<div dir="auto">Hi Andreas,<div dir="auto"><br></div><div dir="auto">Before rebooting, I have tried some performance tuning inorder to prevent timeout errors. As we have sufficient RAM and cpu power, I have increased transport.listen-backlog in Kernel and syn_backlog and max-connections in Kernel. So, I expected that it won't cause a problem. Also the NFS clients are mounted but not being used. And all the nodes are in same network.</div><div dir="auto"><br></div><div dir="auto">My assumption was that some slowness in the beginning can be seen, which will be resolved automatically.</div><div dir="auto"><br></div><div dir="auto">Is it still a base idea to have 72 nodes and starting them at once?</div><div dir="auto"><br></div><div dir="auto">Regards,</div><div dir="auto">Jeevan.</div></div><br><div class="gmail_quote"><div dir="ltr">On Sun, Nov 25, 2018, 6:00 PM Andreas Davour <<a href="mailto:ante@update.uu.se">ante@update.uu.se</a> wrote:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><br>
72 nodes!!???<br>
<br>
Can the common wisdom come to the rescue here. Does this even work? Wont <br>
the translator overhead make so many nodes scale terribly?<br>
<br>
Are people building clusters that big, and getting any performance at <br>
all?<br>
<br>
/andreas<br>
<br>
On Sun, 25 Nov 2018, Jeevan Patnaik wrote:<br>
<br>
> Hi,<br>
><br>
> I have different issues:<br>
><br>
> I have restarted glusterd service on my 72 nodes almost parallelly with<br>
> ansible while the gluster NFS clients are in mounted state<br>
><br>
> After that many of the gluster peers went to rejected state. In logs, I see<br>
> msg id 106010 stating that checksum doesn't match.<br>
><br>
> I'm confused which checksum is that and how is it changed after I restart.<br>
><br>
> I restarted because gluster volume status commands gives timeout. I have<br>
> tiering enabled on the volume and was trying to detach. And that too never<br>
> completed. The status shows only in progress even the tiered volume<br>
> contains only a few 100 8MB filese I created for testing.<br>
><br>
> my overall experience with gluster tiering is really bad :(<br>
><br>
> Besides, what's the best way to restore old state if something goes wrong?<br>
> Till now, I have been using no volfile at all.. I only use volume status<br>
> commands to configure my cluster. Do I need to use a volfile inorder to<br>
> restore something?<br>
><br>
> Gluster version is 3.12.15<br>
> I have checked the op Version on all nodes and they all are same.<br>
><br>
><br>
> Regards<br>
> Jeevan।<br>
><br>
<br>
--<br>
"economics is a pseudoscience; the astrology of our time"<br>
Kim Stanley Robinson</blockquote></div>