We ran into an issue where our remote collectors only worked for a couple days and after that we had to reset (reboot) them to start collecting again. Instead of troubleshooting the issue we choose to redeploy the appliances with more resources (large).

We have 2 remote collectors per VCF site and we choose to replace them 1 by 1 so we always have 1 left in operation.

First i collected all the ip data so i can quickly deploy it again after removing it.

Next step is to remove the collector from the collector group so it would stop collecting data from adapters, you do need to check if there are adapters which are directly connected to the collector. In our case we have all adapters connected to collector groups so you do not run into this issue.

When to collector is removed from the group you can bring it offline in the admin console and then remove it

check the box!

After we removed it we triggered an inventory sync from LCM so it also understands it is no longer there, first time we ran it it gave an error about the removed node. Just retry it and it will succeed. Then power off the VM and remove it from disk.

Now we will trigger “Add Components” and add a new remote collector. Make sure you have the installed version of vROPS in your binary mappings.

this will also notify you to do a sync first!
fill in the details and click the advanced button
make sure the correct network and gateway are selected!

We had an issue with the default values of network being on xregion instead of region and the default gateway also being incorrect because of this. So be carefull that all the settings are correct otherwise it is a long wait until the firstboot timeout (30 mins approx).

Once everything is filled out you can finish the wizard and start the process.

Be carefull though because we ran into and issue with the certificate being changed to the Log Insight certs because LCM thought we had that one assigned to vrops which was not the reality and inventory syncs did not help with this so we had to start a “change certificate” action after the process is done to recitfy this. Both of these processes will restart your vROPS cluster!

Once all that is done make sure to put your new collector back into the Collector Group and check your adapters if everything is functioning again. Now you can rinse and repeat for every Collector you want to replace!

Leave a Reply

Your email address will not be published. Required fields are marked *

%d bloggers like this: