/Cluster Update

Created Tue, 28 Mar 2023 11:21:05 +0100
255 Words

Cluster Update

My kubernetes cluster runs on a mix of RaspberryPI 4s, and x86(AMD64) machines, but because I built it originally with only the raspberrys, my control plane was on a PI

Well that was working fine for nearly 2 years, but the SD card on the control plane died, which wasn’t a good time

So I decided to what I always do, and go nuts

Spending spree

So I purchased a few things to tackle this issue

  • DL360g8
    • This is a server that is slightly overkill, it has 24 cores (2x Xeon E5-2420), and currently 32Gb of RAM (128Gb on order)
  • R715
    • If the last one was overkill, it has 32cores (2x Opteron 6200), and currently 32Gb of RAM (128Gb on order)
  • DL160g6
    • This server is mainly used as a database runner, it has 16 cores (2x Xeon L5630), and has 96Gb of RAM
  • EX2200-48p-4g
    • This is a fully managed POE switch, but I am considering changing the entire network to Ubiquity

Changes

With these purchases I decided to reconfigure the cluster, so rather than 4 worker PIs, 1 control PI, 2 x86 workers I now have the following

  • 5 Raspberry PI4s as workers
  • DL360g8, as control, and master (4Tb Longhorn)
  • R715, as worker and master (4Tb Longhorn)
  • DL160g6, as database runner
  • ML110g7, as a worker (3Tb Longhorn)
  • N40L, as a worker (2Tb Longhorn)

At the same time I switched to a FTTP provider, and now I have 1gb synchronous instead of 1gb/30mb

Hopefully this means I should have less issues, and less downtime

Series: kubernetes
Previous