<br><font size=2 face="sans-serif">Hi Sunil,</font>
<br>
<br><font size=2 face="sans-serif">my lotus notes choked on the table from
excel... So the two nodes have the following nodenumbers:</font>
<br><font size=2 face="sans-serif">Node ocfs2
crs/css</font>
<br><font size=2 face="sans-serif">byaz05 0
2</font>
<br><font size=2 face="sans-serif">byaz10 1
1</font>
<br>
<br><font size=2 face="sans-serif">Greets,</font>
<br><font size=2 face="sans-serif">Alex</font>
<br>
<br>
<br><tt><font size=3>>In such a situation, ocfs2 fences the higher node
number. afaik,<br>
>css does the same. What are the css node numbers for the two nodes?<br>
<br>
</font></tt><a href="http://oss.oracle.com/mailman/listinfo/ocfs2-users"><tt><font size=3 color=blue><u>>alexandra.strauss
at bayerbbs.com</u></font></tt></a><tt><font size=3> wrote:<br>
>><i><br>
</i>>><i> Hello,<br>
</i>>><i><br>
</i>>><i> I refer to you hoping you may help me with my problem...
We have got <br>
</i>>><i> an issur here and opened a SR at Metalink but until now,
we got no <br>
</i>>><i> useful information in solving our problem. SR-Number is
6855815.994...<br>
</i>>><i><br>
</i>>><i> We wanted to protect 9i Single-Instance Databases with
10g Clusterware <br>
</i>>><i> following the third-party-tool approach. There are no RAC-databases
<br>
</i>>><i> involved. But we want to achieve high availability as the
databases <br>
</i>>><i> are business critical systems. We want to make the systems
able to<br>
</i>>><i> relocate to another machine in case of failure to keep
downtimes <br>
</i>>><i> low... To achieve this we want to use OCFS2 for the filesystem.
<br>
</i>>><i> Relocate is done by script with help of CRS.<br>
</i>>><i><br>
</i>>><i> So we took two systems (byaz05 and byaz10) and installed
the following <br>
</i>>><i> software: 10g CRS (10.2.0.3) and Oracle Software 9.2.0.8
and OCFS2 1.2.8<br>
</i>>><i><br>
</i>>><i> We found the following Metalinknotes and adjusted the heartbeat
and <br>
</i>>><i> timeouts for OCFS2: Metalink Note 395878.1: Heartbeat/Voting/Quorum
<br>
</i>>><i> Related Timeout Configuration for Linux, OCFS2, RAC Stack
to avoid <br>
</i>>><i> unnessary node fencing, panic and reboot<br>
</i>>><i> Metalink Note 391771.1: OCFS2 - FREQUENTLY ASKED QUESTIONS
(hier <br>
</i>>><i> insbesondere der Abschnitt zu Fencing und Quorum)<br>
</i>>><i> Metalink Note 434255.1: Common reasons for OCFS2 Kernel
Panic or <br>
</i>>><i> Reboot Issues<br>
</i>>><i> Metalink Note 457423.1: OCFS2 Fencing, Network, and Disk
Heartbeat <br>
</i>>><i> Timeout Configuration<br>
</i>>><i><br>
</i>>><i> We did no changes to the CRS/CSS default settings until
now.<br>
</i>>><i><br>
</i>>><i> During HA-testing we watched unexpected behaviour of the
system. We <br>
</i>>><i> deactivated the bond for private interconnect and expected
only one <br>
</i>>><i> node to go down. But we faced both nodes going down. As
it seems to me <br>
</i>>><i> one node was rebooted from OCFS2 and the other one from
CRS/CSS.<br>
</i>>><i><br>
</i>>><i> Timestamp<br>
</i>>><i> --------------------------------------------------------------------------------------------------------------
<br>
</i>>><i><br>
</i>>><i> 10:21:06 bond1 disabled (eth1)<br>
