Hi,
I have a service which has previously been running without a problem.
I came back to it yesterday, and find out it’s not running anymore, and I cannot redeploy it.
I’ve tried to redeploy it with and without the build step. The build step runs without a hitch, but when it comes to the deployment there are no logs, and it just says starting.
Every now and again the deployment panel will print out:
Instance is stopping.
Instance stopped.
But that will disappear after some time, and then it will just be blank. And now it just seems to be stuck in “starting”.
Any help would be kindly accepted 
What instance type are you using? E.g. N300s, Free, Nano, etc.
Hi Bjarnos,
I was running small. Now I just tried redeploying again, and now it seems to work without a hitch, which is uncomfortably odd.
The Koyeb status panel displayed issues with deployment in Frankfurt earlier today, which is now resolved. Was your instance perhaps in the Frankfurt region?
Hey Bjarnos,
Yes, it was indeed! I saw the deployment issues, but there wasn’t anything specific reported in the uptime graph, only a “frankfurt issues” that wasn’t connected to the actual graph afai remember, when I had the problems, so it seemed odd.
But yes you’re right, that is the case, so might just have been me who read it wrong.
Thank you
Hello, i have same issues. And deployed in the Frankfurt region as well.
Yesterday the application was automatically redeployed (I have no idea why), the codebase did not change, the build, configuration, or env variables have remained unchanged for more than a year. However, the application still redeployed itself, and now it does not work. There are no logs available during the application launch, so I cannot identify the problem. I also do not have access to the Console when the instance does not start successfully. After 30 minutes, the instance terminates, then the logs appear, where it seems that the application is running fine. I thought it was a bug in Koyeb in healthchecks (so I temporarily added a container to heartbeat, and even now the healthchecks do not pass successfully). I have the entire build in docker-compose.yaml and I run it via Dockerfile.koyeb (on localhost the application works without a problem). The problem is probably in the Koyeb platform. Please urgently resolve my request, as this is a production incident with the highest priority and is blocking the client’s business. I tried to increase instance from Medium to Large, but it did not help.
Hey @Jan_Drnek ,
Sadly, without additional info like error logs, I cannot provide more support, as I’m not a Koyeb team member. I hope @Lukasz_Oles or @edouard can help you further.
Hi,
thank you for the reply.
I managed to resolve the production incident myself by migrating the application from the Frankfurt region to the Paris region. The application is now running correctly in Paris.
Based on this and the very similar symptoms reported in this thread, it strongly looks like the issue was related to the Frankfurt region / deployment runtime rather than the application itself.
Thanks.