<div dir="ltr">Ah, your part power is too high for that few devices. I would have recommend 14 rather than 18. Which is good enough to scale up to ~500 devices before you might worry about maybe running into balancing issues that *could* make it difficult to run your cluster more than 70-80% full, but at that point it's 20-40 servers anyway and your disks are probably 6TB or more. Anyway, thats in the future. In the meantime you've got your object notes sorta overloaded and thrashing...<div>
<br></div><div>You could set run_pause in the [object-replicator] section of object-server.conf to like 300 or 600 or something to put them to sleep after a cycle if you don't expect many changes/failures going on in the "cluster".</div>
<div><br></div><div>How much data do you have? Can you migrate it to a new cluster with a ring better suited for that smaller scale?<br><div><div><br></div><div><br></div></div></div></div><div class="gmail_extra"><br><br>
<div class="gmail_quote">On Tue, Jul 1, 2014 at 3:19 PM, Pete Zaitcev <span dir="ltr"><<a href="mailto:zaitcev@redhat.com" target="_blank">zaitcev@redhat.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div class="">On Tue, 1 Jul 2014 10:50:15 +0100<br>
Diogo Vieira <<a href="mailto:dfv@eurotux.com">dfv@eurotux.com</a>> wrote:<br>
<br>
> Can you tell me if this is normal behaviour? If so, how will<br>
> this scale when I add more objects? Will it keep getting more<br>
> and more CPU usage?<br>
<br>
</div>Dunno if it's normal or not, but clusters installed with default<br>
parameters do that. My nodes in a similar cluster make about 200<br>
REPLICATE ops every second. The noise is annoying, but it only<br>
adds up to 80 KB/s.<br>
<span class="HOEnZb"><font color="#888888"><br>
-- Pete<br>
</font></span></blockquote></div><br></div>