I am currently implementing the SCOM monitoring solution for the NetApp devices on a customer's estate. There is a debate as to whether monitoring using OnCommand 3.0 and the supplied Management Packs or the 'old fashioned' solution using the SNMP trapping method from the NetApp Operations Manager should be implemented. Are the pros and cons of each solution documented anywhere, especially regarding the types of devices supported (if there are any limitations)? My personal view is that the SNMP type monitoring should not be used as there is a new and improved method but the storage support team are concerned that this means that alerts from the NetApp Operations Manager console will be missed. I am afraid I am not a storage expert so do not know if this is the case or not.
Use OC server to do the monitoring and in storage system add OC server as SNMP server. Doing this way will make sure that your OC is pooling the filers at periodic interval to get current status and at the same time if anything goes wrong at filer it will send the SNMP trap to OC server notifying the current status.
Say if you have made volume full scan in OC server to run every 10 min. and last run was at 11:30 but a volume went full at 11:35 so until next pooling interval which is 11:40 you will not come to know but having SNMP server configured on filer, filer will send a SNMP trap to OC server at 11:35 itself and OC server will do pooling of that volume at 11:35 only.