<div dir="ltr">I agree with Valentin. And on his 2nd point, recommendation is usually to set +A parameter to at least 12 threads per core on which your node is deployed on. e.g. 128 on an 8 core platform should be fine.<br><div class="gmail_extra"><br clear="all"><div><div class="gmail_signature" data-smartmail="gmail_signature"><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><br></div><div dir="ltr">Best regards,<div>Ayanda</div><div><br></div><div><font face="monospace, monospace">Erlang Solutions Ltd.</font></div><div><font face="monospace, monospace"><br></font></div></div></div></div></div></div></div></div></div></div>
<br><div class="gmail_quote">On 15 May 2017 at 08:51, Valentin Micic <span dir="ltr"><<a href="mailto:v@pharos-avantgard.com" target="_blank">v@pharos-avantgard.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div style="word-wrap:break-word">Just a thought… <div><br></div><div>1. Check your interface for errors (e.g. collisions etc.)<div>This usually happens when your local ethernet interface and adjacent switch have different configuration (e.g. one is configured to be half-duplex, and another as full-duplex).</div><div>In this case, as the traffic increases, a number of errors (e.g. collisions) on the interface increases as well, thus causing nodes to get disconnected.</div><div><br></div><div>2. Check the disk I/O. </div><div>Sometimes, excessive disk I/O operations may cause "uninterruptible sleep". When this happens, your CPU is prevented from doing anything else but wait for disk.</div><div>If this is the case, start your run-time (where excessive disk I/O is taking place) with +A attribute (this will increase a number of threads responsible for disk I/O).</div><div><br></div><div>Kind regards</div><div><br></div><div>V/</div><div><br></div><div><br><div><div><div class="h5"><div>On 15 May 2017, at 6:56 AM, Arun wrote:</div><br class="m_6367042351819859907Apple-interchange-newline"></div></div><blockquote type="cite"><div><div class="h5">
<div text="#000000" bgcolor="#FFFFFF">
Hi,<br>
<br>
I am running my application on a distributed way, with one
server node and multiple client nodes and all nodes are started in <b>visible</b>
mode. A SNMP agent is running in the server node and I have a GUI
through which I can access all the nodes in the distributed system.
The SNMP query will be landing in the server node and which further
does rpc to the specific client node and serve the query. <br>
<br>
The problem what I am facing is that after the complete nodes are up
while accessing properties of some client nodes, the server node is
getting node down messages for most of the client nodes connected to
it with reason "<strong><span class="m_6367042351819859907code">connection_closed</span></strong>"
. The physical connection is intact and there is no ping break
between the nodes. <br>
<br>
What could be the reason for this error. ? and what are the
circumstances this error can come.?<br>
<br>
Note : The tick time is set to 5 Second in all the nodes.<br>
<br>
Regards,<br>
Arun P<br>
</div></div></div>
______________________________<wbr>_________________<br>erlang-questions mailing list<br><a href="mailto:erlang-questions@erlang.org" target="_blank">erlang-questions@erlang.org</a><br><a href="http://erlang.org/mailman/listinfo/erlang-questions" target="_blank">http://erlang.org/mailman/<wbr>listinfo/erlang-questions</a><br></blockquote></div><br></div></div></div><br>______________________________<wbr>_________________<br>
erlang-questions mailing list<br>
<a href="mailto:erlang-questions@erlang.org">erlang-questions@erlang.org</a><br>
<a href="http://erlang.org/mailman/listinfo/erlang-questions" rel="noreferrer" target="_blank">http://erlang.org/mailman/<wbr>listinfo/erlang-questions</a><br>
<br></blockquote></div><br></div></div>