<html>
<head>
<style><!--
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
font-size: 12pt;
font-family:Calibri
}
--></style></head>
<body class='hmmessage'><div dir='ltr'>Hi,<br>We are experiencing allot of filesystem issues, for example:<br> kernel: EXT3-fs error (device drbd0): ext3_free_blocks_sb: bit already cleared for block 5776787<br>These occur randomly after several consequative failovers between the cluster nodes.<br>We are using drbd with ext3 filesystem with a write-through controller.<BR>Are there any preferences what elevator we should use?<br>According to <a href="http://www.drbd.org/users-guide/s-latency-tuning.html">http://www.drbd.org/users-guide/s-latency-tuning.html</a> <br> both cfq/deadline are fine with preference to deadline.<br>But both re-order writes, right? possibly breaking file system/journal on standby.<BR> <BR>Please advise,<BR> <BR>Setup<BR> drbd with ext3 filesystem <BR> write-through controller<BR> Two nodes cluster<br> Physical Appliances<br> CentOS 6.2 with 2.6.32-220<br> kmod-drbd83-8.3.12-1.el6.i686<BR> <BR>Thanks,<br>Oren<BR> </div></body>
</html>