<p dir="ltr">Hi Simone,</p>
<p dir="ltr">> According to gluster administration guide:<br>
> <a href="https://docs.gluster.org/en/latest/Administrator%20Guide/Network%20Configurations%20Techniques">https://docs.gluster.org/en/latest/Administrator%20Guide/Network%20Configurations%20Techniques</a>/<br>
>  <br>
> in the "when to bond" section we can read:<br>
> network throughput limit of client/server \<\< storage throughput limit<br>
><br>
> 1 GbE (almost always)<br>
> 10-Gbps links or faster -- for writes, replication doubles the load on the network and replicas are usually on different peers to which the client can transmit in parallel.<br>
><br>
> So if you are using oVirt hyper-converged in replica 3 you have to transmit everything two times over the storage network to sync it with other peers.<br>
><br>
> I'm not really in that details, but if <a href="https://bugzilla.redhat.com/1673058">https://bugzilla.redhat.com/1673058</a> is really like it's described, we even have an 5x overhead with current gluster 5.x.<br>
><br>
> This means that with a 1000 Mbps nic we cannot expect more than:<br>
> 1000 Mbps / 2 (other replicas) / 5 (overhead in Gluster 5.x ???) / 8 (bit per bytes) = 12.5 MByte per seconds and this is definitively enough to have sanlock failing especially because we don't have just the sanlock load as you can imagine.<br>
><br>
> I'd strongly advice to move to 10 Gigabit Ethernet (nowadays with a few hundred dollars you can buy a 4/5 ports 10GBASE-T copper switch plus 3 nics and the cables just for the gluster network) or to bond a few 1 Gigabit Ethernet links.</p>
<p dir="ltr">I didn't know that.<br>
So , with 1 Gbit network everyone should use replica 3 arbiter 1 volumes to minimize replication traffic.</p>
<p dir="ltr">Best Regards,<br>
Strahil Nikolov</p>