On Mon, Mar 29, 2004 at 01:35:02PM -0800, Dave Lowenstein wrote:
> inexplicably this started working. It may have something to do with me
> having to kill -9 heartbeat and ldirectord because /etc/init.d/heartbeat
> restart wasn't responding.
There have been a _lot_ of problems with this in the past.
I believe that 1.2.0 actually works quite well with regards to this.
Though in practice, you don't really restart much heartbeat except
during testing.
> my newest problem is that ipfail doesn't seem to be working. The version
> of heartbeat is 1.0.4. When I pull out the ethernet cable on the master
> director, no messages are sent over the serial cable to the slave and
> nothing fails over.
This may be an iteraction with the nice_failback option.
It needs to be off for ipfail to function.
In 1.2.0 it is possible to get nice_failback like behaviour -
the option has been changed to be called auto_failback - and
ipfail to work together.
[snip]
--
Horms
|