/ESXi 6.0 fix corrupted host imageprofile

ESXi 6.0 fix corrupted host imageprofile

In a previous article, we have fixed some VUM errors and the ESXi profile issues.

Again, we had some issues with the host when applying updates, and it was impossible to remediate them.

In the initial reboot(after applying the updates) host did not power up. It froze with this error:

ESXi 6.0 fix corrupted host imageprofile

We did a problem before. I wrote another article regarding this “fatal error 15”.

Need to reboot the ESXi host again, then the ESXi did start, but no updates were installed. When we try again to stage the updates, we get the “error code 15”. I discuss this issue in this article. I see this type of error in exupdate.log

Cannot locate source for payload uc_amd of VIB VMware_bootbank_cpu-microcode
Failed to send vob install.stage.error: [Errno 27] File too large

esxupdate: BootBankInstaller.pyc: DEBUG: Creating an empty ImageProfile for bootbank /bootbank
esxupdate: HostImage: DEBUG: Live image has been updated but /altbootbank image has not. This means a reboot is not safe.
esxupdate: HostImage: DEBUG: Live image has been updated but /altbootbank image has not. This means a reboot is not safe.

Cannot locate source for payload b of VIB VMware_bootbank_esx-base_6.0.0-2.37.3825889

So it seems that the ESXi image profile was corrupted. Following the tasks from the previous articles did not fix the issue. The problem is in the host boot image.

So I did start to troubleshoot the host image profile.

No host profile in the ESXi host, even when copy files from a working ESXi host after a reboot, did not fix the issue. Always get the same problem.

So we need then check the /bootbank image (imgdb.tgz).

First, we need to check where the imgdb.tgz is located but run a find command.

The system found two imgdb, need to check which one is the correct one (the other is just a backup) and see if the file is corrupted or not.

I see the one with only 183 bytes is the one used in the /bootbank. So I immediately suspected that this one was a corrupted file since the file is at least 60Kb.
So I check the same image from a working ESXi and see the size.

So as we can see, this is the standard size for a working imgdb file from a working ESXi.

So to fix this, we will copy the same file from a working ESXi to our fault ESXi.

Before copying the image from the working ESXi host, let’s make a copy of the corrupt image to the tmp folder and remove the corrupt file.

Before copying the files between hosts, don’t forget again to set the firewall rule to allow the copy. For security reasons, after you do this task, disable it again.

Just copy the file over to the corrupted host.

Note: Even we have already copied the Profile and VIBs files from the working ESXi (tasks explained in the previous articles and above), we can also do the next step to copy those files into the corrupted ESXi host using the imgdb.tgz file. For that, we need to extract the file and copy the folders.

You could skip the next step if you copied the files already following the previous articles and tasks.

Next, do a backup of the changes we made to the bootbank host image profiles.

Next, reboot the ESXi host, apply the patches, or use the VUM again.

In this case, since we copy an image and profile from an ESXi that had already the bugfix ESXi600-20160500 applied, now in this fixed host, it recognizes this bugfix is already installed. That is not correct. The option was to use the bugfix again manually.

Just upload to our ISO Datastore and run the vib install command.

But then we did get another issue: [Errno 2] No such file or directory: ‘/bootbank/conrep.v00’

Checking the file “conrep.v00” I see that doesn’t exist in the /bootbank of the corrupted ESXi. So decided to copy this again from the working ESXi this file.

Run the patch and install it again.

No errors were found this time.

Reboot the host, and everything was ok again with this ESXi and is up to date.

I hope this can help you bypass this error in the future.

Share this article if you think it is worth sharing. If you have any questions or comments, comment here or contact me on Twitter.

©2016 ProVirtualzone. All Rights Reserved
By | 2022-03-29T12:40:06+02:00 July 3rd, 2016|VMware Posts, vSphere|12 Comments

About the Author:

I am over 20 years’ experience in the IT industry. Working with Virtualization for more than 10 years (mainly VMware). I am an MCP, VCP6.5-DCV, VMware vSAN Specialist, Veeam Vanguard 2018/2019, vExpert vSAN 2018/2019 and vExpert for the last 4 years. Specialties are Virtualization, Storage, and Virtual Backups. I am working for Elits a Swedish consulting company and allocated to a Swedish multinational networking and telecommunications company as a Teach Lead and acting as a Senior ICT Infrastructure Engineer. I am a blogger and owner of the blog ProVirtualzone.com


  1. Brandon 19/08/2016 at 18:57

    Thanks for the post! The corrupt imgdb.tgz was my problem. Running ESXi 6.0 Update 2, HP custom ESXi image.

