Symptoms:
-
Very poor SAN performance on Cisco UCS B200 M3 Blades.
-
Deploying VMs from templates and cloning VMs hang.
-
CPU stats in a VM indicated 100% cpu in wio state
Background:
-
Cisco UCS provides a QoS feature to allow prioritization of certain types of traffic.
-
Cisco UCS supported multi-hop FCoE from release 2.1 onwards.
Resolution:
Enable Packet Drop (no pause) for any enabled priority (in my case below, it was Platinum) when using UCS 2.1+ i.e. UCS with multi-hop FCoE. This was not required in earlier versions of UCS using native FC.
Root Cause:
I am not a networking expert to provide much clarity here, but according to Jeremy Waldrop, packet drop cannot be enabled on 2 different QoS groups on the same interfaces. I thought this wouldn’t matter because the QoS groups are assigned to different CoS groups, but I don’t know much about this to discuss further.
(1) The solution above describes a successful problem-solving experience and may not be applicable to other problems with similar symptoms.
(2) Your rating of this post will be much appreciated as it gives me and others who read this article, an indication of whether this solution has worked for people other than me. Also, feel free to leave comments.
