Consistent ZFS freeze while replacing.

[Attila Nagy]

Hi,

On 04/15/11 06:30, Pawel Tyll wrote:
> Hi list,
>
> I'm running FreeBSD 8.2-STABLE #0: Wed Apr 13 03:03:08 CEST 2011
> (amd64) along with
> http://people.freebsd.org/~mm/patches/zfs/v28/stable-8-zfsv28-20110317.patch.xz
>
> I replaced one device with another, and when replace was finishing,
> pool froze. After two reboots (first one resulted in freeze while
> importing the pool), system came up with pool in following state:
>
>    pool: storage
>   state: ONLINE
>   scan: resilvered 689G in 19h35m with 0 errors on Thu Apr 14 23:37:10 2011
> config:
>
>          NAME                   STATE     READ WRITE CKSUM
>          storage                ONLINE       0     0     0
>            raidz1-0             ONLINE       0     0     0
>              da16               ONLINE       0     0     0
>              da17               ONLINE       0     0     0
>              da18               ONLINE       0     0     0
>              da19               ONLINE       0     0     0
>              da20               ONLINE       0     0     0
>              da21               ONLINE       0     0     0
>              da22               ONLINE       0     0     0
>              da23               ONLINE       0     0     0
>            raidz1-1             ONLINE       0     0     0
>              gpt/ST-5XW25KAR    ONLINE       0     0     0
>              da6                ONLINE       0     0     0
>              replacing-2        ONLINE       0     0     0
>                da1              ONLINE       0     0     0
>                gpt/ST-5XW24JDN  ONLINE       0     0     0
>              gpt/ST-5XW0PKCM    ONLINE       0     0     0
>              da5                ONLINE       0     0     0
>              da3                ONLINE       0     0     0
>              da7                ONLINE       0     0     0
>              da4                ONLINE       0     0     0
>            raidz1-3             ONLINE       0     0     0
>              gpt/ST-5XW10ZKX    ONLINE       0     0     0
>              gpt/ST-5XW1A9L6    ONLINE       0     0     0
>              gpt/ST-5XW1BNGJ    ONLINE       0     0     0
>              gpt/ST-5XW1ALPW    ONLINE       0     0     0
>              gpt/ST-5XW11AQ5    ONLINE       0     0     0
>              gpt/ST-5XW1330J    ONLINE       0     0     0
>              gpt/ST-5XW18MD4    ONLINE       0     0     0
>              gpt/ST-5XW10VB0    ONLINE       0     0     0
>          logs
>            mirror-2             ONLINE       0     0     0
>              gpt/ssd1-log       ONLINE       0     0     0
>              gpt/ssd2-log       ONLINE       0     0     0
>          cache
>            gpt/ssd1-cache       ONLINE       0     0     0
>            gpt/ssd2-cache       ONLINE       0     0     0
>
> This happened to me twice already and I have five more drives to
> replace. Any ideas?
I'm not sure it's the same, but I can consistently reach this situation 
when replacing while a scrub is in progress (or replacing more than one 
drive).
So far I could only get a working pool in these cases, if I stopped (in 
single user, even the pool import freezes every operation on zf) scrub, 
or replace, so only one left.