Search the VMware Knowledge Base (KB)
View by Article ID

ESX fails to boot with the error: Error 15: Could not find file (1004574)

  • 36 Ratings

Symptoms

  • Cannot start the ESX host after applying patches using Update Manager
  • Starting the ESX host after applying patches using Update Manager fails
  • vCenter Update Manager upgrades host, but does not reboot the host
  • The ESX host stops responding and you are unable log in to the host using the vSphere Client.
  • You see the error: 

    Error 15: Could not find file

  • The ESX host stops responding to the vCenter Update Manager task, but reboots normally and then experiences an error attempting to update VMware Tools on guests
  • You see an error which states that the VMware Tools files cannot be found due to lack of disk space
  • You see an error on the console screen (ALT-F12) indicating a lack of free inodes
  • vCenter Update Manager may report the below error code when this issue is encountered.

    Error Code: 19

Resolution

This issue may occur if the VMware Support bundles have been created and extracted on the host without first being cleaned. This can confuse the installation process and ESX modifies GRUB to point to the wrong location on the host. 
 
To work around this issue: 
  1. Reboot the ESX host.
  2. Select Service Console Only mode (ESX 3.5) or Troubleshooting mode (ESX 4.x) on the Grub menu.
  3. Verify that /usr/lib/vmware/hostd/docroot/downloads does not contain any previous vm-support dumps.

    To locate these files, run this command:

    find /usr/lib/vmware -name vm-support -print
     
  4. To remove these files, run the command:

    rm -rf /usr/lib/vmware/hostd/docroot/downloads/vm-support-*
     
  5. Log in as root, and run these commands:

    • In ESX 3.x only – esxcfg-boot -p 
    • In all versions of ESX – esxcfg-boot -b
    • In ESX 3.x only – esxcfg-boot -r 

       
  6. Reboot the ESX host.

Additional Information

If running the esxcfg-boot -b command fails, run vdf -h on the host to see how much free space is available in /boot. If there is very little space, such as 100 MB or less, check the /tmp and /vmimages folder for any old files or very large ISOs and remove them. Re-run esxcfg-boot -b. If you are able to run this command, reboot the host.
 
Note: ESX requires at least 24MB to boot properly.
 
You may also experience this issue after a successful upgrade, such as upgrading from ESX 3.5 to 4.0 and then performing an upgrade from 4.0 to 4.1. To resolve this issue, go to the troubleshooting mode and then run this command to clean the old ESX 3.5.x files:
 
cleanup-esx3 –f
 
Note: After you use this command, you cannot revert to ESX 3.5.x.
 
For more information on troubleshooting this issue, see   Troubleshooting an ESX host that does not boot (10065)
.

For ESX 4.x, check the /boot/trouble directory for any files that may be present and remove them to free up space:

To remove files from 
the /boot/trouble directory, run these commands:

#cd /boot/trouble
#ls
initrd.img  vmlinuz

#rm -r *


Note: VMware High Availability agents may not correctly configure after the failure to boot and in the Summary tab, you can see that the HA configuration failed. In this case, in vCenter Server, right-click the host and click Reconfigure HA.

See Also

Update History

04/01/2011 - Expanded details on what causes this issue. 12/01/2011 - changed /var/lib/vmware/hostd/docroot to /usr/..., this was validated. 05/17/2012 - Added steps to check /boot/trouble in additional information

Request a Product Feature

To request a new product feature or to provide feedback on a VMware product, please visit the Request a Product Feature page.

Feedback

  • 36 Ratings

Did this article help you?
This article resolved my issue.
This article did not resolve my issue.
This article helped but additional information was required to resolve my issue.

What can we do to improve this information? (4000 or fewer characters)




Please enter the Captcha code before clicking Submit.
  • 36 Ratings
Actions
KB: