<html>
<head>
<meta http-equiv="content-type" content="text/html; charset=ISO-8859-1">
</head>
<body bgcolor="#FFFFFF" text="#000000">
Hello,<br>
<br>
We are experiencing "PingAck timeout" on a system where multiple
DRBD resources are configured (more exactly a pair of active/active
Lustre MDS servers):<br>
<br>
A --- drbd0 --- B [nfs-data] idle<br>
A --- drbd1 --- B [nfs-apps] idle<br>
A --- drbd2 --- B [nfs-tmp] idle<br>
A --> drbd3 --> B [mdt1] heavy load<br>
A <-- drbd4 <-- B [mdt2] heavy load<br>
A --- drbd5 --- B [mgs] idle<br>
<br>
Our environment is DRBD 8.4.4, with "ping-int = 10s" and
"ping-timeout = 25" (2.5s)<br>
<br>
The link between the two servers is 20Gb/s Infiniband (configured in
datagram mode).<br>
<br>
Strangely, the timeout occurs on an idle resource (e.g. drbd1) when
two of the other resources ('mdt1' and 'mdt2') are heavily loaded
(and displaying no connection/timeout problem what-so-ever).<br>
<br>
Looking at the source code, I believe that DRBD cannot know about
the potentially "congested" link (because of the heavily loaded
resources, 'mdt1' and 'mdt2') and the potentially resulting PingAck
timeout it may spawn for another idle one (e.g. 'drbd1'). Am I right
?<br>
<br>
Is there a way to circumvent this problem ?<br>
<br>
Thanks and best,<br>
<br>
Cédric Dufour<br>
<div class="moz-signature">-- <br>
<meta http-equiv="Content-Type" content="text/html;
charset=ISO-8859-1">
<style type="text/css">
DIV.signature {FONT:normal 11px sans-serif;COLOR:#000000;}
DIV.signature P {MARGIN:5px 0px;FONT:bold 13px sans-serif;COLOR:#000050;}
</style>
<div class="signature">
<p>Cédric Dufour @ Idiap Research Institute</p>
</div>
</div>
</body>
</html>