Hi Lon,
On Fri, 2007-01-12 at 16:05 -0500, Lon Hohberger wrote:
> On Fri, 2007-01-12 at 14:59 +0100, Simone Gotti wrote:
> > Hi all,
> >
> > On a 2 node openais cman cluster, I failed a network interface and
> > noticed that it didn't failed over the other node.
> >
> > Looking at the rgmanager-2.0.16 code I noticed that:
> >
> > handle_relocate_req is called with preferred_target = -1, but inside
> > this function, there are 2 checks to see if the preferred_target is
> > setted, the check is a 'if (preferred_target != 0)' so the function
> > thinks that a preferred target is choosed. Then, inside the cycle, the
> > only one target that really exists is "me" (as -1 isn't a real target)
> > and there a "goto exausted:", the service is then restarted only on the
> > locale node, where it fails again and so it's stopped. Changing these
> > checks to "> 0" worked.
> >
> > Before writing a patch I noticed that in the RHEL4 CVS tag is used a
> > NODE_ID_NONE instead of the numeric values, so the problem (not tested)
> > probably doesn't happen.
> > Is it probably a forgotten patch on HEAD and RHEL5?
>
> https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=222485
>
> Please attach your patch if you have it; I wrote one, but yours is
> already tested :) (or you can send it here, too)
My patch is the same as yours but you also modified the type of "target"
and "me" variables :D
I have a little doubt:
I did (as you said) a check with ">= 0" but probably it can also be ">
0" as looking in cmanccs.c:read_ccs_nodes I see:
[...]
if (check_nodeids && nodeid == 0) {
char message[132];
sprintf(message, "No node ID for %s, run 'ccs_tool
addnodeids' to fix", nodename);
log_msg(LOG_ERR, message);
write_cman_pipe(message);
return -1;
}
[...]
so looks like a nodeid should be > 0 (as 0 looks like it's not
accepted).
What do you think?
>
> > The same problem is in the ip.sh resource scripts as it's missing the
> > patch for "Fix bug in ip.sh allowing start of the IP if the link was
> > down, preventing failover (linux-cluster reported)." in 1.5.2.16 of
> > RHEL4 branch.
>
> https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=222484
>
> This one's already got a fix as you said in the RHEL4 branch; we'll use
> it.
Thanks!
Bye!
>
> -- Lon
>
> --
> Linux-cluster mailing list
> Linux-cluster@xxxxxxxxxx
> https://www.redhat.com/mailman/listinfo/linux-cluster
--
Simone Gotti
--
Email.it, the professional e-mail, gratis per te: http://www.email.it/f
Sponsor:
Video Corsi GRATIS - Scopri come imparare velocemente e senza stress (Internet, Informatica, Web Marketing, Hobby
Clicca qui: http://adv.email.it/cgi-bin/foclick.cgi?mid=5145&d=14-1
Index: src/daemons/rg_state.c
===================================================================
RCS file: /cvs/cluster/cluster/rgmanager/src/daemons/rg_state.c,v
retrieving revision 1.26
diff -u -b -B -p -r1.26 rg_state.c
--- src/daemons/rg_state.c 14 Dec 2006 22:18:07 -0000 1.26
+++ src/daemons/rg_state.c 14 Jan 2007 16:51:32 -0000
@@ -1308,7 +1308,7 @@ handle_relocate_req(char *svcName, int r
return RG_EFORWARD;
}
- if (preferred_target != 0) {
+ if (preferred_target >= 0) {
allowed_nodes = member_list();
/*
@@ -1380,7 +1380,7 @@ handle_relocate_req(char *svcName, int r
//count_resource_groups(allowed_nodes);
}
- if (preferred_target != 0)
+ if (preferred_target >= 0)
memb_mark_down(allowed_nodes, preferred_target);
memb_mark_down(allowed_nodes, me);
--
Linux-cluster mailing list
Linux-cluster@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/linux-cluster
