Thanks, those links will be useful. I'm going to give RH cluster a try.
It seems to me that the HA-LVM implementation uses LVM tags and the
volume_list feature of LVM to implement the exclusivity I was looking
for. Clever.
If for some reason RH cluster won't work for me I can probably use the
same trick to create reasonably newbie-proof manual failover scripts.
Thanks to everyone for their help.
On 03-Feb-10 10:36, Zoran Salahovic Lendra wrote:
> Heya,
> I think that you shoud definitely use Conga to create your initial
> configuratin files and test the cluster. It's easy and helpfull,... later
> you can edit them and also use command line, scripts,... and so on. Take a
> look at this small article to see what Conga provides, and also it's
> architecture:
> http://magazine.redhat.com/2007/03/19/teaching-your-cluster-and-storage-systems-to-dance-an-introduction-to-conga/
>
>
> more info here:
> http://www.redhat.com/docs/en-US/Red_Hat_Enterprise_Linux/5.4/html/Cluster_Suite_Overview/s1-clumgmttools-overview-CSO.html#s2-conga-overview-CSO
>
>
> you can try creating a resource and see if you like it.
>
> "... shared resources to be used by high-availability services,... consist
> of file systems, IP addresses, NFS mounts and exports, and user-created
> scripts that are available to any high-availability service in the
> cluster. "
>
> About your main fear I would sugest a good User manuals for the people who
> is going to work whit it,... And also Panadol hehe
>
>
> Reagards,
> Zoran
>
>
>
>
> "urgrue"<***@bulbous.org>
>
> Enviado por: redhat-list-***@redhat.com
> 02/02/2010 18:34
> Por favor, responda a
> General Red Hat Linux discussion list<redhat-***@redhat.com>
>
>
> Para
> "General Red Hat Linux discussion list"<redhat-***@redhat.com>, "General
> Red Hat Linux discussion list"<redhat-***@redhat.com>
> cc
>
> Asunto
> Re: HA-LVM vs CLVM
>
>
>
>
>
>
> On Tue, 02 Feb 2010 05:50 -0700, "Matt Iavarone"
> <***@gmail.com> wrote:
>
>> You don't need ricci or luci for either of these, but they do make it
>> easier to build and manage your clusters. You can use
>> system-config-cluster in their place.
>>
> I'll take a look at system-config-cluster. The manual just made it sound
> like its kinda deprecated in favor of conga. And I'm not so crazy about
> conga, it seems I'm able to get strange behaviour out of it even doing
> the simplest things in the most pristine setups, which partly explains
> my reluctance to use RHC in all its glory.
>
>
>> You can use, I assume, use just ha-lvm and gfs2 without rgmanager or
>> cman, but how will you manage the filesystem if the node fails? Will
>> you manually mount it on your backup node? And there are many options
>> for a fence device. A red hat cluster using clvm and gfs2 is simple
>> and easy to manage.
>>
> Manual mount/activate on the backup node is fine and in fact required in
> my case, for the same reason that it needs to be ext3 - my employer is
> ultra-conservative. So I'm not really sure where all this leaves me...
>
> The main fear, and the reason I don't want to just "not mount it" on the
> passive node as suggested by others, is that some helpdesk newbie or
> careless person goes and starts it on the passive node - very probably
> destroying the whole thing. Without any measure in place to at the very
> least produce a warning, it's a bit too plausible a scenario.
>
>
--
redhat-list mailing list
unsubscribe mailto:redhat-list-***@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/redhat-list