Monitoring without Operations Manager

[tvmo_tvmo]

Hello,

How can I monitor NetApp filers without Operations Manager? My company has SAN Screen, but I'm not sure it will do what i want. for example:

1. Volume full

2. Aggregate full

3. Snapshot Reserver full

4. SnapVault replica out of date

5. SnapMirror lag

6. Snapshot failed

7. SMVI/SQL/Exchange

8. Filer down

9. Port offline

10. etc.

Thanks

[ostiguy]

SANscreen 6.0 will understand thin provisioning for Flex/trad vols and LUNs (we will know allocated capacity, consumed (on the NetApp array) capacity, and raw capacity)

In Flex/trad vols, we will also understand used capacity, dedupe ratio, snapshot reserve/used.

6.0 has a "Storage Pool" object in the data model - we will map NetApp aggregates to it, and understand raw, usable, used, snapshot reserve

So, I think that covers a chunk of 1-3, but perhaps not all. We will not throw alerts for snapshot reserve full, but I believe we will show "Snapshot Used" > 100% when that is the case.

We don't tackle 4--7

For filer down, we will see if its ports go down, but if the array panicked, and the ports stayed up, we would not know - our datasource polling the NetApp would fail, but when that happens we do not assume the array is down.

9, port offline, that we certainly do.

Etc - If you own Application Insight: Array Performance, FAS and vSeries (and IBM N Series) are supported now.

Hope this helps

Matt

[steve_francis]

If you want a commercial solution, www.LogicMonitor.com does a great job of monitoring all the NetApp things you mention, plus a lot more. And it's monitoring by default - so as soon as you add a volume, it's latency and requests and throughput will be monitored per volume. Add a PAM II card, and it's utilization will be monitored, etc.

And has nice Exchange and SQL server monitoring too (plus everything else).

[tvmo_tvmo]

Additonal purchase is not an option, Im stuck with what I've got

For the items not covered by SANscreen, what are my options? I'm used to seeing Operations Manager, so I'm a bit stuck without it.

Thanks

[steve_francis]

You can cobble together something with cacti and nagios, but unless you have lots of idle time, or your time is free, it's far cheaper to use something automated like LogicMonitor.

Not to mention the cost of things breaking as nagios wasn't updated manually to be aware of the new snapmirror, so no alerts were triggered, etc.

[tvmo_tvmo]

Thanks Steve.

As s synopsis then:

1. Volume full - SANscreen

2. Aggregate full - SANscreen

3. Snapshot Reserver full - SANscreen will alert when snapshot used space is 100%

4. SnapVault replica out of date - No free NetApp alerting/monitoring

5. SnapMirror lag - No free NetApp alerting/monitoring

6. Snapshot failed - No free NetApp alerting/monitoring

7. SMVI/SQL/Exchange - No free NetApp alerting/monitoring

8. Filer down - No free NetApp alerting/monitoring

9. Port offline - SANscreen

Thanks

[steve_francis]

probably the other big thing I always like to have monitored is per volume latency. Which Ops Manager does nicely, cacti can do, dont know about nagios. LogicMonitor does.

And I shoukld have mentioned that LogicMonitor is only $300/month for up to 25 systems - it's SaaS, so you can try it for a month.

[fjohn]

Take a look at the Data ONTAP PowerShell Toolkit (which is free to NetApp Customers and Partners).  If you find something you can't do in SANScreen 6.0, you'll likely be able to do it using the toolkit.

J

[tvmo_tvmo]

Thanks guys, is it possible that my list can be confirmed please, just so as I'm absolutely clear?

1. Volume full - SANscreen

2. Aggregate full - SANscreen

3. Snapshot Reserver full - SANscreen will alert when snapshot used space is 100%

4. SnapVault replica out of date - No free NetApp alerting/monitoring

5. SnapMirror lag - No free NetApp alerting/monitoring

6. Snapshot failed - No free NetApp alerting/monitoring

7. SMVI/SQL/Exchange - No free NetApp alerting/monitoring

8. Filer down - No free NetApp alerting/monitoring

9. Port offline - SANscreen

I haven't done any scripting, are there any good resources especially for NetApp work?

Thanks

[ostiguy]

On #3, we will not throw an alert - you have visibility into snapshot space utilization, but we don't throw alerts.

[tvmo_tvmo]

Does SANscreen do any alerting?

What about the other points?

Re the scripting, any good NetApp resources, example scripts, etc?

Thanks

[ostiguy]

Our violations are visible in the SANscreen client, and if SNMP traps have been configured, we can throw those to the SNMP monitoring solution of your choice.

We have switch fabric alerts (configurable for utilization, error, CRC, Loss of Signal/Sync, etc) as well, and again, if SNMP traps have been configured, we will throw traps for these as well. The alerts are viewable and dismissible in the client as well

[dmorera]

For #7 you could use snapdrive alerting. I use this to identify issues with SME and SMSQL as well as SMVI.