[zfs-discuss] Please help to fix zpool.

Arman Khalatyan arm2arm at gmail.com
Wed Oct 16 07:05:26 EDT 2013


I am using this pool as a mirror of the lustre file system. I just tested
with RAIDZ3,RAIDZ2 and striped. the most performance I got with this
configuration. Checking 9TB with ~500K fileswas taking almost day which is
not acceptable.
The Part of the raid is used for Backuppc, part for rsync. The rsync data
is stored with zfs set compression=lzj. Whole pool is backuped every month
to the TAPE. If zfs fails at least I will have one working copy from
TAPE(worst case a month old).
Second issue probably I should distribute backups on different zfs/hosts.
Looks like on parallel IO time ZFS is CPU hungry.
a.


On Wed, Oct 16, 2013 at 8:34 AM, Micky Martin <mickylmartin at gmail.com>wrote:

>
>
> On Wednesday, October 16, 2013 1:28:29 AM UTC+5, Gregor Kopka wrote:
>>
>> Micky already quoted me, so only this comment left:
>>
>
> :)
>
>
>> This backup system will be toast the moment one block with data isn't
>> readable, there is no redundancy in the pool.
>>
>
> Good catch. That's logical.
>
>  To unsubscribe from this group and stop receiving emails from it, send an
> email to zfs-discuss+unsubscribe at zfsonlinux.org.
>

To unsubscribe from this group and stop receiving emails from it, send an email to zfs-discuss+unsubscribe at zfsonlinux.org.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://list.zfsonlinux.org/pipermail/zfs-discuss/attachments/20131016/4194de56/attachment.html>


More information about the zfs-discuss mailing list