Update / reorganize resources as imagecat moves to the nomad infrastructure

[acozine]

What maintenance needs to be done?

Level of urgency

• High
• Moderate
• Low

Why is this maintenance needed?

DLS has moved the imagecat project over to the nomad infrastructure.

Acceptance criteria

• Decommission imagecat-prod1, imagecat-staging1, and imagecat-libserv37, checking the CPU and memory usage on all
• Reapportion the resources from the decommissioned VMs to the nomad infrastructure
• Increase the RAM on nomad-client-prod1 from 8GB to 16GB
• Increase the RAM on nomad-client-prod2 from 8GB to 16GB
• Create nomad-client-prod3 with 16GB of RAM and 8 cores
• Create nomad-client-prod4 with 16GB of RAM and 8 cores
• Increase the RAM on nomad-client-staging1 to 16 GB
• Create nomad-client-staging2 with 16GB of RAM and 8 cores
• Create nomad-client-staging3 with 16GB of RAM and 8 cores
• Increase the RAM on dpul-collections-staging1 to 16GB (this is the box that runs the logging infrastructure)

Implementation notes, if any

Double-check both the total amount of resources and the distribution across data centers and across VMware hosts, and note the overall delta here in the ticket.

[acozine]

Decommissioned resources:
imagecat-libserv37 had 1 CPU and 4 GB memory on host 3b and 200 GB disk space on 14_05TB disk
imagecat-prod1 had 8 CPUs and 64 GB memory on host 2b with 30 GB disk space on 12_05TB disk

[acozine]

Created nomad-client-prod3/4 today, both are on the .lib domain and on the private network. I can only SSH to one of them, but I'm hoping that's just a patience issue. Also opened pulibrary/princeton_ansible#5663 for the inventory side of things.