zfs discuss - Jan 2010 - raidz1 pool import failed with missing slog

In osol 2009.06 - rpool vdev was dying but I was able to do the clean export of
the data pool. The data pool''s zil was on the failed HDD''s
slice as well as slog''s GUID. As the result I have 4 out of 4 raid5
healthy data drives but cannot import zpool to access the data. This is
obviously a disaster for me.

I found two discussions about similar issues:
http://opensolaris.org/jive/thread.jspa?messageID=233666 and
http://opensolaris.org/jive/thread.jspa?messageID=420073
But I don''t think the recipes in these threads can help to import the
inconsistent
pool.

Is there any way to ignore missing ZIL devices during the import?
I expected that this is the case, since
http://www.solarisinternals.com/wiki/index.php/ZFS_Best_Practices_Guide says:
[b]Hybrid Storage Pools (or Pools with SSDs)[/b] 
...
If a separate log device is not mirrored and the device that contains the log
fails, storing log blocks reverts to the storage pool.

If not - can I somehow reassemble the pool using [b]zpool import -D[/b] option,
or do anything else to get my data back?

Please help!
-- 
This message posted from opensolaris.org
-------------- next part --------------
A non-text attachment was scrubbed...
Name: zpool_import.log
Type: application/octet-stream
Size: 522 bytes
Desc: not available
URL:
<http://mail.opensolaris.org/pipermail/zfs-discuss/attachments/20091231/90caca65/attachment.obj>

Hi Yuriy,

A couple of options are:

1. If you can boot this system from a system or DVD that is running
at least build 128, then you can try to import the pool using the
following syntax:

zpool import -F -c <old zpool.cache file> <pool>

The -F option provides a way to rollback the last few transactions.

I''m not sure I understand all the implications of this option because
I haven''t used it myself.

2. If you can provide us access to this system or we can provide some
code bits that could resolve this. Email me separately if you are
interested in pursuing this option.

Thanks,

Cindy

On 12/31/09 19:02, Yuriy Vasylchenko wrote:
> In osol 2009.06 - rpool vdev was dying but I was able to do the clean
export of the data pool. The data pool''s zil was on the failed
HDD''s slice as well as slog''s GUID. As the result I have 4 out
of 4 raid5 healthy data drives but cannot import zpool to access the data. This
is obviously a disaster for me.
> 
> I found two discussions about similar issues:
> http://opensolaris.org/jive/thread.jspa?messageID=233666 and
> http://opensolaris.org/jive/thread.jspa?messageID=420073
> But I don''t think the recipes in these threads can help to import
the inconsistent
> pool.
> 
> Is there any way to ignore missing ZIL devices during the import?
> I expected that this is the case, since
http://www.solarisinternals.com/wiki/index.php/ZFS_Best_Practices_Guide says:
> [b]Hybrid Storage Pools (or Pools with SSDs)[/b] 
> ...
> If a separate log device is not mirrored and the device that contains the
log fails, storing log blocks reverts to the storage pool.
> 
> If not - can I somehow reassemble the pool using [b]zpool import -D[/b]
option, or do anything else to get my data back?
> 
> Please help!
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> zfs-discuss mailing list
> zfs-discuss at opensolaris.org
> http://mail.opensolaris.org/mailman/listinfo/zfs-discuss