When we have a maintenance window and need to do our work in the Virtual Environment, we always find new issues. This time was in one of the ESXi 6.0 hosts.

This is an HP DL360 G7 with the ESXi 6.0 build 3620759.

When trying to use VMware update manager in this host, the scan runs without any problem, but when trying to Stage the updates, we get:

Stage patches to entity esxi01.localdomain. The host returns esxupdate error code:99.
An unhandled exception was encountered. Check the Update Manager log files and esxupdate log files for more details.

Looking at the esxpudate.log(in ESXi var/log/), I see a lot of failed tries to upload VIB files to the repository. Also this:

2016-04-24T15:43:54Z esxupdate: root: ERROR: Traceback (most recent call last):
2016-04-24T15:43:54Z esxupdate: root: ERROR:   File “/usr/lib/vmware/esxcli-software”, line 462, in <module>
2016-04-24T15:43:54Z esxupdate: root: ERROR:     main()
2016-04-24T15:43:54Z esxupdate: root: ERROR:   File “/usr/lib/vmware/esxcli-software”, line 453, in main
2016-04-24T15:43:54Z esxupdate: root: ERROR:     ret = CMDTABLE
2016-04-24T15:43:54Z esxupdate: root: ERROR:   File “/usr/lib/vmware/esxcli-software”, line 173, in ProfileGetCmd
2016-04-24T15:43:54Z esxupdate: root: ERROR:     raise Exception(“No host image profile defined”)
2016-04-24T15:43:54Z esxupdate: root: ERROR: Exception: No host image profile defined

I needed to check the Host Software profile(not the same as the Host Profile) and check the issue.

Then I noticed that the ESXi somehow lost its software profile, and it was not possible to stage any update or even install a VIB manually.

When troubleshooting the issue, I found out that there was no Software profile in this host.

I needed to check the vibs folder and profiles folder in /var/db/esximg. I noticed the two folders were empty. Also, checking the VIB list in this host, I get this:

When checking the same information from a working ESXi host, I get this:

And also this:

Searching on the internet for the same type of issue, I cannot find anything similar. VMware KB2043170 has a solution for the “error code 99” by Initializing VUM without reinstalling. But since this is a problem in one host, this could not be the solution for this issue.

Then found an article with a similar issue in Patrick Terlisten blog.

Since I didn’t want to reinstall the ESXi host at the moment, so I needed to try this solution. Even was not 100% similar, it was almost the same.

I followed Patrick’s recommendation to copy the esximg files/folders from a working host(esxi-02) to the corrupted ESXi host(esxi-01).

First, if you don’t have your SSH Client enabled in the host firewall, you need to allow to do the next task using the SCP command.

To enable SSH Client in the source ESXi host:

Note: Don’t forget to disable SSH Client after finishing these tasks.

After you run the SCP command, you will be prompted for the root password of the remote host, and once you have successfully authenticated, the file will copy.

After copying the files, check if the files are in the ESXi host(esxi01).

Host Software profile:

After checking that the files are copied into the destination, I rerun the VUM Scan into the host and after Stage, and no errors were found. After that, I remediated the ESXi host, and it is fully updated.

Again, thanks to Patrick for his solution that put me on the right track.

I hope this can help.

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