[Linux-HA] Colocation confusion

[Andrew Beekhof]

On 6/18/07, Max Hofer <max.hofer at apus.co.at> wrote:
> On Monday 18 June 2007, Andrew Beekhof wrote:
> > On 6/17/07, Christian Rishøj <christian at rishoj.net> wrote:
> > > On 6/16/07, Andrew Beekhof <beekhof at gmail.com> wrote:
> > > > > I am *assuming* clones works with colocation, because I can do:
> > > >
> > > > you can. but later versions have gotten better in that respect.
> > > >
> > > > I'm guessing you're using 2.0.8 right?
> > >
> > > No, 2.0.7-2.
> > >
> > >
> > > > you might want to try the packages available at:
> > > >     http://software.opensuse.org/download/server:/ha-clustering/
> > >
> > > Running Ubuntu, so I'm afraid the Suse packages is not an option.
> >
> > if you clicked the link you'd see there are also fedora, ubuntu,
> > debian and mandriva packages there
>
> What changeset (tag) is this version?

SLE10-SP1 aka 2.1.0 (see: http://hg.linux-ha.org/dev/tags)

plus a bunch of debian packaging changes from horms

> Horms: will there be a backport for Debian Sarge? (if you need help let me
> know)
>
> >
> > > Would you recommend compiling from source or CVS?
> > >
> > >
> > > > since 2.0.8 came out, a colocation constraints such as:
> > > >     <rsc_collocate from=A to=B score=INFINITY/>
> > > >
> > > > should be read as:
> > > >      figure out where B is/goes, and then put A there too.
> > > >
> > > > so in your case, from=... should be your IP address and to=... the
> > > > webserver and database clones
> > >
> > > Aha! I have been interpreting the colocation in the opposite way.
> > > Would you happen to know which behavior applies to 2.0.7.x?
> >
> > i don't believe 2.0.7 supports the behavior you're after
> > _______________________________________________
> > Linux-HA mailing list
> > Linux-HA at lists.linux-ha.org
> > http://lists.linux-ha.org/mailman/listinfo/linux-ha
> > See also: http://linux-ha.org/ReportingProblems
> >
>
>