[zfs-discuss] ARC settings on backup dataset.

Gordan Bobic gordan.bobic at gmail.com
Mon Jan 4 15:30:41 EST 2016


Unless you have millions of small files, you should probably set
primarycache=metadata on the backup data set.

On Mon, Jan 4, 2016 at 7:11 PM, David C. Miller via zfs-discuss <
zfs-discuss at list.zfsonlinux.org> wrote:

> I just setup a new system that has two identical SAS2 JBOD's that I have
> ZOL using in two separate storage pools. One JBOD is the pool for a live
> single dataset which replicates to the second JBOD pool for a hot backup
> every hour. The server has 128GB of ram and I have set the zfs_arc_max to
> only use 64GB. Before I start putting real data on this systems I was
> wondering if ZFS will automatically put data from the backup dataset into
> the ARC? Since the backup dataset does not need to be in the ARC, would it
> be beneficial to set the primarycache property to none on the backup pool
> dataset?
>
> David Miller.
> _______________________________________________
> zfs-discuss mailing list
> zfs-discuss at list.zfsonlinux.org
> http://list.zfsonlinux.org/cgi-bin/mailman/listinfo/zfs-discuss
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://list.zfsonlinux.org/pipermail/zfs-discuss/attachments/20160104/98c23325/attachment.html>


More information about the zfs-discuss mailing list