Enthusiast
Enthusiast

[vRA 7.3] - Upgrade 7.3.1 failed

Getting this right at the start of an upgrade attempt 7.3-7.3.1 from the .iso.

Vendor:

VMware Inc.

Appliance Name:

VMware vRealize Appliance

Appliance Version:

7.3.0.536 Build 5610496

(

Details...

)

Last Check:

Thursday, 2018 April 19 15:16:39 UTC-4 (Using update CD found on: /dev/sr0)

Pre-install: failed

The Management Agent was not upgraded on the nodes listed below. Make sure the Management Agent is up and running on them and that it has connectivity to the VA.

<servername>

RPM Status 1: Pre install script failed, package test and installation skipped.

Post-install: failed

Update failed (code 1-1). Check logs in /opt/vmware/var/log/vami or retry update later.

Last Install:

Failed to install updates(Error while running pre-install scripts and post-install scripts) on Thursday, 2018 April 19 15:20:31 UTC-4

Pre-install: failed

The Management Agent was not upgraded on the nodes listed below. Make sure the Management Agent is up and running on them and that it has connectivity to the VA.

<servername>

RPM Status 1: Pre install script failed, package test and installation skipped.

Post-install: failed

Update failed (code 1-1). Check logs in /opt/vmware/var/log/vami or retry update later.

0 Kudos
3 Replies
Immortal
Immortal

I've run into this one before. Although the presentation is the same, I cannot be 100% in your case. What you want to check is the management agent binary versions on the IaaS node(s). What I saw happening was that the management agent was, in fact, being updated, but something in the config was hosed which caused it to fail when checking back into the VA. This caused the updater to believe, since it couldn't contact the agent, the upgrade had failed. If this turns out to be your case, I'd open a SR with GSS. I just stood up a new version after trying lots of things to clear it up.

0 Kudos
Enthusiast
Enthusiast

Just saw this and had the same issue (except going from 7.3.0 to 7.4.0).  It took me a while to get through, but here is what I found...

Update Fails to Upgrade the Management Agent

This doc shows that if you have management agent issues, reinstall the agent.  (BTW, in step 1, the correct URL is https:// va-hostname.domain.name:5480/installer ,not https:// va-hostname.domain.name:5480/install).

We attempted the install, which failed the same way yours did.  (management agent not responding). to resolve, we downloaded the agent and reinstalled.  I believe that since part of the upgrade actually succeeded, the msi at 5480:installer was version 7.4.0 (not 7.3.0 which the appliance was still, which kinda bothered me since if this failed, I expected the installer to undo anything it did).  When we installed, it only asked for the user name and password for the service to run.  The agents installed successfully and showed communicating in the vami, but when we re-ran the install, it failed at the next spot.  Here is a snippet of the log:

Pre installed agents seemed to work:

+ /etc/bootstrap/preupdate.d/00-00-03-upgrade-management-agents 7.3.0.536 7.4.0.645

Current version is 7.3.0.536

Found 3 IaaS nodes with ids: {agentids removed}

The new version for iaas nodes is: 7.4.0.13461

Nodes to upgrade the management agent on:

There are no management agents that need upgrading

+ log '/etc/bootstrap/preupdate.d/00-00-03-upgrade-management-agents done, succeeded.'

++ date '+%Y-%m-%d %H:%M:%S%:z'

+ echo '2018-04-23 16:23:40-07:00 /etc/bootstrap/preupdate.d/00-00-03-upgrade-management-agents done, succeeded.'

+ for script in '"${bootstrap_dir}"/*'

But now the IaaS failed.

+ echo

+ '[' '!' -e /etc/bootstrap/preupdate.d/00-00-04-prereq-iaas ']'

+ '[' '!' -x /etc/bootstrap/preupdate.d/00-00-04-prereq-iaas ']'

+ log '/etc/bootstrap/preupdate.d/00-00-04-prereq-iaas starting...'

++ date '+%Y-%m-%d %H:%M:%S%:z'

+ echo '2018-04-23 16:23:40-07:00 /etc/bootstrap/preupdate.d/00-00-04-prereq-iaas starting...'

+ /etc/bootstrap/preupdate.d/00-00-04-prereq-iaas 7.3.0.536 7.4.0.645

+ res=1

+ echo 'Script /etc/bootstrap/preupdate.d/00-00-04-prereq-iaas failed, error status 1'

+ exit 1

Just as I was about to follow daphnissov's advice and rebuild (why has save me many, many times), I noticed in the vami update tab:

Pre-install: failed (code p-1)

The prerequisites of the following nodes cannot be completed:

servername1

No valid endpoints found in the Management Agent's configuration. Details: Trusted connectivity validation failed for address "https://server1:5480/": Thumbprint mismatch for address "https://servername1:5480/" and thumbprint "{thumbprintRemoved}".

servername2

No valid endpoints found in the Management Agent's configuration. Details: Trusted connectivity validation failed for address "https://server2:5480/": Thumbprint mismatch for address "https://servername2:5480/" and thumbprint "{thumbprintRemoved}".

RPM Status 1: Pre install script failed, package test and installation skipped.

Post-install: failed

Update failed (code 1-1). Check logs in /opt/vmware/var/log/vami or retry update later.

We decided to re-install the agent again, but when we ran it, it only gave us the option to remove it. We uninstalled, rebooted, then reinstalled.  This time it asked us to load and confirm the thumbprint (which was different).  After we verified the agents reporting (the only time the agents stopped reporting was when they were uninstalled), we re-ran the installation, and it completed.

So in short, try uninstalling the Management agents from the endpoints, reboot, then reinstall the agents, making sure that the thumbprints match. Hope this helps.

0 Kudos
Enthusiast
Enthusiast

@MikeNox

Will give that a try... still working out a failed 7.4 upgrade on the same environment with support.

We only have so many vra deployments to test this on.

:smileylaugh:

Thanks for the feedback.

0 Kudos