You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
For testing i am using a zpool with 400 vdevs (files) and 5 spares, i replaced 3 vdevs with spares triggering resilvering.
While resilvering with a replaced spare, the identifier for the spare in use/replaced seem to be wrong numbered.
Suddenly there is a spare-18 and a spare-19 and a spare-0. For my feeling there should only be spare-0 through spare-5 if no more spares are defined.
The code here is working as intended but the numbering may be non-obvious so let me explain. The suffix appended to interior vdevs (mirror, raidz, spare, etc) is the child number in the vdev parent. It has nothing to do with the specific spare which was used. Notice that in your example above the number always reflects the location in parent raidz3 vdev. Here's a simpler example.
pool: tank
state: ONLINE
scan: resilvered 40K in 0h0m with 0 errors on Fri Mar 1 13:58:30 2013
config:
NAME STATE READ WRITE CKSUM
tank ONLINE 0 0 0
v0 ONLINE 0 0 0
v1 ONLINE 0 0 0
spare-2 ONLINE 0 0 0
v2 ONLINE 0 0 0
v3 ONLINE 0 0 0
spares
v3 INUSE currently in use
I'm closing this issue because this is the expected behavior and must remain this way to be consistent with the other implementations.
For testing i am using a zpool with 400 vdevs (files) and 5 spares, i replaced 3 vdevs with spares triggering resilvering.
While resilvering with a replaced spare, the identifier for the spare in use/replaced seem to be wrong numbered.
Suddenly there is a spare-18 and a spare-19 and a spare-0. For my feeling there should only be spare-0 through spare-5 if no more spares are defined.
The text was updated successfully, but these errors were encountered: