[Pacemaker] handling the case-insensitive hostname

Junko IKEDA tsukishima.ha at gmail.com
Wed Apr 10 01:34:59 UTC 2013


sorry for the confusion again...

case 5 is for Corosync.

* Heartbeat
Hostname is always converted into lower case.
http://hg.linux-ha.org/dev/file/add12b838ef4/heartbeat/heartbeat.c#l576

hostname(LOWER) command(LOWER) -> OK
hostname(UPPER) command(LOWER) -> OK
hostname(LOWER) command(UPPER) -> NG
hostname(UPPER) command(UPPER) -> NG

Some crm_xxx commands reject UPPER hostname,
and attached patch would handle this.

* Corosync
It seems that there is no case-sensitive restriction, so it's strange to
see case 5.







2013/4/10 Andrew Beekhof <andrew at beekhof.net>

>
> On 09/04/2013, at 7:28 PM, Junko IKEDA <tsukishima.ha at gmail.com> wrote:
>
> > Sorry, that patch means, Heartbeat can't handle this situation, so
> Heartbeat is needed to be modified.
> > Corosync doesn't need any modification.
>
> What about
>
> >>  * case 5 -> ng (lower-case to upper-case at remote)
> >>
> >> [root at GUEST03 ~]# crm_resource -C -r dummy -N guest04 -Q
> >> Cleaning up dummy on guest04
> >> Waiting for 1 replies from the CRMdNo messages received in 60 seconds..
> aborting
> >>
> >>
> >> Do you have any ideas to manage case 5 ?
>
>
> Was that for corosync or heartbeat? or both?
>
> > I'll file this into the bugzilla.
> >
> > Thanks,
> > Junko
> >
> >
> > 2013/4/9 Andrew Beekhof <andrew at beekhof.net>
> >
> > On 08/04/2013, at 5:14 PM, Junko IKEDA <tsukishima.ha at gmail.com> wrote:
> >
> > > Hi,
> > >
> > > Yes, Corosync version is v2.3.1, so I thought that this should work
> well.
> > > Heartbeat is needed to modified to handle this.
> >
> > It doesn't make sense to me how adding heartbeat specific code helps
> when running corosync.
> > Can you create a bug for this?  I'm just about done clearing my email
> backlog and bugzilla is what I'll be working through next :)
> >
> > >
> > > Thanks,
> > > Junko
> > >
> > >
> > > 2013/4/8 Andrew Beekhof <andrew at beekhof.net>
> > > This is using corosync 2.0?
> > >
> > > On 04/04/2013, at 9:55 PM, Junko IKEDA <tsukishima.ha at gmail.com>
> wrote:
> > >
> > > > Hi,
> > > >
> > > > I run the latest pacemaker + corosync, and check some
> case-insensitive behaviors.
> > > > Hostname should be handled in a case-insensitive manner,
> > > > and there might be the following pattern.
> > > >
> > > > example;
> > > > hostname = GUEST03
> > > > CLI input = guest03 (option for crm_resource command)
> > > >
> > > > [root at GUEST03 ~]# crm_mon -1
> > > > Last updated: Thu Apr  4 17:11:48 2013
> > > > Last change: Thu Apr  4 17:11:12 2013 via cibadmin on GUEST03
> > > > Stack: corosync
> > > > Current DC: GUEST03 (3232242816) - partition with quorum
> > > > Version: 1.1.9-e8caee8
> > > > 2 Nodes configured, unknown expected votes
> > > > 1 Resources configured.
> > > >
> > > >
> > > > Online: [ GUEST03 GUEST04 ]
> > > >
> > > >  dummy  (ocf::pacemaker:Dummy): Started GUEST03
> > > >
> > > >
> > > >
> > > >  * case 1 -> ok (without hostname)
> > > >
> > > > [root at GUEST03 ~]# crm_resource -C -r dummy
> > > > Cleaning up dummy on GUEST03
> > > > Cleaning up dummy on GUEST04
> > > > Waiting for 1 replies from the CRMd. OK
> > > >
> > > >  * case 2 -> ok (uppper-case to upper-case at local)
> > > >
> > > > [root at GUEST03 ~]# crm_resource -C -r dummy -N GUEST03 -Q
> > > > Cleaning up dummy on GUEST03
> > > > Waiting for 1 replies from the CRMd. OK
> > > >
> > > >  * case 3 -> ok (uppper-case to upper-case at remote)
> > > >
> > > > [root at GUEST03 ~]# crm_resource -C -r dummy -N GUEST04 -Q
> > > > Cleaning up dummy on GUEST04
> > > > Waiting for 1 replies from the CRMd. OK
> > > >
> > > >  * case 4 -> ok (lower-case to upper-case at local)
> > > >
> > > > [root at GUEST03 ~]# crm_resource -C -r dummy -N guest03 -Q
> > > > Cleaning up dummy on guest03
> > > > Waiting for 1 replies from the CRMd[root at GUEST03 ~]#
