It used to just be a “stats show –i <minutes>” but it may be something from the Java API now, I’m not sure. Stats show still works and can provide the same information, getting it to write out is a little confusing.

 

Frankly it’s disappointing that NetApp has emphasized support on the thicker method for performance monitoring, SNMP + SSH worked well when you have dozens or more filers to handle. We don’t have a common directory so creating a single local ID with a ssh key is much easier for us from a management point of view.

 

Too bad Ontap’s command line makes AIX look good L

 


From: Fred Grieco [mailto:fredgrieco@yahoo.com]
Sent: Monday, January 31, 2011 12:51 PM
To: Page, Jeremy; toasters@mathworks.com
Subject: Re: latency measurement in DFM/Management Console/etc

 

Thanks, this makes sense.  But...

The graph seems to roll along at 1 minute intervals.  This is in "Management Console" which I assume is also called Performance Advisor.  The default interval for this in DFM though is 5 minutes (perfAdvisorPollInterval). Is there some other setting for this? 


From: "Page, Jeremy" <jeremy.page@gilbarco.com>
To: Fred Grieco <fredgrieco@yahoo.com>; toasters@mathworks.com
Sent: Mon, January 31, 2011 8:54:18 AM
Subject: RE: latency measurement in DFM/Management Console/etc

Frequently this is from too short of a sample period.

 

______________________________________________________________________________________________________________________________________________________

* email:Jeremy.Page@gilbarco.com| Servers and Storage  | Gilbarco Veeder-Root

( phone: 336.547.5399  | ( cell: 336.601.7274  | 6 fax: 336.547.5163


From: owner-toasters@mathworks.com [mailto:owner-toasters@mathworks.com] On Behalf Of Fred Grieco
Sent: Monday, January 31, 2011 8:44 AM
To: toasters@mathworks.com
Subject: latency measurement in DFM/Management Console/etc

 

How does the Netapp or DFM measure volume and lun latency?  Does it just measure how long it's taking to service requests from the back ports to the front ports?  Or does it measure some sort of fibre ping to the host?

I am looking in the management console in DFM/Ops Mgr and seeing spikes of very high volume and lun latency (seconds...) on an N-Series IBM gateway filer.  I've looked at all the usual suspects, including misaligned VMs, high correlated IOPs, delays on the backend disk, deferred consistency points.  Nothing is matching up.  Most of the connected systems are on a Cisco UCS with NPIV.  The systems with the lowest IOs have the highest latency numbers.

Thoughts?

TIA,

Fred

Please be advised that this email may contain confidential
information. If you are not the intended recipient, please notify us
by email by replying to the sender and delete this message. The
sender disclaims that the content of this email constitutes an offer
to enter into, or the acceptance of, any agreement; provided that the
foregoing does not invalidate the binding effect of any digital or
other electronic reproduction of a manual signature that is included
in any attachment.

Please be advised that this email may contain confidential
information. If you are not the intended recipient, please notify us
by email by replying to the sender and delete this message. The
sender disclaims that the content of this email constitutes an offer
to enter into, or the acceptance of, any agreement; provided that the
foregoing does not invalidate the binding effect of any digital or
other electronic reproduction of a manual signature that is included
in any attachment.