    • Luciano Patrao 20/08/2016 at 00:03

      Glade to help Brandon.

    • Luciano Patrao 11/06/2017 at 18:46

      Hi Brandon,

      Only today I notice that my reply is not sending emails to users that comment on my blog. So now I am just FYI to you regarding my comment.

      Thank You

      Luciano Patrao

  2. Josh 08/11/2016 at 15:16

    Thanks for this! This solved my issue quickly in a production environment!

    • Luciano Patrao 09/11/2016 at 09:09

      Hi Josh,

      Thanks. Glade that I could help.

      PS: Share if you think can help others.

      Thank You

      Luciano Patrao

    • Luciano Patrao 11/06/2017 at 18:46

      Hi Josh,

      Only today I notice that my reply is not sending emails to users that comment on my blog. So now I am just FYI to you regarding my comment.

      Thank You

      Luciano Patrao

  3. Fida 15/11/2016 at 02:10

    We have two HP Proliant BL460c Gen8 and one HP Proliant DL380 Gen( configured in a HA cluster. We upgraded BL460c from ESXi 5.5 to ESXi 6.0 (HP Customized Image) successfully. But we could not upgrade HP DL380 to ESXi 6.0 as it kept giving an error message ” Could not obtain module order from esx image db: (“/vmfs/columes/e88c0deb-2ab95e1a-d61f-0b6dbf28e3a5/imgdb.tgz’, “Error writing tar database: [Errno 2] No such file or directory: ‘/vmfs/volumes/e88c0deb-2ab95e1a-d61f-0b6dbf28e3a5/imgdb.tgz.new’.”) We tried to upgarde HP DL380 to HP Custom image and also tried Vmware normal ESXi 6.0 image. And we also tried a fresh install of ESXi 6.0 but we get the same error again and again.
    Could you please assist us in resolving the issue and what could be the cause of issue.? I would be grateful to you.

    • Luciano Patrao 16/11/2016 at 15:06


      Have you check that partition? Does the file exist?
      Are you doing a manual upgrade, or using VMware Update manager?

      Did you follow the recommendations and procedures of this article to correct the imgdb.tgz file?

      Can you provide the logs to investigate deeper?

  4. Ytsejamer1 26/07/2017 at 20:00

    God bless you sir! This was a lifesaver. I’d been hopping from one code 99 error (used your other article – http://www.provirtualzone.com/esxi-6-0-reports-vm-during-stage-patches-to-entity-vum-operation/), to error code 15 (used your other article – http://www.provirtualzone.com/esxi-6-0-reports-error-code-15-during-remediate-update-in-vum-operation/ ), to finally here to resolve the last error code 15. After rebuilding the bootbank (https://kb.vmware.com/kb/2033564), I ran over this article to help me with the last point.

    VMware support was poking at it with me and I found that my NetAppNa.v00 file needed to be moved over from a good host to the problematic one. Once I did that we were finally able to stage the updates and install the updates normally using VUM.

    For the record, I have HP BL460c G9 servers which used the HP U3 iso to install ESXi.

  5. Peter G. 19/09/2017 at 01:24

    Thank you, thank you, thank you! Had a 6.5.0 host that VUM (on VCSA 6.5.0u1) kept failing to scan. Host’s /bootbank/imgdb.tgz was about 150 bytes (not kBytes!) and both /var/db/profiles and /var/db/vibs were empty.

    Solved it by copying over /bootbank/imgdb.tgz from a “good” host to the bad one, extract the profile and VIBs into those empty directories in /var/db, run /sbin/auto-backup.sh as you described in this article, and everything worked again.

    • Luciano Patrao 19/09/2017 at 03:47

      Hi Peter,

      Glad to help.
      Don’t forget to backup, reboot and then rerun the scan(apply any updates needed) and again restart the ESXi host. Also always copy the profile and VIBs from an ESXi host that is running in the same build number(that as not been updated yet). If you copy from a newer build, the corrupted ESXi host thinks is updated and is not.

      Because sometimes the issue with host image profile is not fixed. Only after some reboots, we are sure that is fix.

      Thank you for feedback.

      Luciano Patrao

Leave A Comment