<p>Actually I'm studying this technology. This is only a test. I'm trying to understand all possible configuration and at the moment I have some problems to understand the differences among file systems. Now I'm trying to use ocfs2 and it seems to work well but what is the best alternative?</p>
<p><span lang="en"><span>My big question is when should I use one rather than another one? I know glusterfs and, If I'm right, it doesn't need drbd but it is slower than gfs2 and ocfs2. If I understood you can have the best performance using ocfs2, so I'm trying to use it. In case split-brain happens what is the best?<br>
</span></span></p>
And with mysql. To realize two node active/active there are many ways. you can use mysql master/master ndb replication or put data folder in a drbd partition on ocfs2 with the option "external locking" activated. what is the fastest and the safest way? in the case of split-brain happen what is the best? in my opinion the best is drbd. maybe you will loose <br>
<br>As you can see, at the moment I only need to make experiences to understand these concepts.<br><br>Thank you very much for your help.<br><br><br>
<div class="gmail_quote">Il giorno 19/feb/2012 21:52, "Florian Haas" <<a href="mailto:florian@hastexo.com" target="_blank">florian@hastexo.com</a>> ha scritto:<br type="attribution"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
On 02/18/12 10:59, diego fanesi wrote:<br>
> are you saying I can install drbd + gfs2 + pacemaker without using cman?<br>
> It seems that gfs2 depends on cman...<br>
<br>
Only on RHEL/CentOS/Fedora. Not on Debian.<br>
<br>
> I want to realize active/active cluster and I'm following the document<br>
> "cluster from scratch" that you can found on this website.<br>
><br>
> I don't know if there are other ways to realize it.<br>
<br>
Here's a reference config; we use this in classes we teach (where we run<br>
the Pacemaker stack on Debian because that's the only distro that<br>
supports all of Pacemaker, OCFS2, GFS2, GlusterFS and Ceph). This makes<br>
no claims at being perfect, but it works rather well.<br>
<br>
primitive p_dlm_controld ocf:pacemaker:controld \<br>
params daemon="dlm_controld.pcmk" \<br>
op start interval="0" timeout="90" \<br>
op stop interval="0" timeout="100" \<br>
op monitor interval="10"<br>
primitive p_gfs_controld ocf:pacemaker:controld \<br>
params daemon="gfs_controld.pcmk" \<br>
op start interval="0" timeout="90" \<br>
op stop interval="0" timeout="100" \<br>
op monitor interval="10"<br>
group g_gfs2 p_dlm_controld p_gfs_controld<br>
clone cl_gfs2 g_gfs2 \<br>
meta interleave="true"<br>
<br>
Here's the corresponding DRBD/Pacemaker configuration.<br>
<br>
primitive p_drbd_gfs2 ocf:linbit:drbd \<br>
params drbd_resource="gfs2" \<br>
op monitor interval="10" role="Master" \<br>
op monitor interval="30" role="Slave"<br>
ms ms_drbd_gfs2 p_drbd_gfs2 \<br>
meta notify="true" master-max="2" \<br>
interleave="true"<br>
colocation c_gfs2_on_drbd inf: cl_gfs2 ms_drbd_gfs2:Master<br>
order o_drbd_before_gfs2 inf: ms_drbd_gfs2:promote cl_gfs2:start<br>
<br>
Of course, you'll have to add proper fencing, and there are several DRBD<br>
configuration options that you must remember to set. And, obviously, you<br>
need the actual Filesystem resources to manage your GFSs proper.<br>
<br>
That being said, it's entirely possible that a GlusterFS based solution<br>
would solve your issue as well, and be easier to set up. Or even<br>
something NFS based, backed by a single-Primary DRBD config for HA. You<br>
didn't give many details of your setup, however, so it's impossible to<br>
tell for certain.<br>
<br>
Hope this helps.<br>
Florian<br>
<br>
--<br>
Need help with High Availability?<br>
<a href="http://www.hastexo.com/now" target="_blank">http://www.hastexo.com/now</a><br>
<br>
_______________________________________________<br>
Pacemaker mailing list: <a href="mailto:Pacemaker@oss.clusterlabs.org" target="_blank">Pacemaker@oss.clusterlabs.org</a><br>
<a href="http://oss.clusterlabs.org/mailman/listinfo/pacemaker" target="_blank">http://oss.clusterlabs.org/mailman/listinfo/pacemaker</a><br>
<br>
Project Home: <a href="http://www.clusterlabs.org" target="_blank">http://www.clusterlabs.org</a><br>
Getting started: <a href="http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf" target="_blank">http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf</a><br>
Bugs: <a href="http://bugs.clusterlabs.org" target="_blank">http://bugs.clusterlabs.org</a><br>
</blockquote></div>