OCUM6.0: launching a script after an event/alert triggers in a Clustered ONTAP config

[javierb]

Hello people

I have being viewing some interesting videos about OCUM6.0 located here communities:

How to Monitor Events and create Alerts in OnCommand Unified Manager 6.0

https://communities.netapp.com/videos/3428

Monitoring cluster resource availability in OnCommand Unified Manager 6.0

https://communities.netapp.com/videos/3426

I would like to know if the scenario i am thinking of it's possible or not in CDOT8.2+OCUM6.0.

Based on the availability of any HA pair existing in a Clustered  configuration (even number of nodes-controllers),

I would like to trigger some commands or script  in case of node failure (node as RESOURCE, and, failure as EVENT) after the failover process has been carried out and finished.

Basically I would issue some commands after failover is done.

Is it possible to do this with OCUM6.0, or do i need additionally WFA (Work Flow Automation)?

in case of being doable, could anybody "picture" describe the steps to put this to work?

Thxs & Rgds

[venkat3]

Hi Javier,

Welcome to communities

>Basically I would issue some commands after failover is done.

I am curious to know more on the commands/script part.

What commands/script would you want to run as soon as a node failure happens?

>Is it possible to do this with OCUM6.0,

AFAIK running custom scripts would be made available in upcoming releases.

It would help if you elaborate on the custom scripts you plan to run.

Today OCUM6.0 sends alerts/snmp traps for Storage Failover events

-srinivas

[javierb]

Hello Srinivas

i try to explain what i have in my mind to be accomplished.

Supposing a 4-node cDOT 8.2 configuration.

So, we have two FAS HA pairs, where every node has a defense against contingencies and events backed up by the other HA pair node.

Supposing it happens one  failover and takeover in one of the HA pairs.

It is in that situation and after the failover process has been completed where I would like to issue one script (or some commands),

The commands would be some kind of 'vol move' like below

  clusterenvironment1> volume move start -volume vs7_vol01 -destination-aggregate aggr2_cluster01_01

or perhaps some kind of SnapVault update

  clusterenvironment1> snapmirror update -vserver vs7 -volume vs7_vol01

That's my idea and the origin of my question.

is it Not possible to do that as soon as a node failure happens in this release of OCUM6.0?

If not, are you aware of when this is gonna be available in OCUM?

Thxs & Rgds

[javierb]

anybody has news on this?

thxs

[msaravan]

There is no in-built functionality available in UM6.0 to trigger scripts whenever event (detection of failure conditions) is triggered.

Whenever such events happen, UM6.0 has the capability to notify admins via e-mails and SNMP traps. You can achieve your expected behavior, if you know any ways to trigger scripts upon receiving e-mails or traps.

Regards,

Saravanan

[adaikkap]

Hi,

    If you are looking for the functionality that you had in OCUM 5.x in OCUM 6.0, then the answer is, its not available currently in the product.

Regards

adai

[javierb]

Thanks Saravanan and Adai

Well, after reading that, I see it it possible to use for instance Nagios to catch the trap or traps generated and send some rsh/ssh commands to one oa more controllers in a cDOT.

The point is I will have to make more work of parsing some information after receiving the traps(s)s and before issuing some commands because I will need some volumes details and ...

Not so easy , I mean it is not one or two mouse clicks.

When will be this feature available in OCUM 6.0 (like we do/did in OC5)?

thxs & rgds