hastd Failover with ucarp

Freddie Cash fjwcash at gmail.com
Mon Feb 21 17:38:52 UTC 2011

On Mon, Feb 21, 2011 at 7:55 AM, Christian Vogt
<christian.vogt at haw-hamburg.de> wrote:
> Thanks for the great work, I like this straight-forward FreeBSD a lot from what I experienced untill now. I used the HAST How-To from http://wiki.freebsd.org/HAST and it works perfectly if I use "pkill -USR2 -f 'ucarp -B'" to initiate the failover. The secondary node becomes primary and the carp-interface is switched over to it.
> But if I do a hard shutdown of the primary node it doesn't work, the secondary node doesn't get primary. The ucarp-up script on the secondary node is executed, but it fails because of the still running secondary worker process  (Secondary process for resource test is still running after 30 seconds). Is the secondary process expected to end automatically, when the primary process fails?
> Thanks for your help

Does it work correctly if you use the built-in carp(4) + devd(8)
instead of the port for ucarp?

That will narrow down whether it's an issue with ucarp or with hastd(8).

Personally, I prefer to use the tools that ship with FreeBSD, as it
gives it a more integrated feel.

http://blather.michaelwlucas.com/?p=241 Has some pre-configured
scripts for hooking into devd.conf(5) to glue together carp(4) and
hastd(8).  Works for ZFS and UFS.

Freddie Cash
fjwcash at gmail.com

More information about the freebsd-fs mailing list