> > > >
> > > >  * case 5 -> ng (lower-case to upper-case at remote)
> > > >
> > > > [root at GUEST03 ~]# crm_resource -C -r dummy -N guest04 -Q
> > > > Cleaning up dummy on guest04
> > > > Waiting for 1 replies from the CRMdNo messages received in 60
> seconds.. aborting
> > > >
> > > >
> > > > Do you have any ideas to manage case 5 ?
> > > > By the way, it seems that "crm_failcount" works well in all cases.
> > > > "crm_resource -m/M/U" works, too.
> > > >
> > > >
> > > > [root at GUEST03 ~]# cd /home/ikedaj/src/pacemaker/
> > > > [root at GUEST03 pacemaker]# git log | head -n 10
> > > > commit e8caee88c9e078fccf98a9da05543b73d4696f04
> > > > Merge: 288bfeb 203554b
> > > > Author: Andrew Beekhof <andrew at beekhof.net>
> > > > Date:   Thu Mar 28 17:19:56 2013 -0700
> > > >
> > > >     Merge pull request #268 from gao-yan/node-cache
> > > >
> > > >     Fix: legacy: cl#5148 - Correctly remove a node that used to have
> a different nodeid
> > > >
> > > > commit 288bfeb1b84a736a379a1ddfb58cd5e90c59421e
> > > >
> > > > [root at GUEST03 pacemaker]# cd ../corosync/
> > > > [root at GUEST03 corosync]# git log | head -n 10
> > > > commit 005e7fd3b9de79bc32894f9f15431a7cf32a3b27
> > > > Author: Andrei Belov <defanator at gmail.com>
> > > > Date:   Thu Mar 28 14:24:41 2013 +0000
> > > >
> > > >     Improved POSIX-compliant handling of getpwnam_r() and
> getgrnam_r().
> > > >
> > > >     Signed-off-by: Andrei Belov <defanator at gmail.com>
> > > >     Reviewed-by: Jan Friesse <jfriesse at redhat.com>
> > > >
> > > > commit 55d9db902cdd5155abc269f3e64a4251f00ef853
> > > >
> > > >
> > > > Thanks,
> > > > Junko IKEDA
> > > > _______________________________________________
> > > > Pacemaker mailing list: Pacemaker at oss.clusterlabs.org
> > > > http://oss.clusterlabs.org/mailman/listinfo/pacemaker
> > > >
> > > > Project Home: http://www.clusterlabs.org
> > > > Getting started:
> http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
> > > > Bugs: http://bugs.clusterlabs.org
> > >
> > >
> > > _______________________________________________
> > > Pacemaker mailing list: Pacemaker at oss.clusterlabs.org
> > > http://oss.clusterlabs.org/mailman/listinfo/pacemaker
> > >
> > > Project Home: http://www.clusterlabs.org
> > > Getting started:
> http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
> > > Bugs: http://bugs.clusterlabs.org
> > >
> > > <hostname.patch>_______________________________________________
> > > Pacemaker mailing list: Pacemaker at oss.clusterlabs.org
> > > http://oss.clusterlabs.org/mailman/listinfo/pacemaker
> > >
> > > Project Home: http://www.clusterlabs.org
> > > Getting started:
> http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
> > > Bugs: http://bugs.clusterlabs.org
> >
> >
> > _______________________________________________
> > Pacemaker mailing list: Pacemaker at oss.clusterlabs.org
> > http://oss.clusterlabs.org/mailman/listinfo/pacemaker
> >
> > Project Home: http://www.clusterlabs.org
> > Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
> > Bugs: http://bugs.clusterlabs.org
> >
> > _______________________________________________
> > Pacemaker mailing list: Pacemaker at oss.clusterlabs.org
> > http://oss.clusterlabs.org/mailman/listinfo/pacemaker
> >
> > Project Home: http://www.clusterlabs.org
> > Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
> > Bugs: http://bugs.clusterlabs.org
>
>
> _______________________________________________
> Pacemaker mailing list: Pacemaker at oss.clusterlabs.org
> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
>
> Project Home: http://www.clusterlabs.org
> Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
> Bugs: http://bugs.clusterlabs.org
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.clusterlabs.org/pipermail/pacemaker/attachments/20130410/e287af7b/attachment.htm>


More information about the Pacemaker mailing list