topic Re: Failed MSA Controller cannot remove in MSA Storage

Failed MSA Controller cannot remove

pythagoras — Fri, 22 Feb 2019 16:34:43 GMT

I have a trustworthy P2000 G3 SAN which has had a controller fail. I have no intention of ever replacing it but i cant see anywhere on how to disable it. My intention is to use the SAN until it dies.

Controller B keeps moaning A is missing and showing as degraded.

Anyone know how to tell B to never look for A anymore?

Thanks

web or CLI commands would be great.

Re: Failed MSA Controller cannot remove

Torsten. — Fri, 22 Feb 2019 17:15:09 GMT

There is a CLI command to set single controller, the remaining controller should be A.

Re: Failed MSA Controller cannot remove

Shawn_K — Fri, 22 Feb 2019 17:50:22 GMT

Hello,

If one of the controllers has failed it should be replaced in order to provide redundancy. Otherwise a single controller becomes a single point of failure. However, if you decide to run in single controller mode, my suggestion would be to do the following.

1. Remove the failed controller A - you do not want it plugged into the backplane. Otherwise the system knows the controller is there and will continue to alert you on the failed controller. Your health status will also show as degraded.

2. Set single controller mode - using the CLI issue the following command: #set advanced-settings single-controller

Once the controller is set to single controller mode you should observe all the vdisks as owned by the B controller. Optimally you should properly power down the system and move the good working B controller to the top slot. You will also need to check your host mappings so they are properly mapped to the array and not looking for the old controller host ports.

Cheers,
Shawn

I work for Hewlett Packard Enterprise. The comments in this post are my own and do not represent an official reply from HPE. No warranty or guarantees of any kind are expressed in my reply.

Re: Failed MSA Controller cannot remove

pythagoras — Mon, 25 Feb 2019 10:17:40 GMT

Wow that worked. No reboot either. Issued the command via CLI, refreshed the web page, all green again.

I'll move the controller to the top slot and remap once the servers go back into production.

Thanks alot!!