VMware Cloud Community
BNixon1980
Contributor
Contributor
Jump to solution

vROPS 6.7 Successful Upgrade

Running a single node with two disks at 1TB each. Database is currently using roughly 70% or 1.3TB of space. Opened a ticket with VMware as our upgrade continuously ran for a couple of days without completing. Their tech had a similar situation where the customer's upgrade took 3 days and had a 500GB database. We are currently on Day 5 of our second attempt, and it is still running. We would love a progress bar, with a percentage completed.

Anyone run into similar issues? What was your final outcome? VMware has instructed us to wait it out for a few more days, they have seen a few failed upgrade attempts per week and VMware should be working on a 6.7.1 in the very near future to remediate those issues.

1 Solution

Accepted Solutions
BNixon1980
Contributor
Contributor
Jump to solution

VA Upgrade

Procedure

Before installing console into VA >  service liagentd stop > chkconfig > liagentd OFF (to disable service from restarting during reboot)

1.In a Web browser, navigate to the vRealize Operations Manager administration interface at https://master-node-name-or-ip-address/admin.

2.Log in with the admin username and password for the master node.

3.On the left, click Software Update.

4.Click Install a Software Update.

5.Follow the wizard to locate and install your copy of update-filename.pak.  Installation completes in a couple of minutes, and the administrator interface logs you out. If you are not logged out automatically after 5 minutes, refresh the page in your browser.

6.Log back in to the master node administrator interface, and click Software Update again.

7.Verify that update name appears on the right. If the update does not appear, wait a few minutes, and refresh the page in your browser.

This allowed our team to successfully upgrade to 6.7. On a side note, once cluster was online and vROPS was accessible again collections took a very long time to go green/green for collection state and collection status. There was also an issue with the vCenter Adapter needing the thumbprint to be re-registered before collections would begin again.

View solution in original post

Reply
0 Kudos
6 Replies
OsburnM
Hot Shot
Hot Shot
Jump to solution

I've had problems in the past where the upgrade just appeared to be stuck and not really doing anything.  Turned out to be the loginsight rpm installer was hung on the vROps data nodes. You could see in TOP on the datanodes the liagent rpm installer process was just sitting there.  Just killing the process wouldn't let the upgrade resume; so we reverted back our snapshots and after stopping the liagent process BEFORE starting the upgrade, it took right off.

BNixon1980
Contributor
Contributor
Jump to solution

So we let the upgrade continue all weekend without success. Just reverted to our snapshot and are currently upgrading the OS portion. Going to follow your advice and kill "liagent" process before installing the VA upgrade .pak. Lets see if this works...

Reply
0 Kudos
BNixon1980
Contributor
Contributor
Jump to solution

VA Upgrade

Procedure

Before installing console into VA >  service liagentd stop > chkconfig > liagentd OFF (to disable service from restarting during reboot)

1.In a Web browser, navigate to the vRealize Operations Manager administration interface at https://master-node-name-or-ip-address/admin.

2.Log in with the admin username and password for the master node.

3.On the left, click Software Update.

4.Click Install a Software Update.

5.Follow the wizard to locate and install your copy of update-filename.pak.  Installation completes in a couple of minutes, and the administrator interface logs you out. If you are not logged out automatically after 5 minutes, refresh the page in your browser.

6.Log back in to the master node administrator interface, and click Software Update again.

7.Verify that update name appears on the right. If the update does not appear, wait a few minutes, and refresh the page in your browser.

This allowed our team to successfully upgrade to 6.7. On a side note, once cluster was online and vROPS was accessible again collections took a very long time to go green/green for collection state and collection status. There was also an issue with the vCenter Adapter needing the thumbprint to be re-registered before collections would begin again.

Reply
0 Kudos
BNixon1980
Contributor
Contributor
Jump to solution

Thank you for your help, we had two failed attempts in almost two weeks and this was the fix!

OsburnM
Hot Shot
Hot Shot
Jump to solution

Awesome! Can you PM me the SR you had with VMware (if you opened one)?  We actually ran into this on two different version upgrades and support never did figure it out themselves. Seeing others with the same problem makes me think it’s Now bigger than just us and I’d like to get our TAM to see about getting a KB built for this.

BNixon1980
Contributor
Contributor
Jump to solution

Not sure if my PM worked, here you go: VMware Support Request 18785024904

Reply
0 Kudos