<html dir="ltr">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=Windows-1252">
<style type="text/css" id="owaParaStyle"></style>
</head>
<body fpstyle="1" ocsi="0">
<div style="direction: ltr;font-family: Tahoma;color: #000000;font-size: 10pt;">So we've backed out a bunch of network tuning parameters we had set (based on the GPFS wiki pages), they've been set a while but um ... maybe they are causing issues.
<div><br>
</div>
<div>Secondly, we've noticed in dump tscomm that we see connection broken to a node, and then the node ID is usually the same node, which is a bit weird to me.</div>
<div><br>
</div>
<div>We've also just updated firmware on the Intel nics (the x722) which is part of the Skylake board. And specifically its the newer skylake kit we see this problem on. We've a number of issues with the x722 firmware (like it won't even bring a link up when
plugged into some of our 10GbE switches, but that's another story).</div>
<div><br>
</div>
<div>We've also dropped the bonded links from these nodes, just in case its related...</div>
<div><br>
</div>
<div>Simon</div>
<div><br>
</div>
<div>
<div style="font-family: Times New Roman; color: #000000; font-size: 16px">
<hr tabindex="-1">
<div id="divRpF662645" style="direction: ltr;"><font face="Tahoma" size="2" color="#000000"><b>From:</b> gpfsug-discuss-bounces@spectrumscale.org [gpfsug-discuss-bounces@spectrumscale.org] on behalf of jlewars@us.ibm.com [jlewars@us.ibm.com]<br>
<b>Sent:</b> 17 January 2019 14:30<br>
<b>To:</b> Tomer Perry; gpfsug main discussion list<br>
<b>Cc:</b> Yong Ze Chen<br>
<b>Subject:</b> Re: [gpfsug-discuss] Node expels<br>
</font><br>
</div>
<div></div>
<div><font size="2" face="sans-serif">></font><font size="2" face="Calibri">They always appear to be to a specific type of hardware with the same Ethernet controller,
</font><br>
<br>
<font size="2" face="sans-serif">That makes me think you might be seeing packet loss that could require ring buffer tuning (the defaults and limits will differ with different ethernet adapters). </font><br>
<br>
<font size="2" face="sans-serif">The expel section in the slides on this page has been expanded to include a 'debugging expels section' (slides 19-20, which also reference ring buffer tuning):<br>
</font><a href="https://www.ibm.com/developerworks/community/wikis/home?lang=en#!/wiki/General%20Parallel%20File%20System%20(GPFS)/page/DEBUG%20Expels/comment/7e4f9433-7ca3-430f-b40b-94777c507381" target="_blank" rel="noopener noreferrer"><font size="2" color="blue" face="sans-serif">https://www.ibm.com/developerworks/community/wikis/home?lang=en#!/wiki/General%20Parallel%20File%20System%20(GPFS)/page/DEBUG%20Expels/comment/7e4f9433-7ca3-430f-b40b-94777c507381</font></a><font size="2" face="sans-serif"><br>
</font><br>
<font size="2" face="sans-serif">Regards,<br>
John Lewars <br>
Spectrum Scale Performance, IBM Poughkeepsie<br>
</font><br>
<br>
<br>
<br>
<font size="1" color="#5f5f5f" face="sans-serif">From: </font><font size="1" face="sans-serif">Tomer Perry/Israel/IBM</font><br>
<font size="1" color="#5f5f5f" face="sans-serif">To: </font><font size="1" face="sans-serif">gpfsug main discussion list <gpfsug-discuss@spectrumscale.org></font><br>
<font size="1" color="#5f5f5f" face="sans-serif">Cc: </font><font size="1" face="sans-serif">John Lewars/Poughkeepsie/IBM@IBMUS, Yong Ze Chen/China/IBM@IBMCN</font><br>
<font size="1" color="#5f5f5f" face="sans-serif">Date: </font><font size="1" face="sans-serif">01/17/2019 08:28 AM</font><br>
<font size="1" color="#5f5f5f" face="sans-serif">Subject: </font><font size="1" face="sans-serif">Re: [gpfsug-discuss] Node expels</font><br>
<hr noshade="">
<br>
<br>
<font size="2" face="sans-serif">Hi,</font><br>
<br>
<font size="2" face="sans-serif">I was asked to elaborate a bit ( thus also adding John and Yong Ze Chen).</font><br>
<br>
<font size="2" face="sans-serif">As written on the slide:</font><br>
<font size="3">One of the best ways to determine if a network layer problem is root cause for an expel is to look at the low-level socket details dumped in the ‘extra’ log data (mmfs dump all) saved as part of automatic data collection on Linux GPFS nodes.
