![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
Got a Zpool with 5 mirrored pairs of 2tb SATA disks and 2 hot spares.
No errors, all working, but a bit slower than expected. Anyway when checking out the iostats we see the following:
See that sd1 is always busier (& slower service time) than the others? And it's always like this under any load. Database, VMWare, sequential transfers.
No SMART warnings or any errors as far as I can tell. It's not normal ZFS behaviour is it?
We have 2 hot spares in the pool (sd12 & sd13). Is there any risk-free way to soft-fail the sd1 disk and set it to be a hot spare instead of an active disk?
No errors, all working, but a bit slower than expected. Anyway when checking out the iostats we see the following:
Code:
extended device statistics
device r/s w/s Mr/s Mw/s wait actv svc_t %w %b
cmdk0 1.5 15.9 0.0 0.0 0.0 0.0 1.4 0 2
cmdk1 2.7 15.5 0.0 0.0 0.0 0.0 1.0 0 1
[COLOR="Pink"][B]sd1 15.7 20.3 0.6 1.5 0.0 1.1 31.9 0 47[/B][/COLOR]
sd2 18.1 20.4 0.7 1.5 0.0 0.8 22.0 0 35
sd3 17.2 22.8 0.5 1.4 0.0 0.6 15.9 0 31
sd4 16.7 23.2 0.9 1.4 0.0 0.9 21.4 0 39
sd5 20.6 25.1 0.7 1.6 0.0 0.7 15.0 0 33
sd6 18.5 24.6 0.7 1.6 0.0 0.8 18.3 0 36
sd7 15.2 20.5 0.7 1.6 0.0 0.7 18.9 0 32
sd9 14.4 20.6 0.7 1.6 0.0 0.7 18.6 0 29
sd10 21.9 16.2 0.7 1.5 0.0 0.6 14.7 0 24
sd11 20.0 16.3 0.8 1.5 0.0 0.6 16.1 0 26
sd12 0.1 0.0 0.0 0.0 0.0 0.0 0.5 0 0
sd13 0.1 0.0 0.0 0.0 0.0 0.0 0.5 0 0
extended device statistics
device r/s w/s Mr/s Mw/s wait actv svc_t %w %b
cmdk0 3.6 13.9 0.1 0.0 0.0 0.0 1.4 0 2
cmdk1 4.8 13.6 0.1 0.0 0.0 0.0 1.5 0 2
[COLOR="Pink"][B]sd1 25.4 32.7 0.8 1.7 0.0 1.4 23.5 0 63[/B][/COLOR]
sd2 21.6 32.5 0.8 1.7 0.0 1.0 17.8 0 43
sd3 25.5 30.2 0.7 1.7 0.0 0.8 14.8 0 39
sd4 20.4 29.4 0.6 1.7 0.0 1.0 20.8 0 46
sd5 18.1 22.0 0.8 1.5 0.0 0.7 17.0 0 30
sd6 20.2 22.5 0.6 1.5 0.0 0.7 17.1 0 32
sd7 20.9 19.7 0.8 1.5 0.0 0.7 16.3 0 31
sd9 24.1 19.8 0.7 1.5 0.0 0.7 17.0 0 33
sd10 22.6 26.1 0.8 1.5 0.0 0.8 16.0 0 36
sd11 30.3 25.9 0.8 1.5 0.0 0.8 14.6 0 36
sd12 1.1 0.4 0.0 0.0 0.0 0.0 0.0 0 0
sd13 1.1 0.4 0.0 0.0 0.0 0.0 0.0 0 0
extended device statistics
device r/s w/s Mr/s Mw/s wait actv svc_t %w %b
cmdk0 3.6 20.4 0.1 0.2 0.0 0.0 1.6 0 3
cmdk1 5.1 20.2 0.1 0.2 0.0 0.0 1.4 0 3
[COLOR="Pink"][B]sd1 17.9 22.2 0.8 1.4 0.0 1.1 27.8 0 47[/B][/COLOR]
sd2 17.7 22.1 0.7 1.4 0.0 0.7 18.8 0 32
sd3 13.7 23.0 0.4 1.6 0.0 0.7 18.7 0 31
sd4 22.2 22.9 0.9 1.6 0.0 0.9 20.0 0 40
sd5 17.7 22.5 0.7 1.6 0.0 0.7 17.7 0 33
sd6 24.9 22.2 0.9 1.6 0.0 0.7 15.2 0 33
sd7 20.6 21.1 0.8 1.6 0.0 0.7 17.5 0 32
sd9 21.5 21.0 0.7 1.6 0.0 0.7 17.4 0 34
sd10 21.8 22.7 0.9 1.5 0.0 0.8 16.9 0 34
sd11 26.3 23.2 0.7 1.5 0.0 0.7 14.5 0 32
sd12 1.1 0.4 0.0 0.0 0.0 0.0 0.0 0 0
sd13 1.1 0.4 0.0 0.0 0.0 0.0 0.0 0 0See that sd1 is always busier (& slower service time) than the others? And it's always like this under any load. Database, VMWare, sequential transfers.
No SMART warnings or any errors as far as I can tell. It's not normal ZFS behaviour is it?
We have 2 hot spares in the pool (sd12 & sd13). Is there any risk-free way to soft-fail the sd1 disk and set it to be a hot spare instead of an active disk?