On Tue, Dec 21, 2010 at 11:43:23AM -0600, Dimitri Maziuk wrote:
> Dimitri Maziuk wrote:
> >   - doesn't seem to do anything other than print
> > hb_standby[6096]: Going standby [all].
> > 
> > Neither does hb_takeover on the other node. They aren't even logging 
> > anything.
> > 
> > Is this the expected behaviour (this is r1-style setup w/ haresources)?
> The winning answer is "Standby mode only implemented when nice_failback 
> on" -- logged by 2.1.4, but not by 3.0.3. Also undocumented anywhere 
> that I know of.

Uhm, no, that message is still there.
The code in question has not been changed.

Though "nice_failback" was deprecated in 2.x already,
and has been replaced by "auto_failback" (on, off) instead.

And regardless of whether auto_failback is on or off, "nice_failback" is
considered to be on (it's always nice, wether it's on of off; only
"legacy" is not nice ;-> ...  because that's confusing that "nice" name
was deprecated a long time ago).

As auto_failback unfortunately still defaults to "legacy",
I'd expect 3.0.3 to log just the same as 2.1.whatever.
Unless of course you have auto_failback explicitly configured
(which you should).
Then standby would be expected to work.

Wether it does what you want it to do depends on a number of other
things, internal state tracking for one thing,
but also on the quality of the used resource scripts,
and on wether or not haresources is identical on both nodes.

: Lars Ellenberg
: LINBIT | Your Way to High Availability
: DRBD/HA support and consulting http://www.linbit.com

DRBD® and LINBIT® are registered trademarks of LINBIT, Austria.
Linux-HA mailing list
See also: http://linux-ha.org/ReportingProblems

Reply via email to