Re: zfs backups (was: Re: Best way to simulate a failure)

On Jan 21, 8:29 pm, "DoN. Nichols" <dnich...@xxxxxxxxxxx> wrote:
On 2008-01-21, Casper H.S  *** <Casper....@xxxxxxx> wrote:





Mike <n00s...@xxxxxxxxxxx> writes:

I have read that the STOP-A key sequence can cause Solaris OS file
system corruption.  How would you compare the risk of using STOP-A to
the risk of pulling the power cord?  I want a realistic failover test
but I don't want to destroy the system in the process.  The system
needs to be usable after my test has finished.

Stop-A is no more damaging than pulling th epowercord (and not
as bad for your hardware as cycling power).

Some filesystems types can be corrupted if the system goes away in the
middle of an update.  ZFS does not have that problem.
All filesystems can lose data (without corruption) when power goes out.
(Only data which has not yet been acknowledged by the system
as being committed to stable storage)

        Out of curiosity -- speaking of zfs -- I seem to have a problem
with backing up zfs filesystems on Exabyte tapes using amanda.  Here is
a part of the daily status e-mail from a recent backup run:

 ======================================================================
HOSTNAME     DISK        L ORIG-kB  OUT-kB  COMP%  MMM:SS   KB/s MMM:SS   KB/s
-------------------------- ------------------------------------- -------------

        [ ... ]

burke        --usr-share 1     230      64   27.8    0:05    7.7   0:02   35.7
burke        -e-p/home-1 0      10      32  320.0    0:00    4.0   0:02   17.8
burke        -e-p/home-4 0      10      32  320.0    0:00    4.2   0:02   19.1
burke        -e-p/home-6 0      10      32  320.0    0:00    4.0   0:02   17.4
burke        -/workplace 1   22970    2080    9.1    1:33   22.3   0:05  398.0
burke        /opt        1   30640    3072   10.0    0:52   58.8   0:06  530.8
burke        /package02  0  430960  175744   40.8    1:45 1668.9   0:19 9147.6
burke        /photos/CSU 0      10      32  320.0    0:00    4.1   0:02   17.8
burke        /photos/DoN 0      10      32  320.0    0:00    4.1   0:02   17.8
burke        -os/Dolores 0      10      32  320.0    0:00    3.9   0:06    5.2
burke        /src        0 2580440 1576576   61.1   12:20 2130.0   2:23 11053.3

        [ ... ]

 ======================================================================

        The size of the backup (10 kB) is the same whether it is a
level-0 or a level-1 backup.  FWIW, the backups are being done with
gtar, not ufsdump and the same works fine for non-zfs filesystems.  (The
zfs ones are the "-e-p/home-?" ones and the /photos/* and "-os/Dolores"
ones.  The first three filesystems are on one zfs pool, and the next
three are on another.  Both pools are healthy according to "zpool
status" and the filesystems show no problems with "zfs list"

        If it matters, the OS is Solaris 10 U3, the hardware is a Sun
Fire 280R, and both pools are built from 18GB drives -- SCA drives in a
D-1000 for one, and 68-pin drives in a Kingston rack-mount JBOD in the
other -- both with a hot spare drive sitting there.

What version of Amanda are you using?


        Has anyone else experienced a similar problem?

        Thanks,
                DoN.

--
 Email:   <dnich...@xxxxxxxxxxx>   | Voice (all times): (703) 938-4564
        (too) near Washington D.C. |http://www.d-and-d.com/dnichols/DoN.html
           --- Black Holes are where God is dividing by zero ---- Hide quoted text -

- Show quoted text -

.