Hi,
On Mon, Sep 19, 2011 at 11:47:50AM -0400, Ken Dechick wrote:
> Ahh, OK now we are getting somewhere! Dejan you are absolutely correct, logd
> is in the cluster-glue. Installed the source rpm and found this
> in the SPEC file:
>
> # Don't package pieces we wont support
> rm -rf %{buildroot}/%{_libdir}/stonith
> rm -f %{buildroot}/%{_libdir}/heartbeat/ha_logd
> rm -f %{buildroot}/%{_sbindir}/ha_logger
> rm -f %{buildroot}/%{_sbindir}/hb_report
> rm -f %{buildroot}/%{_sbindir}/meatclient
> rm -f %{buildroot}/%{_sbindir}/stonith
> rm -f %{buildroot}/%{_sbindir}/sbd
> rm -f %{buildroot}/%{_datadir}/cluster-glue/ha_cf_support.sh
> rm -f %{buildroot}/%{_datadir}/cluster-glue/openais_conf_support.sh
> rm -f %{buildroot}/%{_datadir}/cluster-glue/utillib.sh
> rm -f %{buildroot}/%{_datadir}/cluster-glue/combine-logs.pl
> rm -f %{buildroot}/%{_datadir}/cluster-glue/ha_log.sh
> rm -rf %{buildroot}/%{_datadir}/doc/cluster-glue/stonith
> rm -rf %{buildroot}/%{_mandir}/man1/
> rm -rf %{buildroot}/%{_mandir}/man8/meatclient.8
> rm -rf %{buildroot}/%{_mandir}/man8/ha_logd.8
> rm -rf %{buildroot}/%{_mandir}/man8/stonith.8
> rm -f %{buildroot}/%{_sysconfdir}/init.d/logd
>
> That explains the STONITH and logd issues quite well, wish there were deatils
> on why though ..
>
> So for some reason logd is not supported under EL6. I wonder why. Can a dev
> help me out here? Why would logd not be supported? Heartbeat is complaining
> that it needs to have
I guess that's within the distribution's prerogative, to support
or not to support parts of software. Whether users are happy
with that decision is another matter.
As for logd, I'd recommend using it, but it is optional. If
components are complaining that they cannot connect to it and it
is not available, then you can configure Heartbeat differently.
IIRC, the ha.cf directive is called use_logd.
Thanks,
Dejan
> logd running. Do I need to consider migrating over to corosync? HA is so
> complicated that I would prefer to stick with what I know rather than start
> fresh with a new message
> layer - been using heartbeat/pacemaker for over 2 years now.
>
> Until I hear more about this issue I am going to make some changes to the
> SPEC file and try rebuilding the cluster-glue myself to include logd. Wish me
> luck,
> I will continue to report my results.
>
>
> Hmm, just came across this when searching bug 553386 :
>
> Description of problem:
>
> This is a list of issues for the logd init script when compared with the
> criteria in https://fedoraproject.org/wiki/Packaging/SysVInitScript
> Version-Release number of selected component (if applicable):
> cluster-glue-1.0-0.11.b79635605337.hg.fc12.i686
>
>
> * init script name conflicts with logd provided by upstart package
> * man page for "logd" is for upstart's logd
> * ha_logd is the man page for cluster-glue's logd
> * suggest renaming init script to "ha_logd"
>
> * logd init script is not added during package installation
>
> * init script is installed to the wrong location. It should be installed to
> /etc/rc.d/init.d instead of /etc/init.d
>
> [root@morph-03 init.d]# rpm -ql cluster-glue | grep init
> /etc/init.d/logd
>
> * missing required actions:
> - reload
> - force-reload
> - condrestart | try-restart
>
> * wrong exit status from status action when service is stopped. Got 1,
> expected 3.
>
> * subsystem locks are not used
>
> * /etc/rc.d/init.d/functions is not used for standard reporting
>
> * wrong exit status for usage and invalid arguments. Got 1, expected 2.
>
>
>
> Andrew Beekhof , can you chime in here? Running EL6 how should I be
> configuring my ha logging if we aren't using logd anymore? Really confused!
> My system logs are filling
> up with this over and over again:
>
> Sep 19 11:45:03 demoHA1 attrd_updater: [6830]: WARN: Initializing connection
> to logging daemon failed. Logging daemon may not be running
> Sep 19 11:44:46 demoHA2 crm_attribute: [16578]: WARN: Initializing connection
> to logging daemon failed. Logging daemon may not be running
>
> P.S. - I double posted the firs email I sent on this iss this morning -
> sincere apologies for that. I messed up my left arm and have a splint on my
> wrist - very difficult
> to adjust to using one hand when your job involves sitting here at a PC and
> typing all day!
>
> -Thanks
>
>
> >> For quite some time logd has been part of the cluster-glue
> >> package. It's the one containing the clplumbing libraries and
> >> lrmd for instance. But I'm not familiar with the EPEL packaging,
> >> so it could be different there.
> >>
> >> Thanks,
> >>
> >> Dejan
>
> -
>
> Kenneth M DeChick
> Linux Systems Administrator
> Community Computer Services, Inc.
>
> This message and any attachments may contain information that is protected by
> law as privileged and confidential, and is transmitted for the sole use of
> the intended recipient(s). If you are not the intended recipient, you are
> hereby notified that any use, dissemination, copying or retention of this
> e-mail or the information contained herein is strictly prohibited. If you
> received this e-mail in error, please immediately notify the sender by
> e-mail, and permanently delete this e-mail.
>
> _______________________________________________
> Linux-HA mailing list
> [email protected]
> http://lists.linux-ha.org/mailman/listinfo/linux-ha
> See also: http://linux-ha.org/ReportingProblems
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
