However what if you want to scale up your node or nodes, going from say small to medium?
- Perhaps your POC has proved so useful you want to move it into production?
- You have lots of really useful historical on your existing smaller build and you don’t want to redeploy (therefore losing your historical data)
- You have limitations where scaling out is simply not an option but you need vROps to take on more work.
There are likely other compelling reasons for you to want to scale up your vROps nodes so lets look at how we can do that.
First of all it’s important to know it is indeed supported to carry out this action but you really need to give it some careful consideration. Before electing to do this work I would urge you to take a look at the sizing spreadsheets VMware have made available so you can be sure you are making the right decision.
Once you know what size you are going up to you will need to consider the Hardware version, for instance the medium sized nodes is version 7 limiting it to 8vCPU so you will NEED to increase the Hardware version to accommodate any increase in node size.
Finally the process to to increase the node size is quite straight forward.
- First of all take the cluster offline in the admin section. (this is critical)
- Gracefully shutdown each of the nodes until all of the vROps appliances are powered off.
- Increase the vCPU, RAM and disk where required.
- Power on your Master node first allow a few minutes..
- Power on your HA replica (if you have one) and wait a few minutes..
- Power on the data nodes..
- Once they are all up and running verify the cluster can see all of the nodes and that they are ready for the cluster to be switched back on.
- Switch on the vROps cluster.
- Job done.
It goes without saying vROps is a beast so make sure your hardware can handle the increase (hosts and storage) and be mindful of the reference architecture.
At present I am not aware of any KB article from VMware for this.. but as soon as I find one I will update this post.