Active IQ Unified Manager Discussions

Alert upon reaching 80% volume space utilization

egono
15,526 Views

Hi,

I am new here, please understand if i am on the wrong place posting my concern.

And this is my concern, i have a LUN and it is 50GB of space residing on a volume with 50GB usable space. The space guarantee of the volume was set to volume. The LUN does not have space reserve. The LUN is mapped on a SQL server and function as repository of the Database Log. There was a time that the LUN was offline due to lack of sapce on the volume. I increased the size of the volume as well as the LUN size and put the LUN on online state again. I don't want to encounter the same problem again and I'm thinking if there is a possibility that the filer will alert me if the volume reaching 80% space utilization?

Thanks.

The filer has Data ONTAP 7.3.1 installed.

10 REPLIES 10

tanmoy
15,495 Views

Hi Edwin,

Operations Manager has volume level thresholds which can help you in this case. The default thresholds for volumes are set at 80%( nearly full threshold--warning alert) and 90% ( full threshold-error alert). In your case if the LUN were space reserved and the size of the LUN were same as volume size; then you would have got always error alerts because it's 100% used.

But as your LUN is not space reserved; you should see the alerts as warning or error when LUN used space approaches to 80% or 90% of volumes space. If you are using Operations Manager then you can get those alerts easily.

Thanks

Tanmoy

egono
15,495 Views

Hi Tanmoy,

Thanks for your support, by the way is there any other options that may address my concern?

Edwin Gono

rbalaji
15,495 Views

Hi Edwin,

You can tweak the global options 'volFullThreshold' (by default 90) or 'VolNearlyFullThreshold' (by default 80) to suit your environment.

(Refer dfm option set help).

And set the alarms (e.g: could be email) from DFM for these event types.

You can as well configure these options at individual volume level (dfm volume set help).

egono
15,495 Views

Hi Balaji,

Thanks for your support.

I think DFM or Data Fabric Manager is an older version of Operations Manager.

It was already suggested by Sir Tanmoy on his post.

How about using the autosupport tool to issue an alert if I am nearing the 80% volume space utilization. Is there a way that I can set a threshold on volume options? Do you think this option can possibly work?

Thanks.

rbalaji
15,495 Views

Hi Edwin,

Operations Manager is the new name for the DataFabric Manager web interface.

<<<< How about using the autosupport tool to issue an alert if I am nearing the 80% volume space utilization.

Which autosupport tool are you referring to ?

Instead you can use DFM alarms.

From Operations Manager Web UI, Navigate to 'Control Center'->Home->Setup->Alarms and create an alarm for the required event (Volume Almost Full in your case).

<<<< Is there a way that I can set a threshold on volume options?

Its mentioned in my earlier post. You can set these options at volume level. Refer 'dfm volume set help'

pash_hrnic
15,495 Views

I'm trying to do the same thing as the OP - confugre alerting w/out using the operations manger...

I want to configure autosupport to alert when ANY volume on a filer reaches a certain threshold...

dfm volume set help yields nothing...

Am I missing something?

I'm fairly certain that dfm (Data Fabric Man) needs to be in place before 'dfm volume set help' will work.

Like the OP, I require a native alerting solution...

Thanks.

prasads
15,495 Views

HI,

Pasting below an example of how this threshold setting works with DFM.

Hope this helps.

[root@lnx186-92 ~]# dfm volume set deux183-205:/vol0 volNearlyFullThreshold=70 volFullThreshold=80

Changed volume nearly full threshold (%) for volume deux183-205:/vol0 (728) to 70.

Changed volume full threshold (%) for volume deux183-205:/vol0 (728) to 80.

