Just came across this issue: https://youtu.be/USJXEjkQzds. Not sure what’s causing it but I think the service failed and render ends up in a loop attempting to constantly “reconnect” with it.
You are correct. This is caused by an error being thrown when your service starts up that prevent us from establishing a shell connection to your service. I would suggest taking a look at your service logs to see if there are any errors occurring there.
Thanks for confirming! Yeah it just seems like a bad UX, so I figured I’d report.
Agreed. It’s a known UX issue that we are tracking internally. Thanks for calling it out!
I’m still seeing this bug on services that are up and running fine: https://youtu.be/aP1H2-ZAE8s.
What could be going on in this case?
In addition: I suspected that the issues may be due to the fact I was using a docker env, and thought perhaps it was missing some SSH dependencies… I just switched the service over to the officially-supported rust env and I’m still seeing the exact same issue. So far, I have never been able to shell into any of my render services.
Hi Samuela, I took a look at your service, and looks like the service has failed due to an error. Check out the log tab to see what might have gone wrong. I believe your switch to Rust commit must have introduced the issue. Take a look at your events tab to see the timeline of your server failure.
I’m not seeing any failures or any issues in the Logs tab. It’s exactly as displayed in the video. The dashboard says that my service is up and healthy and it’s also responding to requests. What failure did you observe?
The commit switching to the rust env is not the culprit because I encountered the same bug with the docker env. In fact fixing this issue was my primary motivation for the migration to the rust env.
Let’s continue the conversation in direct messaging as I might need to ask you for your service ID.