The upgrade itself completed successfully, but SP failover did not complete in a timely enough fashion to preclude our XenServer High Availability from kicking in and fencing a few of our servers.
EMC Support reiterated the SP failover (i.e. IP addresses normally assigned to one SP being handled by the other SP when the first SP is being upgraded/rebooted, etc.) was supposed to have happened under the time threshold for which we have set our HA timeout, but our HA logs indicate that was not so.
Consequently, EMC Support collected log files and service information, and promsised to get back to us with the maximum amount of time an SP failover can possibly take, so we can adjust our HA failover timeout accordingly.
I'm installing it today so I'll let you know. I am still having an issue where it reboots out of the blue taking everything down. Last time they said 2.1.1 would fix it. Starting to regret getting this thing.
I just upgraded to version 2.1.3.16008 from 2.1.1.14913 early this morning. We were having major issues with our datastores on the VNXe losing connectivity under heavy load i.e. during backups. This fix was in the "Fixed Bugs" section of the release notes. The new version is working great so far and it looks to be working faster for our iSCSI LUNs.
I applied the update with no noticible effects yet.
I shut down all of my virtual machines before doing the update. No matter what I have read, during updates the SP's don't failover fast enough to keep my VM's up, Doing updates are a real pain, but nessessary since I'm desperate to fix these unscheduled reboots from happening. Every time it happens they tell me to upgrade the OS.
Has anyone else experienced a similar problem with viewing their performance data since the upgrade? We are going to be installing our new VNXe 3300's in the next week or so and intend to upgrade to the latest OE as part of it - but am wondering if there are new bugs to watch out for with this version.
engineering-rel
5 Posts
695
0
Posted December 30th, 2011 07:00
We applied this firmware upgrade last night.
The upgrade itself completed successfully, but SP failover did not complete in a timely enough fashion to preclude our XenServer High Availability from kicking in and fencing a few of our servers.
EMC Support reiterated the SP failover (i.e. IP addresses normally assigned to one SP being handled by the other SP when the first SP is being upgraded/rebooted, etc.) was supposed to have happened under the time threshold for which we have set our HA timeout, but our HA logs indicate that was not so.
Consequently, EMC Support collected log files and service information, and promsised to get back to us with the maximum amount of time an SP failover can possibly take, so we can adjust our HA failover timeout accordingly.
Hope that helps others...