</font><br>
<br>
<font size="2" face="sans-serif">So, the idea is that in expel situation, we dump the socket state from the OS ( you can see the same using 'ss -i' for example).</font><br>
<font size="2" face="sans-serif">In your example, it shows that the ca_state is 4, there are retransmits, high rto and all the point to a network problem.</font><br>
<font size="2" face="sans-serif">You can find more details here: </font><a href="http://www.yonch.com/tech/linux-tcp-congestion-control-internals" target="_blank" rel="noopener noreferrer"><font size="2" color="blue" face="sans-serif">http://www.yonch.com/tech/linux-tcp-congestion-control-internals</font></a><br>
<br>
<font size="2" face="sans-serif"><br>
Regards,<br>
<br>
Tomer Perry<br>
Scalable I/O Development (Spectrum Scale)<br>
email: tomp@il.ibm.com<br>
1 Azrieli Center, Tel Aviv 67021, Israel<br>
Global Tel: +1 720 3422758<br>
Israel Tel: +972 3 9188625<br>
Mobile: +972 52 2554625<br>
</font><br>
<br>
<br>
<br>
<br>
<font size="1" color="#5f5f5f" face="sans-serif">From: </font><font size="1" face="sans-serif">"Tomer Perry" <TOMP@il.ibm.com></font><br>
<font size="1" color="#5f5f5f" face="sans-serif">To: </font><font size="1" face="sans-serif">gpfsug main discussion list <gpfsug-discuss@spectrumscale.org></font><br>
<font size="1" color="#5f5f5f" face="sans-serif">Date: </font><font size="1" face="sans-serif">17/01/2019 13:46</font><br>
<font size="1" color="#5f5f5f" face="sans-serif">Subject: </font><font size="1" face="sans-serif">Re: [gpfsug-discuss] Node expels</font><br>
<font size="1" color="#5f5f5f" face="sans-serif">Sent by: </font><font size="1" face="sans-serif">gpfsug-discuss-bounces@spectrumscale.org</font><br>
<hr noshade="">
<br>
<br>
<br>
<font size="2" face="sans-serif">Simon,</font><font size="3"><br>
</font><font size="2" face="sans-serif"><br>
Take a look at </font><a href="http://files.gpfsug.org/presentations/2018/USA/Scale_Network_Flow-0.8.pdf" target="_blank" rel="noopener noreferrer"><font size="2" color="blue" face="sans-serif"><u>http://files.gpfsug.org/presentations/2018/USA/Scale_Network_Flow-0.8.pdf</u></font></a><font size="2" face="sans-serif">slide
13.</font><font size="3"><br>
</font><font size="2" face="sans-serif"><br>
<br>
Regards,<br>
<br>
Tomer Perry<br>
Scalable I/O Development (Spectrum Scale)<br>
email: tomp@il.ibm.com<br>
1 Azrieli Center, Tel Aviv 67021, Israel<br>
Global Tel: +1 720 3422758<br>
Israel Tel: +972 3 9188625<br>
Mobile: +972 52 2554625</font><font size="3"><br>
<br>
<br>
<br>
</font><font size="1" color="#5f5f5f" face="sans-serif"><br>
From: </font><font size="1" face="sans-serif">Simon Thompson <S.J.Thompson@bham.ac.uk></font><font size="1" color="#5f5f5f" face="sans-serif"><br>
To: </font><font size="1" face="sans-serif">"gpfsug-discuss@spectrumscale.org" <gpfsug-discuss@spectrumscale.org></font><font size="1" color="#5f5f5f" face="sans-serif"><br>
Date: </font><font size="1" face="sans-serif">17/01/2019 13:35</font><font size="1" color="#5f5f5f" face="sans-serif"><br>
Subject: </font><font size="1" face="sans-serif">[gpfsug-discuss] Node expels</font><font size="1" color="#5f5f5f" face="sans-serif"><br>
Sent by: </font><font size="1" face="sans-serif">gpfsug-discuss-bounces@spectrumscale.org</font><font size="3"><br>
