Has anyone ever used the Performance Advisor in the NetApp Management Console to troubleshoot performance issues? If so, I'm particularly interested in anyone's comments on these two items:
Network Errors - one of the results indicated "following ifnet objects has violated Network Errors - (list of interfaces). Check the network." Talk about a nondescript error. Anyone have an idea of what this error means?
Atime Updates - this one is a lot more straightforward. The advisor recommends setting the no_atime_update option to on to improve performance. My understanding is that this option simply provides the last access time of a file, such as what you would see when viewing the properties of a file in Windows. My question is: is there any downside to turning this on? Is this option better suited to a certain type of volume, such as NFS (used as a Vmware datastore), iSCSI, and/or CIFS?
Atime updates are absolutely useless for LUN and VMware data stores, you can safely turn them off there. For CIFS it us up to you - do touched to know when file was accessed last time?
Thank you aborzenkov! That is what I was thinking as well, but wanted to verify. I will turn it off for LUNs and VMware, and leave it on for CIFS (and NFS volumes that are connected to Linux servers).
Any ideas on the network errors message? If not I'll open a case with NetApp.
