Monitor Quota Usage in an end-user friendly way?

[Pete]

I'm very new to the Isilon, we had it installed about 2 weeks ago. I was curious if anyone has a good method for monitoring quota usage on the device in a way that's accessible/understandable by the end user.

We're mapping DFS shares to the Isilon shares, so users aren't mapping these as drives and don't have a way to keep track of how much they've used or have left. Currently, we have several 2TB shares off a Windows server and the users will either map these are drives on their PC, or I'll monitor them using What's Up Gold with SNMP or WMI.

I found a great looking project on a blog where the user monitored quota usage within Nagios, by uploading and executing CLI on the cluster. Unfortunately, I'm stuck with What's Up Gold.

I've read a bit about InsightIQ, we're not licensed for this. However, if we were, would I be able to create an end-user friendly interface using the data that would provide?

Thanks!

[padmanabhaiah v]

You can see the detailed report of the Quotas and Usage in the Quotas & Usage filed. which will give you list of the Quotas and current usage and available space. hope this helps.

WebGui --> File System Management --> SmartQuotas -->Quotas & Usage

Thanks, Paddy

[Chris Pepper]

Pete,

First, "isi quota" has many formatting options.

Second, have you considered dumping quota reports in a public location? Our users all mount /ifs/x/common via NFS but don't have shell access to the cluster, so every 10 minutes we dump a few simple status files into a public directory.

You could also put output into a public directory accessible via HTTP. The hostname check is because jobs in /etc/mcp/override/crontab.smbtime run on *every* node; we don't want all our nodes to write the same output over each other.

Unfortunately group quota email notifications don't go to users -- only to admins.

Chris

> luxisi-1# grep status /etc/mcp/override/crontab.smbtime
> */10 * * * * root sh /ifs/x/home/pepper/bin/status.sh
> luxisi-1# cat /ifs/x/home/pepper/bin/status.sh
> #!/bin/sh
> # Record basic cluster health information
>
> if [ ! `uname -n` = luxisi-19 ]
> then
> exit
> fi
>
> PREFIX=/ifs/x/common/cluster/status
>
> isi alert list > $PREFIX/alert.log
> isi status > $PREFIX/status.log
> isi status -q -d > $PREFIX/pool.log
> isi job status -v > $PREFIX/job.log
> isi quota list > $PREFIX/quota.log
> isi quota list|grep -v :|grep -v default- > $PREFIX/quota-short.log
> isi quota list --is-threshold-exceeded > $PREFIX/quota-exceeded.log
> isi snapshot list -l > $PREFIX/snapshot.log
> isi snapshot usage > $PREFIX/snapshot-usage.log
> isi_for_array -s uptime > $PREFIX/uptime.log
> isi_for_array uptime | tr -d :, | awk '{print $12, $1}' | sort -n | awk '{print $2}' > $PREFIX/ordered-nodes.txt

[Pete]

Chris, I think that sounds like a great idea. We'll likely use the output from isi quota list and present it over HTTP. Thank you for posting. 

[Jerry Uanino]

we dump ours into a text table and tell our devs they can parse that, and I think we update it every 5 min.

--
You received this message because you are subscribed to the Google Groups "Isilon Technical User Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to isilon-user-gr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.