Second drive suddenly FAILED state while replacing another drive in Z2 pool?

[Robert Townsend]

I just set a bad drive at ad5 to offline, installed a replacement of same model, and 'replace' the bad drive. The system is currently resilvering for about 3 more hours. In the meantime the pool is 'degraded'. [BTW I had the drives spin down at 10 - probably a bad idea - now they are set always on]

Two issues, one probably easy:

ad7, and mirrored SSD is offline, I suspect I knocked a cable loose. I expect I'll have to reopen the box after the silvering is done
ad1, now is in FAILED state

Is it merely coincidence that ad1 is FAILED? Anything to do other than replace it? I saw one post about running a scrub might clear the issue.

 

[jgreco]

You'll want to check the SMART data to see if the drive has some insights to share with you. FAILED merely means that there was a crapton of I/O failures and ZFS marked it as such. This comes in part from the days before SMART tests, when we were reliant on ZFS itself to identify how screwed up a drive was.

 

[Davvo]

Resilvering heavily stresses your drives.
This is why RAIDZ1 is considered risky. Second parity drive saved your pool.

 

[Etorix]

@Robert Townsend To help us help you, please describe your hardware (per forum rules)
and provide the output of zpool status and smartctl -a /dev/ada7 (as well as any other suspicious drive/all drives), formatted within [ CODE ] tags for readability. (Use a SSH session rather than the shell from web UI, which has longstanding issues with copy and paste.)