[root@lnx186-92 ~]# dfm volume get deux183-205:/vol0

Volume:                                                                          deux183-205:/vol0

Volume Full Threshold (%):                                                       80

Volume Nearly Full Threshold (%):                                                70

Volume Full Threshold Interval:                                                  00:00:00

Enable User Quota Alerts, not available for vserver volumes:                     yes

User Quota Full Threshold (%), not available for vserver volumes:                90

User Quota Nearly Full Threshold (%), not available for vserver volumes:         80

Volume Snap Reserve Full Threshold(%):                                           90

Volume Nearly No First Snapshot Threshold(%), not available for vserver volumes: 80

Volume No First Snapshot Threshold(%), not available for vserver volumes:        90

Volume Space Reserve Nearly Depleted Threshold(%):                               80

Volume Space Reserve Depleted Threshold(%):                                      90

Volume Growth Event Minimum Change (%):                                          1

Volume Quota Nearly Overcommitted Threshold (%):                                 95

Volume Quota Overcommitted Threshold (%):                                        100

Volume Snapshot Count Threshold:                                                 250

Volume Too Old Snapshot Threshold :                                              52 weeks

[root@lnx186-92 ~]# dfm volume set

Valid options are

  volFullThreshold          Volume Full Threshold (%)

  volNearlyFullThreshold    Volume Nearly Full Threshold (%)

  volFullThresholdInterval  Volume Full Threshold Interval

  userEnableAlerts          Enable User Quota Alerts, not available for vserver volumes

  userFullThreshold         User Quota Full Threshold (%), not available for vserver volumes

  userNearlyFullThreshold   User Quota Nearly Full Threshold (%), not available for vserver volumes

  volSnapshotFullThreshold  Volume Snap Reserve Full Threshold(%)

  volNearlyNoFirstSnapThreshold Volume Nearly No First Snapshot Threshold(%), not available for vserver volumes

  volNoFirstSnapThreshold   Volume No First Snapshot Threshold(%), not available for vserver volumes

  volReserveNearlyDepletedThreshold Volume Space Reserve Nearly Depleted Threshold(%)

  volReserveDepletedThreshold Volume Space Reserve Depleted Threshold(%)

  volGrowthEventMinChangePct Volume Growth Event Minimum Change (%)

  volNearlyOvercommittedThreshold Volume Quota Nearly Overcommitted Threshold (%)

  volOvercommittedThreshold Volume Quota Overcommitted Threshold (%)

  volSnapshotCountThreshold Volume Snapshot Count Threshold

  volSnapshotTooOldThreshold Volume Too Old Snapshot Threshold

  volNearlyOverDeduplicatedThreshold Volume Nearly Over Deduplicated Threshold (%)

  volOverDeduplicatedThreshold Volume Over Deduplicated Threshold (%)

[root@lnx186-92 ~]# dfm volume set help

NAME

    set -- change volNearlyFullThreshold and volFullThreshold

SYNOPSIS

    dfm volume set <object> <option-name>=<option-value> ...

DESCRIPTION

    Changes one or more options of a volume.

    All option names will be listed by executing

    this command without any parameters.

pash_hrnic
15,495 Views

Great and all if you own data fabric manager...

Unfortunatly we don't.

Some promising info I've found is to configure swatch on a separate host to monitor the log files and send our alerts as per the desired level...

http://partners.netapp.com/go/techontap/swatch-kiwi.html

Another solution we are looking into is loading the mibs for the filers into our SNMP monitor...

Thanks.

kirk_brady
15,495 Views

Hi,

I wouldn't advise setting up SNMP traps for volumes. I did this and found the volume OID's changing. Then I found this 'feature' is documented here - https://now.netapp.com/Knowledgebase/solutionarea.asp?id=ntapcs17840. Netapp say to never use the OID as a reference to a particular volume since it is subject to change.

I would recommend getting Operations Manager if you can.

If you can't, Nagios can also monitor volume space and alert on thresholds, with the appropriate scripts.

madden
7,493 Views

Hi,

Since you're using thin provisioning why not implement it using vol autogrow and snapshot autodelete to avoid the LUN ever going offline?

Discussions and how to enable are in this tech report:
Thin Provisioning in a NetApp SAN or IP SAN Enterprise Environment

http://media.netapp.com/documents/tr-3483.pdf

I believe the NetApp System Manager (freely downloadable from NOW) even sets up a best practices config for you!  A blog about it is here:

http://www.pkguild.com/2009/04/netapp-nsm-san-v-nas-creation-differences/

Lastly, if you do want to monitor disk used, and don't want to buy Operations Manager (it's not very expensive...might even be free for small configs...ask your salesman) then you could hack "Toasterview" from the toolchest:

http://now.netapp.com/eservice/toolchest?toolid=463

Hope that helps.

Cheers,

Chris

Public