[Linux-HA] nfs-kernel-server in heartbeat 2

[Andrew Beekhof]

On 7/8/07, Adrian Overbury <adrian at inomial.com> wrote:
>
> It's Heartbeat 2.0.2.  I know there's a more recent one (several, in
> fact) but I can't use them in this environment.  This is the Ubuntu
> package for Dapper Drake, and my boss isn't willing for us to use a
> self-compiled or backported version, so 2.0.2 is what I've got to
> work with.



Then your boss is balmy.

I have never understood the logic that, given the same sources, packages
from entityX are any better than entityY.  Especially when one of those
groups has at least 3 people working on heartbeat full-time and the other,
in all likelihood, just checks to see everything compiles.

Consider also where the support is coming from.

>From the logs I can see you're use the crm and as the person responsible for
the majority of that code, there is no way in hell I would use 2.0.2 in
preference to any other version that we have put out in the 1.5 years since
2.0.2 was released.

Many many bugs, including the one you're reporting (i recognize it) have
been fixed in that time... thats why we put out new releases.


Andrew

Regards,
>
> Adrian Overbury
>
> Andrew Beekhof wrote:
>
> > what version of heartbeat?  try seeing if there is a more recent one
> > On 7/6/07, Adrian Overbury <adrian at inomial.com> wrote:
> >
> >> Hi list-denizens.
> >>
> >> I'm sorry if this has been posted to the list before, but my search
> >> didn't turn up anything that seemed applicable after the first few
> >> that I found.
> >>
> >> I've got a two node cluster, nfs1 and nfs2.  It's running drbd 0.7,
> >> drbdlinks, an IPaddr resource, a filesystem resource, and an lsb nfs-
> >> kernel-server resource.  I've got co-location constraints that keep
> >> everything on the same node with the drbd resource, and ordering
> >> constraints like so:
> >>
> >> drbd -> Filesystem -> drbdlinks -> IPaddr -> nfs-kernel-server
> >>
> >> Now, this is (mostly) working.  It correctly discovers that drbd is
> >> already in primary state, mounts the filesystem, starts drbdlinks,
> >> creates the interface alias... and then does nothing else.  nfs-
> >> kernel-server doesn't come up unless I start it myself.  nfs2 wins
> >> the crm election process, so in its syslog, I find a number of these:
> >>
> >> tengine: [22991]: info: mask(tengine.c:cib_action_updated):
> >> Initiating action 5: start nfs_kernel_server on nfs1
> >>
> >> And nfs1 soon after shows:
> >>
> >> crmd: [7092]: WARN: lrm_get_rsc(653): got a return code HA_FAIL from
> >> a reply message of getrsc with function get_ret_from_msg.
> >> crmd: [7092]: ERROR: mask(lrm.c:do_lrm_rsc_op): Triggered dev assert
> >> at lrm.c:746 : type != NULL
> >>
> >> A bit of Googling led me to a bug thread about something similar,
> >> which basically said that the nfs-kernel-server init script wasn't
> >> lsb-compliant because it didn't contain a status function.  I patched
> >> the init script to fix this, but nfs-kernel-server still isn't being
> >> started or managed by heartbeat.  Can anyone tell me where I'm going
> >> wrong/what else I need to try/what other information I need to
> >> provide in order for help to be given?  Please?
> >>
> >> Regards,
> >>
> >> Adrian Overbury
> >> Inomial Pty Ltd
> >> adrian at inomial.com
> >>
> >>
> >>
> >> _______________________________________________
> >> 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
> >>
> >>
> > _______________________________________________
> > 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
>
>
>
> _______________________________________________
> 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
>