIM 8.51.  You define your ToT alarms from a probe's Configuration page opened from the Admin Console.

------------------------------
Kathy Maguire
Technical Support Engineer 4
Broadcom
------------------------------

[erichuang82412668]

thanks, one more question,  this need to config on the second HUB, not Primary, right?

[Kathy Maguire]

The hub managing the probes that you create ToT alarms must have the following set of probes installed on it:

baseline_engine
prediction_engine
ppm
nas

The baseline_engine probe generates the alarms.
The alarm_enrichment probe of the nas processes the alarms and determines when they have met the configured ToT thresholds before publishing them to the nas when they become "visible".
The prediction_engine is responsible for processing the Time To Threshold alarms.
The ppm probe is required in order to be able to configure probes on robots managed by the hub from the Admin Console.

Note that you should configure Forwarding and Replication in the nas probe to forward the alarms from any secondary hubs on up to the primary hub through the nas probe.  You should NOT have an attach queue with a subscription to alarm messages configured on these secondary hubs.

------------------------------
Kathy Maguire
Technical Support Engineer 4
Broadcom
------------------------------

[erichuang82412668]

I found the place to config the event.  For VMware probe, I have to config Time over Threshold on each VM?  Is there any place that I can setup for all VMs. if so, I don't need to config one by one. Please advise.

Thanks,

[Kathy Maguire]

Ideally, you would configure templates from the vmware probe Configure page in the Admin Console (highly recommended).  In the template, you would define your ToT thresholds.  This way, when the template is applied to the monitored vCenter, the ToT alarms will be applied to all discovered VMs.  If you do not use templates from the Admin Console, then the only other way to do this would be to define ToT alarm thresholds for each individual VM (not recommended).  In order to enable templates for the vmware probe, you must enable bulk configuration in the probe's setup dialog.  Bulk configuration is disabled by default.  It is also disable if the probe has been previously configured.  You can only enable bulk configuration when no resources have been configured.  See the vmware AC Apply Monitoring with Templates  techdoc for more information about configuring templates for the vmware probe.  Also note that once you enable bulk configuration, you cannot configure the vmware probe from Infrastructure Manager.

------------------------------
Kathy Maguire
Technical Support Engineer 4
Broadcom
------------------------------

[erichuang82412668]

how could I create a template in Admin console?  you mean web console? you mean in the RAW configure of VMware Probe?   I only create template in infrastructure manager.  Please advise.

Thanks,

[Kathy Maguire]

Yes, the Admin Console is the same thing as the web console.   From a new vmware probe installation (or after deleting all existing resources), you need to enable bulk configuration first by selecting the Enable bulk configuration option after selecting the vmware node from the left-hand pane:

Save the configuration change:

Then the vmware probe's Template Editor becomes available:

You would create your monitoring template by copying one of the default templates, enable the monitoring you want/need.  It is here that you would define the Time Over Threshold alarms for the components/metrics where you need them.  Once you have your template defined and activated, then add your resources to be monitored.  Any template that you create and activate will be applied to the resources configured.  Again, this is documented in the vmware AC Apply Monitoring with Templates mentioned in my last update.

Once you have enabled bulk configuration for the vmware probe, you can no longer configure the probe from Infrastructure Manager.  If you try to open this Configure GUI, you will get the following error:

------------------------------
Kathy Maguire
Technical Support Engineer 4
Broadcom
------------------------------

[erichuang82412668]

Thank you so much, that's great hlep.