VCSA Upgrade from 6.0 to 6.7 - Fails Pre-Upgrade Check Result on Stage 2

[DanPLewis]

Hey all,

Of course nothing ever comes easy! I was in the process of attempting to upgrade our VCSA appliance from 6.0 to 6.7. It creates the new appliance, and on stage to it performs another pre-upgrade check. Unfortunately now I'm failing that check with the following:

Error: Internal error occurs during execution of upgrade process.

Resolution: Please search for these symptoms in the VMware knowledge base for any known issues and possible resolutions. If none can be found, collect a support bundle and open a support request.

Error: Cannot validate target appliance configuration as not enough information from the source appliance can be collected. For more details check out the server logs.

Error: Internal error occurs during pre-upgrade checks.

Resolution: Please search for these symptoms in the VMware Knowledge Base for any known issues and possible resolutions. If none can be found, collect a support bundle and open a support request.

Warning: This ESXi host [HOSTNAME] is managed by vCenter Server [IPADDY].

Resolution: Make sure the cluster where this ESXi host resides is not set to Fully Automated DRS for teh duration of the upgrade process.

With these error messages being pretty generic, I'm able to click "Logs" and download a log TGZ bundle, or click close. I've reviewed some of the logs and nothing popped out at me at all. Any thoughts on where I can begin to troubleshoot this? I have tried setting the target server on two different hosts but that didn't help at all.

[Vijay2027]

Attach logs to this thread. I will take a look.

[DanPLewis]

Thanks Vijay. Any specific logs you're looking for? I'll have to redact them first.

[Vijay2027]

For now pre-checks.json file.

And can you share the list of files that are available in log directory.

[Vijay2027]

and any file CollectRequirements_com.vmwarexxx????

[DanPLewis]

Vijay,

I've gone ahead and redacted / uploaded the prechecks.json file. Unfortunately I cannot find any log that was output with the name CollectRequirements.

Thanks!

[MikeStoica]

Is DRS disabled?

[DanPLewis]

Mike,

It is disabled. We aren't even operating with a cluster, just running a site with three hosts.

Thanks!

[DanPLewis]

There were a few different things that I had tried, that had brought me here.

1. After attempting from a different machine to upgrade and re-reviewing those logs, I noticed the following:
   
   2018-11-08T14:29:38.388Z ERROR __main__ ERROR: Fatal error during upgrade REQUIREMENTS. For more details take a look at: /var/log/vmware/upgrade/requirements-upgrade-runner.log
   
   
2. After reviewing /var/log/vmware/upgrade/requirements-upgrade-runner.log, I noted the following:
   1. 2018-11-08T14:29:10.624Z INFO upgrade.states.component_states eam:CollectRequirements:             "id": "install.ciscommon.service.failstart",
   2. 2018-11-08T14:29:10.625Z INFO upgrade.states.component_states eam:CollectRequirements:             "localized": "An error occurred while starting service 'vmware-eam'",
   3. 2018-11-08T14:29:10.625Z INFO upgrade.states.component_states eam:CollectRequirements:             "translatable": "An error occurred while starting service '%(0)s'"
   4. 2018-11-08T14:29:10.625Z INFO upgrade.states.component_states eam:CollectRequirements: WARNING: VMware ESX Agent Manager may have failed to start
      
      
3. In this log, it requested I look at "CollectRequirements_com.vmware.eam_2018_11_08_09_28.log, which I found and have the following:
   1. "localized": "An error occurred while invoking external command : 'Command: ['/sbin/service', u'vmware-eam', 'start']\nStderr: '",
   2. "translatable": "An error occurred while invoking external command : '%(0)s'"

Thoughts?

[DanPLewis]

Update: This kind of feels like a red herring, this isn't something we're using here. This service doesn't have any information under it, and on our other VMWare Setup it is the same however the service does in fact start. I'm not sure if this is accurate to the issue at hand or not.

[Vijay2027]

Can you verify eam service status and eam.properties file in 6.0

VMware Knowledge Base

[DanPLewis]

Vijay,

The service status: VMware ESX Agent Manager is not running.

The eam.properties file was empty in 6.0. I did try rectifying that by following this, however it still failed:

VMware Knowledge Base

I'm still unsure how big of an issue this is however since we are not using EAM.

[jkghfjkasdhfjka]

What build are you using for 6.7?  6.7 u1?  Was the 6.0 vCenter previously upgraded or migrated to from an older version?

[DanPLewis]

We are using VMWare VCSA 6.7.0-9451876 (Not u1, which for us wouldn't be compatible with VEEAM.

As per the upgrade status, I was informed that this was installed fresh as 6.0, however update 1 and 2 were performed via Update Manager.

[Vijay2027]

Though you do not use the service, if it is not functional  the upgrade will fail with eam firstboot.

Eam service should be corrected before you retry the upgrade.

[DanPLewis]

I guess the next big question is how? I'm continuing to peruse through guides in order to try and find that.

[DanPLewis]

Vijay,

This issue has now been resolved. The issue I was having was I had some additional information inside the eam.properties file that should not have been there. I thought they were comments at the end of the line so I removed them, the service started, and now VCSA is upgraded.

Thanks all for your help!