Take the 2-minute tour ×
Super User is a question and answer site for computer enthusiasts and power users. It's 100% free, no registration required.

I've got a CentOS 6 webserver running in a VM on ESXi 5. After about 133 days of uptime I rebooted it. Now it won't boot. It hangs here for a while:

enter image description here

Then finally:

enter image description here

If this were a real hard drive, I would assume it is dying and replace it. But it's a virtualized drive, and my other VMs are working just fine. Any idea how to troubleshoot this?

share|improve this question
2  
Have you checked the real hard drive? –  harrymc Feb 11 '13 at 6:37
    
Does the VM disk image still exist? Did you change the controller (ATA, Virtuso, SCSI)? –  kobaltz Feb 15 '13 at 2:32
    
@kobaltz: Nothing has changed in the VM configuration. –  Jonathon Reinhart Feb 15 '13 at 3:08
add comment

2 Answers 2

You don't say why you needed to reboot this VM.
Has some serious problem occurred with the host, such as an unplanned power-down?

I would first check for a real hardware disk problem. Try to see if you can copy the VM without any read errors, or if the copy now boots OK.

Secondly, check if the definition of the hard disk in the VM changed somehow. Look for any wrong parameter in the VM's configuration or in its BIOS, or perhaps the virtual disk had filled-up and needs to be enlarged, or it could even be a corrupted disk driver in the VM.

If you have upgraded ESXi in the last 133 days, try this VM on the earlier version.

Lastly, it is entirely possible that during the last 133 days some glitch has corrupted the VM's disk and rendered it unbootable, which you only found out now. It that is the case, you may need to reinstall CentOS or even rebuild the VM.

share|improve this answer
    
I was going to make some software configuration changes in the VM, so I had shut it down to take a snapshot. After powering it back up is when it refused to boot. –  Jonathon Reinhart Feb 17 '13 at 16:43
    
Don't you have a backup copy that works? If you do, you can then compare the VM's configuration file for any changes. –  harrymc Feb 17 '13 at 17:32
add comment
up vote 0 down vote accepted

To fix this, I essentially cloned the whole VM manually, using the ESXi CLI:

  1. Create a new directory, for the new VM, call it recover
  2. Clone the disks using vmkfstools --clonevirtualdisk source.vmdk dest.vmdk This will thick-provision the disks.
  3. Copy the .vmx file, and edit it to point at the new disks (if you renamed them).

Apparently there was just something wrong with the thin-provisioned disks and cloning them took care of things. Hope this helps someone.

share|improve this answer
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.