Any success stories for HAST + ZFS?

Freddie Cash fjwcash at gmail.com
Mon Apr 11 18:26:17 UTC 2011


On Sun, Apr 10, 2011 at 12:36 PM, Mikolaj Golub <trociny at freebsd.org> wrote:
> On Mon, 4 Apr 2011 11:08:16 -0700 Freddie Cash wrote:
>  FC> Once the deadlock patches above are MFC'd to -STABLE, I can do an
>  FC> upgrade cycle and test them.
>
> Committed to STABLE.

Updated src tree to r220537.  Recompiled world, kernel, etc.
Installed world, kernel, etc.  ZFSv28 patch was not affected.

Everything is detected correctly, everything comes up correctly.  See
a new option (reload) in the RC script for hast.

Can create/change role for 24 hast devices simultaneously.

Can switch between master/slave modes.

Have 5 rsyncs running in parallel without any issues, transferring
80-120 Mbps over the network (just under 100 Mbps seems to be the
average right now).

Switching roles while the rsyncs are running succeeds without
deadlocking (obviously, rsync complains a whole bunch while the switch
happens as the pool disappears out from underneath it, but it picks up
again when the pool is back in place).

Hitting the reset switch on the box while the rsyncs are running
doesn't affect the hast devices or the pool, beyond losing the last 5
seconds of writes.

It's only been a couple of hours of testing and hammering, but so far
things are much more stable/performant than before.

Anything else I should test?

-- 
Freddie Cash
fjwcash at gmail.com


More information about the freebsd-current mailing list