vmexplorer

Home Lab Generation 7: Upgrading vSAN 7 Hybrid capacity step by step

Posted on September 21, 2020 Updated on November 9, 2021

My GEN5 Home Lab is ever expanding and the space demands on the vSAN cluster were becoming more apparent.  This past weekend I updated my vSAN 7 cluster capacity disks from 6 x 600GB SAS HDD to 6 x 2TB SAS HDD and it went very smoothly.   Below are my notes and the order I followed around this upgrade.  Additionally, I created a video blog (link further below) around these steps.  Lastly, I can’t stress this enough – this is my home lab and not a production environment. The steps in this blog/video are just how I went about it and are not intended for any other purpose.

Current Cluster:

• 3 x ESXi 7.0 Hosts (Supermicro X9DRD-7LN4F-JBOD, Dual E5 Xeon, 128GB RAM, 64GB USB Boot)
• vSAN Storage is:
  • 600GB SAS Capacity HDD
  • 200GB SAS Cache SDD
  • 2 Disk Groups per host (1 x 200GB SSD + 1 x 600GB HDD)
  • IBM 5210 HBA Disk Controller
  • vSAN Datastore Capacity: ~3.5TB
  • Amount Allocated: ~3.7TB
  • Amount in use: ~1.3TB

Proposed Change:

• Keep the 6 x 200GB SAS Cache SDD Drives
• Remove 6 x 600GB HDD Capacity Disk from hosts
• Replace with 6 x 2TB HDD Capacity Disks
• Upgraded vSAN Datastore ~11TB

Upgrade Notes:

1. I choose to backup (via clone to offsite storage) and power off most of my VMs
2. I clicked on the Cluster > Configure > vSAN > Disk Management
3. I selected the one host I wanted to work with and then the Disk group I wanted to work with
4. I located one of the capacity disks (600GB) and clicked on it
5. I noted its NAA ID (will need later)
6. I then clicked on “Pre-check Data Migration” and choose ‘full data migration’
7. The test completed successfully
8. Back at the Disk Management screen I clicked on the HDD I am working with
9. Next I clicked on the ellipse dots and choose ‘remove’
10. A new window appeared and for vSAN Data Migration I choose ‘Full Data Migration’ then clicked remove
11. I monitored the progress in ‘Recent Tasks’
12. Depending on how much data needed to be migrated, and if there were other objects being resynced it could take a bit of time per drive.  For me this was ~30-90 mins per drive
13. Once the data migration was complete, I went to my host and found the WWN# of the physical disk that matched the NAA ID from Step 5
14. While the system was still running, removed disk from the chassis, and replaced it with the new 2TB HDD
15. Back at vCenter Server I clicked on the Host on the Cluster > Configure > Storage > Storage Devices
16. I made sure the new 2TB drive was present
17. I clicked on the 2TB drive, choose ‘erase partitions’ and choose OK
18. I clicked on the Cluster > Configure > vSAN > Disk Management > ‘Claim Unused Disks’
19. A new Window appeared and I choose ‘Capacity’ for the 2TB HDD, ‘Cache’ for the 200GB SDD drives, and choose OK
20. Recent Task showed the disk being added
21. When it was done I clicked on the newly added disk group and ensured it was in a health state
22. I repeated this process until all the new HDDs were added

Final Outcome:

• After upgrade the vSAN Storage is:
  • 2TB SAS Capacity HDD
  • 200GB SAS Cache SDD
  • 2 Disk Groups per host (1 x 200GB SSD + 1 x 2TB HDD)
  • IBM 5210 HBA Disk Controller
  • vSAN Datastore is ~11.7TB

Notes & other thoughts:

• I was able complete the upgrade in this order due to the nature my home lab components.  Mainly because I’m running a SAS Storage HBA that is just a JBOD controller supporting Hot-Pluggable drives.
• Make sure you run the data migration pre-checks and follow any advice it has.  This came in very handy.
• If you don’t have enough space to fully evacuate a capacity drive you will either have to add more storage or completely remove VM’s from the cluster.
• Checking Cluster>Monitor>vSAN>Resyncing Objects, gave me a good idea when I should start my next migration.  I look for it to be complete before I start. If you have an very active cluster this maybe harder to achieve.
• Checking the vSAN Cluster Health should be done, especially the Cluster > Monitor > Skyline Health > Data > vSAN Object Health, any issues in these areas should be looked into prior to migration
• Not always, but mostly, the disk NAA ID reported in vCenter Server/vSAN usually coincides with the WWN Number on the HDD
• By changing my HDDs from 600GB SAS 10K to 2TB SAS 7.2K there will be a performance hit. However, my lab needed more space and 10k-15K drives were just out of my budget.
• Can’t recommend this reference Link from VMware enough: Expanding and Managing a vSAN Cluster

Video Blog:

Various Photos:

If you like my ‘no-nonsense’ videos and blogs that get straight to the point… then post a comment or let me know… Else, I’ll start posting really boring content!