Now that the UCS Manager Infrastructure Components are complete we need to move over to the HX Data Platform. The next step is to go through the manual bootstrap upgrade process. The following steps will upgrade the HX Data Platform and the Data Platform plug-in.
Open WinSCP and connect to the Cluster Management IP Address and login as the root user. Then we need to copy the HX Data platform upgrade bundle to the /tmp directory of the controller VM.
Next we need to open up Putty and connect to the Cluster Management IP address as the root user. Once logged in browse to the /tmp directory where we copied the installer bundle.
Then run the following command to un-compress the package,
tar –zxvf storfs-packages-3.5.1a-31118.tgz
Then run the following command to bootstrap the package for the upgrade,
You will be prompted to enter the vCenter FQDN and a username and password with administrator privileges. I have provided an image below showing each of the prompts but I have removed my entries.
After the script has run the System Management Service will restart and the bootstrap process will then be complete. You can now check the Plug-in is now updated in the vsphere client.
Next you need to log into the same Cluster Management IP address which will take you to the HX Connect client. You then need to login with the username admin as this has now changed for this new release. On the left hand side menu under Manage click on Upgrade
Then tick the UCS Server Firmware box and the HX Data Platform Box.
Next fill in each of the fields presented underneath UCS Server Firmware and then drag and drop a copy of the upgrade bundle we used earlier to run the manual bootstrap process. Note – The manual bootstrap process will no longer be required for future upgrades beyond the 3.5 release we are upgrading too.
After filling in the UCS Manager details you will need to click on the Discover Button. This will located the current server firmware version and allow you to select the version you would like to upgrade to from the drop down menu.
Once complete you should see the following screen
HX Connect will then go through a validation process and check each component to ensure it will be able to perform the upgrade. If any errors are found it will raise them here and allow you to resolve them before trying again.
Once the validations are complete it will run through upgrading each of the hosts in the cluster one at a time. It will display each of the steps it is running through as the host is upgraded. One thing to note is that if you have any VMs you have pinned to a certain host or if you have other VM rules in place you will need to address this before proceeding as you may get errors in the validation.
The process will take care of the hosts and the VMs in the background automatically. The VMs will be migrated off the host and it will enter HX maintenance mode before performing the upgrade. It will go through each host in the cluster until each one is upgraded.
You can monitor the progress of each host from the vsphere client and UCS Manager. You will still see prompts in UCS Manager about Pending Activities in relation to rebooting hosts but you won’t need to do anything with them. The process will take care of itself. Time to sit back and let the upgrade run its course!
Post Upgrade Checks
Log into UCS Manager and check the running firmware version for all of your components by expanding out each node.
Ensure that everything has been upgraded and there are no pending tasks or errors to resolve. I haven’t included a screenshot of everything we have connected to UCS Manager but expand out and check the servers you have running on your infrastructure.
Also log into HX Connect via the Cluster Management IP Address and check over the main dashboard for any alerts. When you log in you will notice that HX Connect now displays 3.5(1a) under the Hyperflex Logo.
Then to confirm that everything is now back online and healthy switch to your vSphere Web Client. Click On Home, then vCenter Inventory Lists, then Cisco Hyperflex Systems, then Cisco HX Data Platform, then Cluster and then Summary. This should now show the Cluster Status is Online and the Health Status is Healthy.
That’s it your all done! Thanks for checking out my upgrade guide!