ESX 3.5 new install error

[rmaccara]

I've received this error with both 3.5 Update 1 and 2 installs. The system runs without error for a short period of time, then gets this error message:

0:16:58:39:362 cpu0:1024) VMNIX: <0> journal commit I/O error

If I log in, I get this message scrolling continuously:

EXT3-fs error (device sd(8,7)) in start_transaction: Journal has aborted

Once the error starts, I lose the host (not responding) in Virtual Center.

I've got two brand new identical Dell R805 boxes and the error only happens on one of them. They are both connected to a new Dell iSCSI MD3000i box.

I'm very new at this and don't know where to start to hunt this error down (other than searching unsuccessfully for others that have had the same problem).

Can anyone point me in the right direction?

Thanks,

Rob

[Craig_Baltzer]

This looks like one of the disk volumes is having errors. The error is coming from the EXT3 file system, so that points to whatever set of disks you have installed ESX 3.5 on (as opposed to VMFS volumes you've created on the SAN to store your VMs). Prob. the best bet is to check for any bad drives (i.e. something with a red light on it), confirm the appropriate controller settings, BIOS and firmware releases (i.e. matches your other machine that is working), and then look for some kind of "disk exerciser" to stress the disks and verify that they're working correctly over an extended run (i.e. run the utility for 72-96 hours straight).

[Karunakar]

Hi,

Usually these errors occur when the disk is currupted or the file system is gone bad.

but the error itself shows that the file system journal has gone bad.

Have you tried to perform a reboot of the machine after seeing this error, as up on a reboot, the file system check if performed when the machine boots up again.

but back up all the information about the VM's running on that before performing a reboot.

-Karunakar

[rmaccara]

Thanks for your suggestions!

I've tried another re-install today and while doing so wiped the existing drive mirroring and created a new one and ran the controller initialization process.

I still get the same error, so will start over again and double check the controller settings and run the longer more intense initialization test.

It's very frustrating, because I'm new at this and second guess if I've made an error in the setup or there's a hardware error!

Thanks again!

Rob

[Craig_Baltzer · Accepted solution]

I think you're on the right track looking at the hardware or hardware configuration; there really isn't anything in the installation that you can do if you're following through with the partitioning defaults to cause something like this...

[rmaccara]

The only thing I'm doing that is not "default" during the partitioning phase is to add a tmp and home folder and delete the VMFS3 partition.

I re-installed and did not add tmp or home folders, and so far no errors. It occured to me that the errors were on sd8/7 which were the extra folders created during install.

I'll wait till the morning and see if the error has come back. If it does, I'll start pulling hair out then!

Rob