We are running multiple N8N instances in queue mode on GCP and are receiving the following error across our instances on http nodes (GET, POST, any type).
“Bad gateway - the service failed to handle your request
502 Server Error Error: Server Error The server encountered a temporary error and could not complete your request. Please try again in 30 seconds.”
When going into the logs we find the AxiosError: 502 is tied to this error.
Please share your workflow
Workflow independent. Also http request independent.
Information on your n8n setup
n8n version: varying from latest (1.62.5) to 1.48.3
Database (default: SQLite): postgres
n8n EXECUTIONS_PROCESS setting (default: own, main): main
It looks like this might not be an issue with n8n itself as the error you are getting contains HTML so it has connected to something and whatever that is it is unhappy.
Do you have access to the logs for the service you are connecting to that may have more information.
Hey Jon, thank you for being the ever-present, always good hearted community support member.
So, initially I was thinking the same, but since it’s happening across a variety of services in which these http requests are involved makes me think there may be something to the setup that’s bugging out.
Could it be in any shape or form related to certificates, ingresses, or the like since this started happening without any re-configurations or re-deployments. What may also be of interest is that these http requests return this statement all around or exactly on the 1-minute mark. When this starts occuring to a flow (it’s not system wide), only 1 in 10 make it through, and the issue sometimes completely passes away for a while.
Have you seen anyone else with this behavior? I’ve searched through the forum but haven’t found a specific solution that fits this issue.
Having it happen on different services is a bit odd, I wonder if it is a google gateway issue, Are you going direct out to the internet to the services or through some kind of google proxy service?
