We have always manually upgraded everything in the past, but figured we would give the zero downtime cluster upgrade a chance. We have the following infrastructure:2 Virtual MC's2 Physical MM's (7210's)160 AP's mostly 305's and 505'sWe were on 220.127.116.11 and upgraded to 18.104.22.168Followed all the steps outlined and the cluster upgrade proceeded normally through both controllers. However, at some point the AP image preload failed, and then it did the upgrade anyway, so that all AP's were rebooted virtually simultaneously. I am just starting to dive into the logs about this, but if anyone knows where I should look to get more information, I'd appreciate it. The GUI does not provide any error details, and the logs can be cumbersome. This is just one of many, many hats I wear, and I don't use the CLI much at all. Found additional commands to run and their output is attached. Namely:
(host)#show lc-cluster cluster1 upgrade status verbose
(host)#show lc-cluster cluster1 upgrade stats
"Attached a screenshot of the "zero downtime" upgrade results. ThanksEdit: adding additional information and screenshots.
Do you have these being monitored in airwave? Any messages in the logs on airwave? I would assume the messages show show up in the system logs.
A good thought. Unfortunately all I see in the logs for the appropriate timeframe is this
Are those logs for the AP only. Do you see anything in the logs of the Controllers or Mobility Conductor?
I checked both of those and only received very normal and basic events for all MM/MC. I went into the MC's where I kicked off the zero downtime upgrade and do not see anything there either.Next time we need to do an upgrade I'll test it again with a smaller batch of AP's first and see if we run into the same issue.
© Copyright 2023 Hewlett Packard Enterprise Development LPAll Rights Reserved.