</font>
<hr noshade="">
<font size="3"><br>
<br>
</font><font size="2" face="Calibri"><br>
We’ve recently been seeing quite a few node expels with messages of the form:<br>
<br>
2019-01-17_11:19:30.882+0000: [W] The TCP connection to IP address 10.20.0.58 proto-pg-pf01.bear.cluster <c0n236> (socket 153) state is unexpected: state=1 ca_state=4 snd_cwnd=1 snd_ssthresh=5 unacked=5 probes=0 backoff=7 retransmits=7 rto=26496000 rcv_ssthresh=102828
rtt=6729 rttvar=12066 sacked=0 retrans=1 reordering=3 lost=5<br>
2019-01-17_11:19:30.882+0000: [I] tscCheckTcpConn: Sending debug data collection request to node 10.20.0.58 proto-pg-pf01.bear.cluster<br>
2019-01-17_11:19:30.882+0000: Sending request to collect TCP debug data to proto-pg-pf01.bear.cluster localNode<br>
2019-01-17_11:19:30.882+0000: [I] Calling user exit script gpfsSendRequestToNodes: event sendRequestToNodes, Async command /usr/lpp/mmfs/bin/mmcommon.<br>
2019-01-17_11:24:52.611+0000: [E] Timed out in 300 seconds waiting for a commMsgCheckMessages reply from node 10.20.0.58 proto-pg-pf01.bear.cluster. Sending expel message.<br>
<br>
On the client node, we see messages of the form:<br>
<br>
2019-01-17_11:19:31.101+0000: [N] sdrServ: Received Tcp data collection request from 10.10.0.33<br>
2019-01-17_11:19:31.102+0000: [N] GPFS will attempt to collect Tcp debug data on this node.<br>
2019-01-17_11:24:52.838+0000: [N] sdrServ: Received expel data collection request from 10.10.0.33<br>
2019-01-17_11:24:52.838+0000: [N] GPFS will attempt to collect debug data on this node.<br>
2019-01-17_11:25:02.741+0000: [N] This node will be expelled from cluster rds.gpfs.servers due to expel msg from 10.10.12.41 (b<br>
ber-les-nsd01-data.bb2.cluster in rds.gpfs.server<br>
2019-01-17_11:25:03.160+0000: [N] sdrServ: Received expel data collection request from 10.20.0.56<br>
<br>
They always appear to be to a specific type of hardware with the same Ethernet controller, though the nodes are split across three data centres and we aren’t seeing link congestion on the links between them.<br>
<br>
On the node I listed above, it’s not actually doing anything either as the software on it is still being installed (i.e. it’s not doing GPFS or any other IO other than a couple of home directories).<br>
<br>
Any suggestions on what “(socket 153) state is unexpected” means?<br>
<br>
Thanks<br>
<br>
Simon<br>
<br>
</font><tt><font size="2">_______________________________________________<br>
gpfsug-discuss mailing list<br>
gpfsug-discuss at spectrumscale.org</font></tt><font size="3" color="blue"><u><br>
</u></font><a href="http://gpfsug.org/mailman/listinfo/gpfsug-discuss" target="_blank" rel="noopener noreferrer"><tt><font size="2" color="blue"><u>http://gpfsug.org/mailman/listinfo/gpfsug-discuss</u></font></tt></a><font size="3"><br>
<br>
<br>
</font><tt><font size="2">_______________________________________________<br>
gpfsug-discuss mailing list<br>
gpfsug-discuss at spectrumscale.org<br>
</font></tt><a href="http://gpfsug.org/mailman/listinfo/gpfsug-discuss" target="_blank" rel="noopener noreferrer"><tt><font size="2">http://gpfsug.org/mailman/listinfo/gpfsug-discuss</font></tt></a><tt><font size="2"><br>
</font></tt><br>
<br>
<br>
<br>
</div>
</div>
</div>
</div>
</body>
</html>