What are "Other Ops" and "Other Latency"

[thigian007]

1. What is “Other Latency”. I know it is everything OTHER than reads and writes but
   what does that boil down to.
2. Is it bad for it to be high even though read and wright latency is low. I have a
   volume that it is in the 120 ms  on and then read and write is below 10ms.

Another place I've noticed this measurment is on the "lun stats -o"

For a little bit of back ground. These are linux hosts running and accessing these LUNs via FCP.

Any help is appreciated.

[michelvankessel]

In the case of NFS, other OPS are all NFS calls not being read or write.

The command "nfsstat" shows you these calls. I don't know in the case of a LUN. But would like to know, so if you have some more info, please share

regards

Michel

[thigian007]

This is a lun that is directly placed inside a volume. No Qtree. It is access via FCP by a RHEL 5.3 host that is running Oracle. We are running Ontap 8.0.1.

We are not licensed for NFS so nfsstat shows Zero for everything.

Is there any more information that would give you a better picture of what's going on?

[vladimirzhigulin]

Hi,

I have a similar problem with high values for "latency_other" but for LUN's being accessed via iSCSI, in this case you can do the following:

1. priv set diag

2. lun hist

I think the same is applicable for FC protocol as well.

Cheers,

Vladimir

[thigian007]

I'd like to  close this discussion in favor of this one: https://communities.netapp.com/message/56214#56214