David Lee wrote:
> I've recently had a look inside "ping.c" and "ping_group.c".  It wasn't a
> great surprise to find some commonality.  I guess one was derived from
> other.  It means that we've got considerable code duplication, potential
> unnecessary divergence and potential "one-but-miss-the-other" bugfixes).

I just put an identical bug fix into both.  So, I know what you're
talking about...

> Might it be worth attempting to merge them into a more-or-less common
> codebase?  If you think this is worth investigating, then it's probably
> something I could look at, low priority, in the coming months.

I think this would be worth looking at.  Perhaps some ping functions in
the clplumbing library?

But, you know what would be even more interesting, before doing that?

Write a quorum module which used the ping information that heartbeat
collects to declare whether we had quorum or not.

This is why I wrote the ping code in the first place - that and the fact
that it was incredibly simple to do here ;-)  In the meantime, Kevin
found a good use for it, which Andrew replicated.

Here's the bugzilla for that capability:
http://old.linux-foundation.org/developer_bugzilla/show_bug.cgi?id=1497

This would greatly improve the situation for 2-node clusters and in most
cases would eliminate the possibility of mutual-STONITH.

This should be pretty simple to implement, and it would be totally awesome!

-- 
    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/

Reply via email to