HPE StoreVirtual Storage / LeftHand
cancel
Showing results for 
Search instead for 
Did you mean: 

SNMP Monitoring

David Bamber
Occasional Visitor

SNMP Monitoring

We are wanting to develop a very lightweight monitoring tool which is able to report back to us on the usage status of our different storage devices across the network.

One of these devices is our Lefthand SAN which we had hoped to provide a simple report in the form of Provisioned space vs provisionable space which we would monitor via SNMP. Unfortunatly I have not been able to find any appropriate OID's in the MIB files and wondered if anyone else had tried to achieve something similar and found an alternative work-around?

Any help would be much appreciated.

Dave
6 REPLIES
Rich Buckley
Occasional Advisor

Re: SNMP Monitoring

Here are some Dave. I will find more that I have used.

clusVolumeSnapshotClusterUsedPercent
maximum space used by snapshot and writable space (high watermark) as a percent of cluster space
.1.3.6.1.4.1.9804.3.1.1.2.12.101.1.31

Modules:
clusModuleStatsServerIoLatencyTotal
time spent waiting for read and write operations to complete for this server
.1.3.6.1.4.1.9804.3.1.1.2.12.46.1.18

clusModuleStatsIoLatencyWrite
time spent waiting for write operations to complete for this module
.1.3.6.1.4.1.9804.3.1.1.2.12.46.1.27

clusModuleStatsIoLatencyRead
time spent waiting for read operations to complete for this module
.1.3.6.1.4.1.9804.3.1.1.2.12.46.1.26

clusModuleStatsQDepthTotal
number of pending read and write operations
.1.3.6.1.4.1.9804.3.1.1.2.12.46.1.25

clusModuleStatsKbytesWrite
counter of Kbytes written
.1.3.6.1.4.1.9804.3.1.1.2.12.46.1.24

clusModuleStatsKbytesRead
counter of Kbytes read
.1.3.6.1.4.1.9804.3.1.1.2.12.46.1.23

clusModuleStatsIOsWrite
counter of IO write operations
.1.3.6.1.4.1.9804.3.1.1.2.12.46.1.22

clusModuleStatsIOsRead
counter of IO read operations
.1.3.6.1.4.1.9804.3.1.1.2.12.46.1.21

clusModuleStorageStatus
storage status of a module
.1.3.6.1.4.1.9804.3.1.1.2.12.46.1.19

clusModuleRaidStatus
RAID status of a module
.1.3.6.1.4.1.9804.3.1.1.2.12.46.1.10

clusModuleAvailSize
available storage size of module
.1.3.6.1.4.1.9804.3.1.1.2.12.46.1.9

clusModuleSerialNo
serial number of module
.1.3.6.1.4.1.9804.3.1.1.2.12.46.1.6

clusModuleName
hostname of module
.1.3.6.1.4.1.9804.3.1.1.2.12.46.1.4

.1.3.6.1.4.1.9804.3.1.1.2.12.46.1.2


ref:00D0hiyN.50002955d:ref
Sam Bohlin
Occasional Visitor

Re: SNMP Monitoring

The best way to get the provisioned space would be to look at the volume table in the Clustering MIB and add up the space used for all of the volumes.

clusVolumeSize will give the size of the volume

clusVolumeUsedSpace will give the space actually used.

clusVolumeProvisionedSpace will give the amount that was provisioned.
David Bamber
Occasional Visitor

Re: SNMP Monitoring

Hi guys, Thanks for the replies so far.


Unfortunatly im not having any luck with this and heres why...

Im using the V8 VSA and V8 mib files. When browsing the ip address in iReasoning MIB Browser and I do a table view for the clusVolumeTable the OID's for the fields described above do not exist.

Is there any reason why the VSA would not have these values whilst the live units do? (Ive not tested this on the live units yet so Im just asking the question) or do you have to turn these values/configure these values somehow?


Thanks again

Dave
teledata
Respected Contributor

Re: SNMP Monitoring

I'm running SNMP counters against some legacy VSAs (originally 7.0) that were upgraded to 8.0, and they respond to the counters.

I just configured some new VSAs 8.1 and they do NOT respond to those OIDs.

Did you ever find resolution on this?
http://www.tdonline.com
teledata
Respected Contributor

Re: SNMP Monitoring

I think I may have figured this out. It looks like you can only get those stats from a module running a manager.

So let's say this:

NSM-MOD01*
NSM-MOD02*
VM-FOM*
VSA01
VSA02
* - denotes manager
Management Group
Prod_Cluster
Storage Nodes:
(NSM-MOD01,NSM-MOD01,VM-FOM)
Dev_Cluster
Storage Nodes:
(VSA01,VSA02)

In order to pull SNMP stats for the VSA you must query a module in the Management Group that is running a manager. Seems stats are handled at the mgmt level (even though you are querying "cluster" counters) and can only be returned by those modules running a manager.

Can someone from Lefthand/HP confirm this?
http://www.tdonline.com
teledata
Respected Contributor

Re: SNMP Monitoring

What I found is if I queried one of my Modules running a Manager I could get clusterwide stats for both clusters. IE:


check_snmp -H $HOSTADDRESS$ -C $USER7$ -P 2c -o LEFTHAND-NETWORKS-NSM-CLUSTERING-MIB::clusClusterStatsIoLatencyRead.1

SNMP OK - 611613136 ms


check_snmp -H $HOSTADDRESS$ -C $USER7$ -P 2c -o LEFTHAND-NETWORKS-NSM-CLUSTERING-MIB::clusClusterStatsIoLatencyRead.2

= SNMP OK - 88396 ms


http://www.tdonline.com