Connection lost on n8n
kriss2222
PROOP

20 days ago

I can't edit my n8n workflows because it keeps switching from "Connection lost" to "Published". It goes back and forth every few seconds.

In the Railway Network Flow Logs, I see this repeating itself "tcp 66.33.22.230:26387 → 10.250.15.229:38066 TCP_INVALID_SYN"

I also updated my n8n today in Railway by redeploying it after receiving the email "Action required: update to the latest patch version of n8n." I'm not sure if that caused the problem.

Attachments

$20 Bounty

11 Replies

Status changed to Open Railway 20 days ago


Are there any errors in the deployment logs of any of your n8n services?


kriss2222
PROOP

20 days ago

Oh ya, yes.

This is in the primary n8n Deploy log "(node:1) [DEP0169] DeprecationWarning: url.parse() behavior is not standardized and prone to errors that have security implications. Use the WHATWG URL API instead. CVEs are not issued for url.parse() vulnerabilities."

The attached i downloaded from the Postgres Deploy log.

Attachments


Click on your Postgres service, go to Database, and run the following queries:

REINDEX DATABASE railway;
ALTER DATABASE railway REFRESH COLLATION VERSION;

Redeploy n8n and Postgres afterwards.


0x5b62656e5d

Click on your Postgres service, go to Database, and run the following queries:REINDEX DATABASE railway; ALTER DATABASE railway REFRESH COLLATION VERSION;Redeploy n8n and Postgres afterwards.

kriss2222
PROOP

20 days ago

I don't have the option of data or a database?

Attachments


kriss2222

I don't have the option of data or a database?

You'd need to install a database client (eg, Datagrip), connect to your Postgres service through the public URL, and run the queries through there.


kriss2222
PROOP

20 days ago

I installed TablePlus and did that, no luck.
I attached the new log. logs.1776909648006 is the Postgres deployment logs and logs.1776909712056 the N8N deployment logs


Try adding a variable N8N_PROXY_HOPS=1 in your n8n service and redeploy after.


0x5b62656e5d

Try adding a variable N8N_PROXY_HOPS=1 in your n8n service and redeploy after.

kriss2222
PROOP

20 days ago

still no luck. i attached the new deploy logs for the n8n service.

Attachments


Try adding N8N_PUSH_BACKEND=sse.


0x5b62656e5d

Try adding N8N_PUSH_BACKEND=sse.

kriss2222
PROOP

20 days ago

The SSE change made it worse, now every time i try and run any node in n8n it says "Problem running workflow

Lost connection to the server."

I'm going to delete N8N_PUSH_BACKEND entirely


kriss2222
PROOP

19 days ago

Any other ideas? Also thanks for your help! I appreciate it.


Welcome!

Sign in to your Railway account to join the conversation.

Loading...