Our recent upgrades of Ruby put the pdc-discovery VMs in a weird state. See #636 for an example. Rebuild all VMs from scratch to ensure we are using the same versions of everything on all machines.
Acceptance criteria
[ ] staging machines replaced
[ ] production machines replaced
Implementation notes, if any
Build and configure new production VMs and switch the nginx configuration to them before decommissioning the old ones.
What maintenance needs to be done?
Rebuild the pdc-discovery-* VMs.
Level of urgency
Why is this maintenance needed?
Our recent upgrades of Ruby put the pdc-discovery VMs in a weird state. See #636 for an example. Rebuild all VMs from scratch to ensure we are using the same versions of everything on all machines.
Acceptance criteria
Implementation notes, if any
Build and configure new production VMs and switch the nginx configuration to them before decommissioning the old ones.