IBM Storage - Consisten Stopped status does not cause an alert to be sent

GuillePM

I was doing some alerting tests with my IBM FlashSystem. If there is a failure in replication caused by the link between systems being broken, an event is generated in the FlashSystem, which causes the Alerting configuration in Stor2RRD to shoot an email, as expected.

But, when a remote copy consistency group is stopped manually, whichever the state it is in, an event is not generated in the FlashSystem, a red arrow is shown in Stor2RRD but an alert does not get sent.

Is this expected? If it is expected, shouldn't that change? I guess everyone would want to know when a remote copy is stopped, even if intended.

Thanks!

Pavel

alert is sent only when status goes from green to red.

If it is already in red, then no new alert is generated

Does it explain your behaviour?

GuillePM

It does, what I mean is that, if I manually stop a remote copy consistency group, the system does no go to red at all. Stor2rrd recognizes that the consistency group is stopped, it shows it as red in the Mirror tab, but since no event is generated, the system does no go to red. I understand the behaviour, but I think that, even if the system does not go to red, it should still send an alert about the stopped mirror.

Pavel

I understand you, but we do not keep alert history necessary for such things to check if there is a any further change which would case red change.

we do is simple green/red alerting only, nothing else for now.

It is same like for the storage healt status.

If the storage go to red state (some HW error) we do not alert any new critical error which might appear.

I think critical errors should be resolved asap

GuillePM

I completely understand what you're saying. What I'm suggesting is sending an alert whenever a remote copy consistency group is not in "Consistent synchronized" status.

Pavel

I understand you, but is implementated green --> red status only alerting.

We would have to re-implement complete alerting for that.