Eric Haycraft wrote:
Since no one seems to believe that you can expand a raidz pool, I have attached the following output from solaris 11/06 showing me doing just that. The first expanision is with like sized disks, and the second expansion is with larger disks. I realize that the documentation only has examples using mirrors, but raidz and raid2z are fully supported for adding disk space.

# mkfile 64m 1 2 3 4 5 6
# mkfile 128m 01 02 03

# zpool create testpool raidz  `pwd`/1 `pwd`/2 `pwd`/3
# zpool list testpool
NAME                    SIZE    USED   AVAIL    CAP  HEALTH     ALTROOT
testpool                178M    100K    178M     0%  ONLINE     -

# zpool add testpool raidz  `pwd`/4  `pwd`/5  `pwd`/6
#  zpool list testpool
NAME                    SIZE    USED   AVAIL    CAP  HEALTH     ALTROOT
testpool                356M    305K    356M     0%  ONLINE     -

# zpool add testpool raidz  `pwd`/01 `pwd`/02 `pwd`/03
# zpool list testpool
NAME                    SIZE    USED   AVAIL    CAP  HEALTH     ALTROOT
testpool                726M    475K    726M     0%  ONLINE     -
#
This message posted from opensolaris.org
_______________________________________________
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
No, you are NOT adding to a RAIDZ vdev. You are adding to a ZPOOL. Your new pool has 2 stripes of RAIDZ vdevs.

The final "testpool" zpool above consists of 3 RAIDZ vdevs, each with 3 members.

While expanding a zpool in the way you've show is useful, it has nowhere near the flexibility of simply adding single disks to existing RAIDZ vdevs, which was the original desire expressed. This conversation has been had several times now (take a look in the archives around Jan for the last time it came up).

To make myself clear, what we're talking about is some way to move from:

zpool create tank raidz d1 d2 d3

to the equivalent of

zpool create tank raidz d1 d2 d3 d4 d5 d6

in increments. Right now, that's not possible.


Also, I believe there is a bug in zfs list for Solaris 10 U3's zfs - I have to check it out. Because:

# cd /tmp
# mkfile 64m 1 2 3 4 5 6 7 8
# zpool create testpool1 raidz /tmp/1 /tmp/2 /tmp/3 /tmp/4 raidz /tmp/5 /tmp/6 /tmp/7 /tmp/8
# zfs list testpool1
NAME                   USED  AVAIL  REFER  MOUNTPOINT
testpool1               131K   444M    49K  /testpool1
# zpool destroy testpool1
# zpool create testpool2 raidz /tmp/1 /tmp/2 /tmp/3 /tmp/4 /tmp/5 /tmp/6 /tmp/7 /tmp/8
# zfs list testpool2
NAME                   USED  AVAIL  REFER  MOUNTPOINT
testpool2                77K   444M     3K  /testpool2


Which is _clearly_ incorrect. "testpool1" should have a size of approximately 6*64m = 384, not 7*64m = 448m as in "testpool2".

--
Erik Trimble
Java System Support
Mailstop:  usca22-123
Phone:  x17195
Santa Clara, CA
Timezone: US/Pacific (GMT-0800)

_______________________________________________
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Reply via email to