Dejan Muhamedagic wrote:
> Hi,
>
> Dealing with the LRM regression testing, I've noticed that the
> lrmd tries to get the configuration from the environment which is
> empty (configuration-wise: what a word!) if lrmd is started by
> hand and not by heartbeat. Apart from the lrmd's messages ending
> up in a place where they are not supposed to go, the output of the
> ocf_log is simply lost in space.
>
> Which solution do you propose?
>
> Should lrmd be started by itself without heartbeat at all? It
> does have functionality which could be attractive outside of the
> cluster. If that is the case, shouldn't lrmd parse the config
> files itself?
If it parsed the heartbeat configuration file, that would render it not
very useful by itself, since it would require a heartbeat-style
configuration file.
Although the LRM could be useful in some other context, it won't ever be
useful _by itself_. It will always be used with some other smarter
agency that knows how to tell it what to do by C APIs. In that context,
that smarter system ought to know how to start it up properly.
And, reading that environments config files and passing them to the LRM
as environment variables seems better than having two or three config
files in this other environment. Or that's my first thought on the subject.
--
Alan Robertson <[EMAIL PROTECTED]>
"Openness is the foundation and preservative of friendship... Let me
claim from you at all times your undisguised opinions." - William
Wilberforce
_______________________________________________________
Linux-HA-Dev: [email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha-dev
Home Page: http://linux-ha.org/
