clustering with RHEL
[email protected] Sat, 21 Apr 2007 08:30:59 +0200
| Newsgroups | gmane.linux.evms.devel |
|---|---|
| Message-ID | <[email protected]> |
Hi!
We´re running RHEL 4 and 5 clustering since some years now, and would
like to use evms in clustering. Since RHEL is supporting multi-node
clustering, I´m not sure how to use evms. Also, we would obviously not
want to install Linux-HA to be able to use the clustering features.
Steve mentioned the: "admin_mode" is set to "yes" in the "csm"
> section of /etc/evms.conf, which is not recommended since it can be very
> dangerous)
In what way would that be dangerous to use? We´re experienced cluster
admins. Would that help us in any way?
We´re using clustering both with multinode failover, and with multinode
GFS.
So, questions here are: How can we configure shared cluster
containers(for GFS clustering) without installing Linux-HA?
And: How can we get a multinode failover?
Best regards from/Med vänliga hälsningar från
Johan Kragsterman
http://www.capvert.se
"Jose Jerez" <[email protected]>
Sänt av: [email protected]
2007-04-13 14:56
Till: "Steve Dobbelstein" <[email protected]>
Kopia: [email protected]
Ärende: Re: [Evms-devel] iscsi and clustering
Excellent, I installed linux-HA and now the cluster segment manager is
shown as an option :-)
So I managed to create a private segment container, on top of that a
LVM2 container, then a LVM2 region and volumes that can be mounted.
BUT, I could detect the following problem: when the evms init script
is run, which basically executes the evms_activate command, the
following message appears:
abr 13 12:23:11 Engine: The plug-in Novell-HA in module
/lib/evms/2.5.5/hb2-1.0.0.so failed to load. The plug-in's
setup_evms_plugin() function failed with error code 19: No existe el
dispositivo. (that is: device doen't exist)
After starting evmsgui the volume in the private cluster is not
active, so I figured out that linux-HA has to be started before
running evms_activate; but it takes a while for heartbeat to settle
down and allow the evms_activate to run without the previous error.
And this is an annoyance if I want to have the volume mounted from
/etc/fstab. ¿Any suggestion on how to improve this?
Anyway my main concern right now is how to configure the private
cluster container as a resource in heartbeat; I see there's a document
in the web http://evms.sourceforge.net/install/cluster.html but it
seems outdated, at least for me because I'm using heartbeat v2 (2.0.7)
by the way evms is 2.5.5 all in SLES-10. For my initial testing I
won't use STONITH, I'll deal with that later.
So my question again is: is it possible to have this configuration
done (evms 2.5.5 + heartbeat v2 2.0.7) + private cluster container) ?
if yes then
How(hints);
else
Damn(my luck);
Waiting for your wisdom to illuminate me.
On 3/29/07, Steve Dobbelstein <[email protected]> wrote:
>
>
> "Jose Jerez" <[email protected]> wrote on 03/23/2007 05:59:19 AM:
>
>
> > Hello everyone,
> >
> > I'm setting up a test environment with evms to try the clustering
> > capabilities. The common storage device is an iscsi target (exported
> > from a SLES-10) and is accessible from two xen virtual machines (also
> > SLES-10).
> >
> > The iscsi disk appears as sda in the "Available objects" and "Disks"
> > Tabs, and is not shown in the Volumes tab; I can't remove any segment
> > manager from the disk so I assume that it is seen as a clean new disk
> > by evms.
> >
> > Now when I try to add a new segment manager to the disk the only
> > options available are:
> >
> > DOS segment manager
> > Bad block relocation segment manager
> > GPT segment manager
> >
> > No cluster segment manager.
> >
> > ¿Any hints or directions on how to do this, if at all possible?
>
> The Cluster Segment Manager requires that the machine belongs to the
> membership of a cluster (or that "admin_mode" is set to "yes" in the
"csm"
> section of /etc/evms.conf, which is not recommended since it can be very
> dangerous) . It can't safely create a CSM segment on a shared disk
unless
> it can coordinate the creation with the other nodes in the cluster.
>
> Being a member of a cluster implies that you are running some kind of
> cluster software (EVMS supports Linux-HA and IBM's RSCT), and that the
> machine is currently an active node in the cluster. You will need to
make
> sure EVMS is installed on all the nodes in the cluster so that it can
> coordinate the management of shared storage.
>
> Steve D.
>
>
-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share
your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________
Evms-devel mailing list
[email protected]
To subscribe/unsubscribe, please visit:
https://lists.sourceforge.net/lists/listinfo/evms-devel
-------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
_______________________________________________
Evms-devel mailing list
[email protected]
To subscribe/unsubscribe, please visit:
https://lists.sourceforge.net/lists/listinfo/evms-devel