Pool member moved or was relabled during drive replacement.

[fluxnull]

I had a failing drive and I got an identical replacement today.

Code:

 CRITICAL: Dec. 27, 2019, 5:46 a.m. - Device: /dev/da3 [SAT], 1 Currently unreadable (pending) sectors
CRITICAL: Dec. 28, 2019, 10:43 a.m. - Device: /dev/da3 [SAT], ATA error count increased from 0 to 1
CRITICAL: Dec. 28, 2019, 10:14 p.m. - Device: /dev/da3 [SAT], FAILED SMART self-check. BACK UP DATA NOW!
CRITICAL: Dec. 28, 2019, 10:14 p.m. - Device: /dev/da3 [SAT], Failed SMART usage Attribute: 7 Seek_Error_Rate.

Via the web interface I went to the Pools > Cog > Status and set pool member da3 to "offline" as expected the system went to a degraded status. Then I powered down the server and replaced the failing drive with the new one after verifying serial numbers. Powered the server back on I logged in via the GUI. I Went back to Pools > Cog > Status and I saw the new drive hit the ellipsis and selected "Replace" but instead of allowing me to replace da3 it wanted me to replace da5 (the serial number for da5 matches the new drive). I checked smartctl -a /dev/da3 and found that da3 had be replaced with one of the other pool members. I'm guessing this happened due to the reboot ?

I went ahead and started the resilver and as I did I got the following on the console.

Code:

GEOM_MIRROR: Cannot open consumer da0p1 (error=1).
GEOM_MIRROR: Device swap2 destroyed.

I'm currently at 80% of resilver. Does this sound like normal behavior having the da3 auto replace from another pool member and do I need to be concerned about the "Cannot open consumer da0p1" error?

 

[fluxnull]

The resilver looks to be successful. Though da5 is now where da3 was ?

 

[Johnny Fartpants]

da numbers can and do move normally between reboots so you can never rely on them.