2014-02-26 09:09 AM
Using OnCommand Unified Manager for Clustered Data ONTAP 6.0 - I receive a "Cluster Lacks Spare Disks" warning even though the cluster has at least 2 spare disks of every disk type available.
What is the threshold for this alert and how can I amend it? It used to be possible to do this in version 5 on Windows but with the OVA deployment it doesn't seem to be possible to change events like this.
2014-02-26 09:43 AM
Cluster Lacks spare disks event in OCUM is generated based on the result of the clustered ONTAP API aggr-check-spare-low.
If the API returns true for a particular node or for both nodes, the event will be triggered in OCUM.
The event is not only based on disk types but also based on disk size, rpm and certain other parameters.
2014-02-27 07:06 AM
Are the rules that generate this alert documented in the user documentation, online help, or a public support KB article? I've also encountered the error and was not able to determine the cause.
The event should include an explanation of the rules, or a link to an explanation of the rules.
CSE - Storage Management Solutions
2014-02-27 07:56 AM
The triggered condition in the event summary section in the event details page will help us in getting the rule.
Looks like for Cluster Lacks Spare Disk event more documentation is required.
I will communicate this to the IE team.
2014-02-28 07:10 PM
thanks for posting this question.
>I receive a "Cluster Lacks Spare Disks" warning even though the cluster has at least 2 spare disks of every disk type available.
Certainly ontap cluster is running out of spare disk.Hence ocum is showing this event.
can you paste outputs of the below commands for us to help you better.
a) disk show -spare
b) node run -node <nodename> sysconfig -r <--- Event details would show which node is having low spare.Run this command against that node.
for the second part of the question
> What is the threshold for this alert and how can I amend it?
Just curious to know why would to like to change the default behavior . Are you being spammed for alert emails?
IMO if you are getting "Cluster Lacks Spare Disks" event on any of your clusters , you are certainly running out of spares on nodes and it should be corrected.(its certainly not false alarm)
Though disk show -spares show enough spare disks , as pradeep mentioned above its possible that the 2 spares left are of diff RPM
output of above command would help in telling you why "Cluster lacks spare disks" is seen.
hope this helps,