ESXi 5.5 + Cisco UCS = Poor NFS Performance?

If you are running ESXi 5.5 on a UCS chassis, and you get getting poor NFS performance - check your vmkernel.log

We were showing a high level of NFS errors - lock file creation errors, lock file access errors and APD states.

StorageApdHandler: 421: Device or filesystem with identifier [xxxxxxxx-xxxxxxxx has exited the All Paths Down state.

WARNING: NFSLock: 2226: Cannot create lock file .lck-6200000000000000 (4 retries) : Not found

NFSLock: 2929: failed to get lock on file slotsfile

As it turned out, the firmware for the VIC / enet needed to be updated.  An update to the firmware for the UCS blade to version 2.1.3b fixed this issue.

UCS Service Profile Application - "configuration failed due to compute-unavailable,insufficient resources"

I recently deployed a new UCS blade, and I was greeted with the error:

"configuration failed due to compute-unavailable,insufficient resources"

Since it was just a copy of another service profile template with different vNIC settings - I knew there was no changes otherwise.  I simply disassociated the profile, waited for the blade to become available again, and reapplied the service profile...and voila! The error cleared itself and the blade became available.

Nexus 5000 Zoning

If you want to zone in your new FC or FCoE host quickly, use this template.  Obviously, modify the highlighted areas to match your zone names, vsan numbers, and device names.

Fabric A
zone name fabricname-A vsan nn
member pwwn DEVICEWWPN
zoneset activate name zonesetname vsan nn
device-alias database
device-alias name DEVICENAME-A pwwn DEVICEWWPN
exit
device-alias commit
sh fcns database
end
copy run start

Fabric B
zone name fabricname-B vsan nn
member pwwn DEVICEWWPN
zoneset activate name zonesetname vsan nn
device-alias database
device-alias name DEVICENAME-B pwwn DEVICEWWPN
exit
device-alias commit
sh fcns database
end
copy run start