<html>
<head>
<style><!--
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
font-size: 10pt;
font-family:Verdana
}
--></style>
</head>
<body class='hmmessage'>
1) Gbit-Ethernet connection <BR>
Could you verify that it is Gbit connection ?<BR>
<BR>
2) If it is Gbit, try tune "rate 50M" to "rate 100M".<BR>
Just a suggestion.<BR>
<BR>
:-)<BR>
<BR>
edward.<BR> <BR>
<HR id=stopSpelling>
From: klaushahn@siemens.com<BR>To: drbd-user@lists.linbit.com<BR>Date: Tue, 29 Dec 2009 11:58:51 +0100<BR>Subject: [DRBD-user] Poor performance of drbd/heartbeat/oracle cluster<BR><BR>
<STYLE>
.ExternalClass .ecxEmailQuote
{margin-left:1pt;padding-left:4pt;border-left:#800000 2px solid;}
</STYLE>
<FONT face="Arial, sans-serif" size=2>
<DIV>Hi list,</DIV>
<DIV>I've just set up a 2 node cluster: </DIV>
<DIV>Hardare: 1 CPU (current Xeon), 12 GB RAM, RAID 1, 1 *100 Mbit, 1* 1Gbit Ethernet </DIV>
<DIV> Storage: Raid 1 (300GB), partitions for OS, Swap and DRBD-Device (160GB).</DIV>
<DIV>Software: SLES 11, drbd 8.2.7, heartbeat 2.99, Oracle 10.2.0.4</DIV>
<DIV> </DIV>
<DIV>Current drbd settings:</DIV>
<DIV> </DIV>
<DIV>global {</DIV>
<DIV> dialog-refresh 1;</DIV>
<DIV> minor-count 5;</DIV>
<DIV>}</DIV>
<DIV>common {</DIV>
<DIV> syncer {</DIV>
<DIV> rate 50M;</DIV>
<DIV> }</DIV>
<DIV> net {</DIV>
<DIV> after-sb-1pri discard-secondary;</DIV>
<DIV> after-sb-2pri disconnect;</DIV>
<DIV> after-sb-0pri discard-zero-changes;</DIV>
<DIV> }</DIV>
<DIV> startup {</DIV>
<DIV> degr-wfc-timeout 120;</DIV>
<DIV> wfc-timeout 0;</DIV>
<DIV> }</DIV>
<DIV> handlers {</DIV>
<DIV> split-brain "/usr/lib/drbd/notify-split-brain.sh root";</DIV>
<DIV> }</DIV>
<DIV>}</DIV>
<DIV>resource oradata {</DIV>
<DIV> protocol C;</DIV>
<DIV> disk {</DIV>
<DIV> on-io-error pass_on;</DIV>
<DIV> }</DIV>
<DIV> on elsserverbeta {</DIV>
<DIV> device /dev/drbd0;</DIV>
<DIV> address 192.168.5.3:7788;</DIV>
<DIV> meta-disk internal;</DIV>
<DIV> disk /dev/sda3;</DIV>
<DIV> }</DIV>
<DIV> on elsserveralpha {</DIV>
<DIV> device /dev/drbd0;</DIV>
<DIV> address 192.168.5.2:7788;</DIV>
<DIV> meta-disk internal;</DIV>
<DIV> disk /dev/sda3;</DIV>
<DIV> }</DIV>
<DIV>}</DIV>
<DIV> </DIV>
<DIV>DRBD uses the Gbit-Ethernet connection for replication.</DIV>
<DIV> </DIV>
<DIV>The DRBD-device holds the Oracle data (data and redolog files).</DIV>
<DIV> </DIV>
<DIV>The application which runs on top of the HA-Stack uses a lot of small transactions.</DIV>
<DIV>Oracle throws following warning: "log write time 760 ms, size 8 KB".</DIV>
<DIV> </DIV>
<DIV>If the drbd-stack on the second node (elsserverbeta) is stopped, the application is "fast" (2 to 3 times faster).</DIV>
<DIV> </DIV>
<DIV><FONT face="Arial, sans-serif">Any idea ?</FONT></DIV>
<DIV><FONT face="Arial, sans-serif"> </FONT></DIV>
<DIV><FONT face="Arial, sans-serif">Regards, Klaus</FONT></DIV>
<DIV><FONT face="Arial, sans-serif"> </FONT></DIV>
<DIV><FONT face="Arial, sans-serif"> </FONT></DIV></FONT> <br /><hr />Windows Live: Keep your friends up to date <a href='http://www.microsoft.com/middleeast/windows/windowslive/see-it-in-action/social-network-basics.aspx?ocid=PID23461::T:WLMTAGL:ON:WL:en-xm:SI_SB_1:092010' target='_new'>with what you do online.</a></body>
</html>