SOLVED CRITICAL: The following multipaths are not optimal: disk1

[zamana]

I guess for some reason your setup was identified as multipath. To solve present situation you may try to destroy multipath instance first with `gmultipath destroy disk1` and then recover secondary GPT partition table on disk overwritten by gmultipath with `gpart recover adaX`. Though it doesn't explain how have you got into this situation.

I know, I know... three to six years old thread but you know what? You saved my day.

Thanks.

 

[Rabinovitch]

First JBOD we had has been connected to HBA with 4 cables. Recently we used 2 of that 4 cables to connect an additional JBOD to that HBA. So each JBOD is now connected to HBA with 2 cables, and we have 108 errors about "Multipath multipath/diskXY connection is not optimal. Please check disk cables."

Is it safe to destroy these multipaths?
TrueNAS-12.0-U8

 

[mav@]

If you are not going to use that cabling ever again -- sure, you can destroy the multipaths, but you have to do that from command line after exporting the pool. You'll also need to do `gpart recover` after that to move the second GPT label to the end of the physical disk (note that after that you won't be able to return to using multipath again).