In order to support the most recent versions of NodeJS, we need to migrate our web servers from Amazon Linux 2 to Amazon Linux 2023. Making this move precludes us from doing the usual spin-up/down instances, and involves a full migration.
Everything seems to be working smoothly, except for node-based services. What I did was copy .pm2/dump.pm2 to a new server and launch pm2 as a service. It initially looked fine, but it shows none of the services are taking any CPU or RAM, and apache proxy connections are failing to connect, so I think none of them are actually running. I tried pm2 resurrect as well, but the response is the same:
# pm2 resurrect
[PM2] Resurrecting
[PM2] Restoring processes located in /root/.pm2/dump.pm2
┌────┬────────────────────┬──────────┬──────┬───────────┬──────────┬──────────┐
│ id │ name │ mode │ ↺ │ status │ cpu │ memory │
├────┼────────────────────┼──────────┼──────┼───────────┼──────────┼──────────┤
│ 3 │ IDAP Dev │ fork │ 0 │ online │ 0% │ 0b │
│ 4 │ IDAP Website │ fork │ 0 │ online │ 0% │ 0b │
│ 2 │ Mirovoy NextJS │ fork │ 0 │ online │ 0% │ 0b │
│ 5 │ Reanthropize │ fork │ 0 │ online │ 0% │ 0b │
│ 6 │ Reanthropize Stage │ fork │ 0 │ online │ 0% │ 0b │
│ 1 │ lp-api │ fork │ 0 │ online │ 0% │ 0b │
│ 0 │ url-shortener │ fork │ 0 │ online │ 0% │ 0b │
└────┴────────────────────┴──────────┴──────┴───────────┴──────────┴──────────┘
Systemctl also seems to think it's working:
# systemctl status pm2-root
● pm2-root.service - PM2 process manager
Loaded: loaded (/etc/systemd/system/pm2-root.service; enabled; preset: dis>
Active: active (running) since Sun 2023-08-27 10:44:43 UTC; 1h 7min ago
Docs: https://pm2.keymetrics.io/
Process: 135008 ExecStart=/root/.nvm/versions/node/v18.17.1/lib/node_module>
Main PID: 135015 (PM2 v5.3.0: God)
Tasks: 11 (limit: 4568)
Memory: 20.3M
CPU: 925ms
CGroup: /system.slice/pm2-root.service
└─135015 "PM2 v5.3.0: God Daemon (/root/.pm2)"
Aug 27 10:44:43 ip-10-0-6-45.eu-central-1.compute.internal pm2[135008]: ├────┼─>
Aug 27 10:44:43 ip-10-0-6-45.eu-central-1.compute.internal pm2[135008]: │ 3 │ >
Aug 27 10:44:43 ip-10-0-6-45.eu-central-1.compute.internal pm2[135008]: │ 4 │ >
Aug 27 10:44:43 ip-10-0-6-45.eu-central-1.compute.internal pm2[135008]: │ 2 │ >
Aug 27 10:44:43 ip-10-0-6-45.eu-central-1.compute.internal pm2[135008]: │ 5 │ >
Aug 27 10:44:43 ip-10-0-6-45.eu-central-1.compute.internal pm2[135008]: │ 6 │ >
Aug 27 10:44:43 ip-10-0-6-45.eu-central-1.compute.internal pm2[135008]: │ 1 │ >
Aug 27 10:44:43 ip-10-0-6-45.eu-central-1.compute.internal pm2[135008]: │ 0 │ >
Aug 27 10:44:43 ip-10-0-6-45.eu-central-1.compute.internal pm2[135008]: └────┴─>
Aug 27 10:44:43 ip-10-0-6-45.eu-central-1.compute.internal systemd[1]: Started >
Assuming everything is where it's supposed to be and otherwise working, what's the right way to get pm2 to re-launch all of its services on the new machine? There are only 7 services, so I could, of course, just go start them all manually, but ... I would think there's some way to do this automatically.
