I recently upgraded 5 of my production servers to 3.5U2. Four of them have gone with no problems. The last one is causing me major headaches by dropping out of VC with a "not responding" during HA migrations. (I did reformat and setup the servers from scratch, I got tired of non-clean upgrades)
In /var/log/messages I'm seeing the following error over and over again
Sep 15 13:55:27 esx3 cimprovagt: IpmiIfcFruMulti: checksum of header failed. Checksum = 0x16
Sep 15 13:55:28 esx3 last message repeated 3 times
And it's usually followed by:
Sep 14 13:57:41 esx3 cimprovagt: IpmiIfcFruMulti: checksum of header failed. Checksum = 0xb8
Sep 14 13:57:42 esx3 last message repeated 3 times
Sep 15 13:57:37 esx3 watchdog-hostd: '/usr/sbin/vmware-hostd -u' exited after 11588 seconds
Sep 15 13:57:37 esx3 watchdog-hostd: Executing cleanup command '/usr/sbin/vmware-hostd-support'
Sep 15 13:57:41 esx3 watchdog-vpxa: '/opt/vmware/vpxa/sbin/vpxa' exited after 11591 seconds
Sep 15 13:57:41 esx3 watchdog-vpxa: Executing '/opt/vmware/vpxa/sbin/vpxa'
Sep 15 13:57:48 esx3 watchdog-hostd: Executing '/usr/sbin/vmware-hostd -u'
I've disabled ipmi via chkconfig and stopped the ipmi service. Unfortunately, the box keeps dropping out of VC during migrations so I can't get the VMs off it to reboot.
We seem to have a similar problem on an ESX 3.5U3 system!
"Mar 29 03:22:21 esx35302 cimserver: IpmiIfcFruMulti: checksum of header failed. Checksum = 0x4"
This message appears every 5 minutes in /var/log/messages.
I have also installed our ESX-server from scratch. The only reason I think of is that I had an existing VMFS volume where all our virtual machines are running on. Maybe the older VMFS is the reason for this message!?
Which version of VMFS do you have? We have version 3.21. I know that ESX 3.5U3 usually uses VMFS version 3.31. So this could be the problem. I thought of ESX converting to VMFS 3.31 automatically but it didn't. Is there a possibility to convert the old VMFS to a new VMFS without loosing data manually?