Open Stack

Hi Greg,

The alarming feature of ceilometer is intended to be user-oriented,
i.e. focused on the user-view of cloud resources.

It is not intended to be infrastructure-oriented, i.e. focused on the
operator-view of the underpinning datacenter fabric.

So if understand your question correctly, the answer I thing is no,
we are not addressing the ‘general’ cases you had in mind.

Cheers,
Eoghan

----- Original Message -----
> 
> 
> A general question about Alarms in Ceilometer …
> 
> 
> 
> Is the intent that Ceilometer is collecting, managing & distributing
> ‘general’ OpenStack Alarms ?
> 
> · All of the documentation around Alarms in Ceilometer is about PM Threshold
> Alarms;
> i.e. Alarms triggered when a PM Counter crosses a specified threshold
> 
> · But there are other types of Alarms associated with an OpenStack Cloud
> 
> o Alarms related to equipment failures,
> 
> o Alarms related to network failures,
> 
> o Etc.
> 
> · I have not seen any discussions suggesting that Ceilometer be used to
> collect, manage & distribute ‘general’ OpenStack Alarms
> 
> · The architecture for managing Alarms would be very similar to that of PMs …
> 
> o The Ceilometer Architecture of Listener / Pollster Plugins and Publisher
> Plugins applies well to general Alarm Management as well
> 
> § i.e. Listener / Pollster Plugins for listening or monitoring fault
> conditions
> 
> § Publisher Plugins for generating the appropriate Alarms and sending to an
> Alarm Collector
> 
> § An Alarm Collector with backend plugins for Northbound Management
> Interfaces (SNMP, Netconf, etc.)
> 
> 
> 
> Comments ?
> 
> 
> 
> 
> 
> Greg.
> 
> _______________________________________________
> Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
> Post to     : openstack at lists.openstack.org
> Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
>