</i>>><i> */var/log/messages byaz05*<br>
</i>>><i> Apr 25 10:21:06 byaz05 kernel: bonding: bond1: link status
definitely <br>
</i>>><i> down for interface eth1, disabling it<br>
</i>>><i> Apr 25 10:21:06 byaz05 kernel: bonding: bond1: making interface
eth5 <br>
</i>>><i> the new active one.<br>
</i>>><i><br>
</i>>><i> 10:21:09 bond1 disabled (eth5)<br>
</i>>><i> */var/log/messages byaz05*<br>
</i>>><i> Apr 25 10:21:09 byaz05 kernel: bonding: bond1: link status
definitely <br>
</i>>><i> down for interface eth5, disabling it<br>
</i>>><i> Apr 25 10:21:09 byaz05 kernel: bonding: bond1: now running
without any <br>
</i>>><i> active interface !<br>
</i>>><i><br>
</i>>><i> 10:21:23 o2net – no longer connected<br>
</i>>><i> */var/log/messages byaz05*<br>
</i>>><i> Apr 25 10:21:23 byaz05 kernel: o2net: no longer connected
to node <br>
</i>>><i> byaz10.bayer-ag.com (num 1) at 10.190.59.6:7777<br>
</i>>><i> */var/log/messages byaz10*<br>
</i>>><i> Apr 25 10:21:23 byaz10 kernel: o2net: no longer connected
to node <br>
</i>>><i> byaz05.bayer-ag.com (num 0) at 10.190.59.5:7777<br>
</i>>><i><br>
</i>>><i> 10:21:27 CSSD failure 134<br>
</i>>><i> 10:21:29 Reboot initiated by CRS<br>
</i>>><i> */var/log/messages byaz05*<br>
</i>>><i> Apr 25 10:21:27 byaz05 logger: Oracle clsomon failed with
fatal status <br>
</i>>><i> 12.<br>
</i>>><i> Apr 25 10:21:27 byaz05 logger: Oracle CSSD failure 134.<br>
</i>>><i> Apr 25 10:21:27 byaz05 su(pam_unix)[25839]: session closed
for user <br>
</i>>><i> oracle<br>
</i>>><i> Apr 25 10:21:27 byaz05 logger: Oracle CRS failure. Rebooting
for <br>
</i>>><i> cluster integrity.<br>
</i>>><i> Apr 25 10:21:27 byaz05 kernel: md: stopping all md devices.<br>
</i>>><i> Apr 25 10:21:27 byaz05 kernel: md: md0 switched to read-only
mode.<br>
</i>>><i> Apr 25 10:21:29 byaz05 logger: Oracle CRS failure. Rebooting
for <br>
</i>>><i> cluster integrity.<br>
</i>>><i> Apr 25 10:21:29 byaz05 kernel: e1000: eth2: e1000_watchdog_task:
NIC <br>
</i>>><i> Link is Up 1000 Mbps Full Duplex<br>
</i>>><i> Apr 25 10:21:29 byaz05 logger: Oracle init script ceding
reboot to <br>
</i>>><i> sibling 27383.<br>
</i>><i>><br>
</i>><i>> 10:21:58 Reboot initiated by OCFS2(?)<br>
</i>><i>> */var/log/messages byaz10*<br>
</i>><i>> Apr 25 10:21:58 byaz10 su(pam_unix)[4595]: session opened
for user <br>
</i>><i>> oracle by (uid=0)<br>
</i>><i>> Apr 25 10:21:58 byaz10 su(pam_unix)[4595]: session closed
for user oracle<br>
</i>><i>> Apr 25 10:25:58 byaz10 syslogd 1.4.1: restart.<br>
</i>><i>> Apr 25 10:25:58 byaz10 syslog: syslogd startup succeeded<br>
</i>><i>> Apr 25 10:25:58 byaz10 kernel: klogd 1.4.1, log source
= /proc/kmsg <br>
</i>><i>> started.<br>
</i>><i>> Apr 25 10:25:58 byaz10 kernel: Bootdata ok (command line
is ro <br>
</i>><i>> root=/dev/vgroot/_)<br>
</i>><i>><br>
</i>><i>><br>
</i>><i>> We supposed all the time this is a timing problem. But
we don't know <br>
</i>>><i> which settings raise the problem and which steps to do
to to correct <br>
</i>><i>> them. Otherwise we'll have to work over the complete concept
for the <br>
</i>>><i> business critical systems.<br>
</i>>><i> Can anyone help me?<br>
</i>>><i><br>
</i><br>
>><i> Regards,<br>
</i>>><i> Alexandra<br>
</i></font></tt>