Node.js App starting but not responding?

1

I have an app running and I’m making significant changes to it so I created a new app and I’m trying to get that up and running. Everything runs fine locally it’s just when I deploy to fly I’m getting an ERROR 502 (after waiting quite some time).

The app is logging healthcheck requests, though those don’t seem to be working right either. And if I navigate to the healthcheck endpoint I get a 502 there as well.

Here’s the log output on startup:

2022-10-21T23:12:50Z app[0450c2b2] den [info]Starting init (commit: 249766e)...
2022-10-21T23:12:50Z app[0450c2b2] den [info]Mounting /dev/vdc at /data w/ uid: 0, gid: 0 and chmod 0755
2022-10-21T23:12:50Z app[0450c2b2] den [info]Preparing to run: `sh start.sh` as root
2022-10-21T23:12:50Z app[0450c2b2] den [info]2022/10/21 23:12:50 listening on [fdaa:0:23df:a7b:d827:2:6f19:2]:22 (DNS: [fdaa::3]:53)
2022-10-21T23:12:50Z app[0450c2b2] den [info]+ npx prisma migrate deploy
2022-10-21T23:12:51Z app[0450c2b2] den [info]Prisma schema loaded from prisma/schema.prisma
2022-10-21T23:12:51Z app[0450c2b2] den [info]Datasource "db": SQLite database "sqlite.db" at "file:/data/sqlite.db"
2022-10-21T23:12:51Z app[0450c2b2] den [info]2 migrations found in prisma/migrations
2022-10-21T23:12:51Z app[0450c2b2] den [info]No pending migrations to apply.
2022-10-21T23:12:52Z app[0450c2b2] den [info]+ npm run start
2022-10-21T23:12:52Z app[0450c2b2] den [info]> kentcdodds.com@1.0.0 start
2022-10-21T23:12:52Z app[0450c2b2] den [info]> cross-env NODE_ENV=production node --require ./node_modules/dotenv/config ./index.js
2022-10-21T23:12:55Z app[0450c2b2] den [info]Express server listening on port 8080
2022-10-21T23:12:55Z app[0450c2b2] den [info]prisma:info Starting a sqlite pool with 3 connections.
2022-10-21T23:13:04Z app[0450c2b2] den [info]prisma:query - 52ms - SELECT COUNT(*) FROM (SELECT `main`.`PostRead`.`id` FROM `main`.`PostRead` WHERE 1=1 LIMIT ? OFFSET ?) AS `sub`
2022-10-21T23:13:09Z app[0450c2b2] den [info]GET /healthcheck - - - - ms
2022-10-21T23:13:24Z app[0450c2b2] den [info]GET /healthcheck - - - - ms
2022-10-21T23:13:39Z app[0450c2b2] den [info]GET /healthcheck - - - - ms
2022-10-21T23:13:54Z app[0450c2b2] den [info]GET /healthcheck - - - - ms
2022-10-21T23:14:09Z app[0450c2b2] den [info]GET /healthcheck - - - - ms
2022-10-21T23:14:24Z app[0450c2b2] den [info]GET /healthcheck - - - - ms
2022-10-21T23:14:39Z app[0450c2b2] den [info]GET /healthcheck - - - - ms
... continues

Any ideas?

For anyone at Fly, the app is called kcd-staging and the code is in the dev branch on GitHub - kentcdodds/kentcdodds.com: My personal website

2

Hey - as a first step, I’d try to SSH in via fly ssh console and see if anythings running vis something like these commands (if they are available):

# see running processes
ps aux

# check requests, i’m assuming the port here 
curl localhost:8080
3

Thanks. Those commands are not available unfortunately. But I know that the server is running and receiving requests because of the logs. It’s just struggling to respond for some reason. I’ll keep digging.

4

If your image is Debian based, you can install both with:

apt-get update
apt-get install procps curl
5

Ok, so my home page does a ton of work which is cached after the work is done, but it never gets a chance to get cached before the request times out. So I’m working out a way to work around that.

1 Like
6

I figured this one out. It’s a long story. It was my own fault. Feel free to watch this if you’re interested: EpicWeb.dev Live stream 🌌 Finish deploying LiteFS on Fly - YouTube

1 Like