module SNMP MIB performance statistics

[teledata]

I can’t seem to make heads or tails of the LEFTHAND-NETWORKS-NSM-STORAGE-MIB

This query for device temperature:
$USER1$/check_snmp -H $HOSTADDRESS$ -C $USER7$ -o LEFTHAND-NETWORKS-NSM-STORAGE-MIB::storageDeviceTemperature.1
Returns this: 4294967295

What the heck kind of units does this storage mib use? None of them make any sense to me.

Unfortunately LEFTHAND-NETWORKS-NSM-STORAGE-MIB::storageRaidStatsIOsRead.0 is giving me a pretty useless number as well
It’s a COUNTER style, so I took a reading, and a reading 4 seconds later, and the difference was:
3272326219793694744
Subtract 3272323243381358616
=2976412336128

What the heck kind of IOs are these?

I really like the numbers I get from $USER1$/check_snmp -H $HOSTADDRESS$ -C $USER7$ -P 2c -o LEFTHAND-NETWORKS-NSM-CLUSTERING-MIB::clusModuleStatsIoLatencyRead.1

Unfortunately due to SAN/iQ design. I can ONLY query these counters if the module I’m querying is also a SAN/iQ manager….

Is there a way to query useful stats for modules that ARE NOT managers?

I could turn on managers, but then I’d be violating best practices for a max of 5 managers per Mgmt group…. (also creates some exposure, due to VSA’s having managers, so if some VSAs were to go offline I could potentially loose quorum easier…

Thanks!

[Gauche]

What version of SAN/iQ are you asking this about?

[teledata]

8.0 (NSM2120)

and some of them are 8.1 (VSA)

[Saranya123]

We are running into the same issues, did you get an answer for this one from HP may be?

The difference in the counters return a large value after using new OIDs, which don't seem to be IO's value.

It would be great if you can share any information!

Thanks!

[teledata]

This has been my biggest frustration. I've requested information from the Support folks, and made inquiries, and can't seem to get any good information on this.

HP will add/remove/deprecate OIDs from the MIBs as SAN/iQ versions are updated, but there appears to be no notification or available documentation about these changes.

Several OIDs that previously worked are now broken, I also found that several storage level OIDs only work with the newest hardware.. (ie: I can't get node level IOPs, KBps, latency) statistics from the VSA or legacy models (NSM2120), but I can from a new G2 cluster....

Attention HP: PLEASE document and distribute details about the OIDs/MIBs for SAN/iQ 8.5!!!!!

Here's a page on my website that details some HP LeftHand monitoring I've done:
http://www.tdonline.com/training/lefthand/

Here's some sample scripts:
http://www.tdonline.com/training/scripts/p4000-checks.zip