Strange behaviour when my gmirror raid lost one of the disks

To begin with, I really like the gmirror software raid feature in Freebsd. It has helped me to keep my server and my sistes online for a very long time without any large interruption. Since my last strange gmirror problem I haven’t had any problem until now. I checked my log and found out that my raid was DEGRADED, there were only one disk in my raid setup.

The error message was the following. Looks like some kind of disk failure…

Dec 21 22:30:14 kernel: ad6: TIMEOUT - WRITE_DMA48 retrying (1 retry left) LBA=693151663
Dec 21 22:30:24 kernel: ad6: TIMEOUT - WRITE_DMA48 retrying (0 retries left) LBA=693151663
Dec 21 22:30:24 kernel: ad6: FAILURE - WRITE_DMA48 timed out LBA=693151663
Dec 21 22:30:24 kernel: GEOM_MIRROR: Request failed (error=5). ad6[WRITE(offset=354893651456, length=114688)]
Dec 21 22:30:24 kernel: GEOM_MIRROR: Device gm0: provider ad6 disconnected.

However the disk was there and it was ok and I didn’t know why it had been disconnected. I tried the following set of commands:

root@server ~>gmirror deactivate -v gm0 ad6
gmirror: No such provider: ad6.
root@mysql ~> gmirror forget -v gm0
Done.
root@mysql ~> gmirror insert -v gm0 ad6
Done.
root@mysql ~> gmirror status
      Name    Status  Components
mirror/gm0  DEGRADED  ad4
                      ad6 (0%)

Then it started rebuilding at once, very strange. And it works as before again! Maybe this was a warning sign that the disk is going bad or was it some kind of glitch?

Kommentera

E-postadressen publiceras inte. Obligatoriska fält är märkta *

Denna webbplats använder Akismet för att minska skräppost. Lär dig hur din kommentardata bearbetas.