Installation Error

[ddlidded]

The message bad gateway is appearing when going to my server-pointed domain name even though the setup in the terminal yielded a success message.

Ubuntu 20.04.5 LTS

[shihjay2]

It may help if you could post the results of these commands in your server.

docker logs $(docker ps -aqf "name=^nosh$") -f

docker logs $(docker ps -aqf "name=^router$") -f

docker logs $(docker ps -aqf "name=^couchdb$") -f

[hammercuba]

Hi, same issue here, the log:
root@srv1:~/nosh3# docker logs $(docker ps -aqf "name=^router$") -f
time="2024-01-17T00:03:16Z" level=info msg="Configuration loaded from flags."
time="2024-01-17T00:06:25Z" level=error msg="service "nosh" error: unable to find the IP address for the container "/nosh": the server is ignored" container=nosh-nosh3-08c9a9ec3f585a28b10690bafbcf85a1e6e5a31c37c7c538e853e41108c7014a providerName=docker

[shihjay2]

@hammercuba can you post the log for the nosh container?

docker logs $(docker ps -aqf "name=^nosh$") -f

The nosh container is likely not loading hence the absence of an IP address...

[hammercuba]

root@srv1:~/nosh3# docker logs $(docker ps -aqf "name=^nosh$") -f
listening on 4000
GET /start - - - - ms
GET / 404 139 - 11.190 ms
GET / 404 139 - 3.239 ms

/usr/src/app/node_modules/node-fetch/lib/index.js:1491
reject(new FetchError(request to ${request.url} failed, reason: ${err.message}, 'system', err));
^
FetchError: request to https://db.#######.duckdns.org/_users/ failed, reason: connect ETIMEDOUT ...:443
at ClientRequest. (/usr/src/app/node_modules/node-fetch/lib/index.js:1491:11)
at ClientRequest.emit (node:events:513:28)
at TLSSocket.socketErrorListener (node:_http_client:495:9)
at TLSSocket.emit (node:events:513:28)
at emitErrorNT (node:internal/streams/destroy:151:8)
at emitErrorCloseNT (node:internal/streams/destroy:116:3)
at process.processTicksAndRejections (node:internal/process/task_queues:82:21) {
type: 'system',
errno: 'ETIMEDOUT',
code: 'ETIMEDOUT'
}

Node.js v19.9.0
listening on 4000
GET /start - - - - ms

/usr/src/app/node_modules/node-fetch/lib/index.js:1491
reject(new FetchError(request to ${request.url} failed, reason: ${err.message}, 'system', err));
^
FetchError: request to https://db.####.duckdns.org/_users/ failed, reason: connect ETIMEDOUT ..*.:443
at ClientRequest. (/usr/src/app/node_modules/node-fetch/lib/index.js:1491:11)
at ClientRequest.emit (node:events:513:28)
at TLSSocket.socketErrorListener (node:_http_client:495:9)
at TLSSocket.emit (node:events:513:28)
at emitErrorNT (node:internal/streams/destroy:151:8)
at emitErrorCloseNT (node:internal/streams/destroy:116:3)
at process.processTicksAndRejections (node:internal/process/task_queues:82:21) {
type: 'system',
errno: 'ETIMEDOUT',
code: 'ETIMEDOUT'
}

Node.js v19.9.0
listening on 4000
GET /robots.txt 404 149 - 22.641 ms
GET / 404 139 - 7.010 ms
GET / 404 139 - 2.052 ms
GET / 404 139 - 1.972 ms
GET /favicon.ico 404 150 - 2.096 ms

[hammercuba]

the host is a Debian 12 just with docker, under a at&t fiber, with the port redirection 80 and 443, and duckdns.org configured

[shihjay2]

What is the COUCHDB_URL value in the .env file? I'm assuming it's db.####.duckdns.org based on your logs. Can you confirm that db.####.duckdns.org is reachable (ie CNAME is registered for that subdomain).

[hammercuba]

Yes, is reachable

[hammercuba]

this is from a linode server

root@srv8:~# nslookup db.####.duckdns.org
Server: 127.0.0.53
Address: 127.0.0.53#53

Non-authoritative answer:
Name: db.####.duckdns.org
Address: ##.##.##.##

root@srv8:~# telnet db.#####.duckdns.org 443
Trying ##.##.##.##...
Connected to db.####.duckdns.org.
Escape character is '^]'.
get
HTTP/1.1 400 Bad Request
Content-Type: text/plain; charset=utf-8
Connection: close

400 Bad RequestConnection closed by foreign host.
root@srv8:~#

[shihjay2]

Can you check if https://db.####.duckdns.org/_utils/#login is accessible via browser. If so, I have made some updates in the commit cae56c3

[hammercuba]

[hammercuba]

I did a fresh reinstall of everything and it's doing the same, what can I do?

[shihjay2]

@hammercuba can you post the log for the nosh container again as the recent updates should not throw the same type of errors previously shared.

[hammercuba]

I do a docker-compose down -v, then erase all the nosh3 folder and proceed to run the script again, same logs.

[hammercuba]

/usr/src/app/node_modules/node-fetch/lib/index.js:1491
reject(new FetchError(request to ${request.url} failed, reason: ${err.message}, 'system', err));
^
FetchError: request to https://db.******.duckdns.org/keys/ failed, reason: connect ETIMEDOUT *******:443
at ClientRequest. (/usr/src/app/node_modules/node-fetch/lib/index.js:1491:11)
at ClientRequest.emit (node:events:513:28)
at TLSSocket.socketErrorListener (node:_http_client:495:9)
at TLSSocket.emit (node:events:513:28)
at emitErrorNT (node:internal/streams/destroy:151:8)
at emitErrorCloseNT (node:internal/streams/destroy:116:3)
at process.processTicksAndRejections (node:internal/process/task_queues:82:21) {
type: 'system',
errno: 'ETIMEDOUT',
code: 'ETIMEDOUT'
}

[shihjay2]

Looks like a different error - if you used the docker-compose install method in the readme, it should automatically update the docker image (built a few minutes ago) based on the latest commit 97e0368