Tests failing because of slow start of the application

[micaelmalta]

Some webservers like nginx-unit has a very slow start process which cause tests failure

Would it be possible to include a healthcheck with start_period like it s done in docker-compose?

thanks

[NateBrady23]

We ping the application every second for 1 minute. We don't have the capacity to add a feature just for this. We could increase the ping time a little but that would increase a full run of the test suite by (increase * number of tests that don't start properly). How long does it take nginx-unit to start?

[micaelmalta]

Unfortunately, Nginx-Unit has a 2 steps start and It depends on the number of processes.

For what I experienced, it takes about 2s per process per step

So if we want to use the full capacity of the server, we could have for ex, 64 processes which can take a long long time to start

Also, there is a restart during the 2 steps which can cause issue with the way the health check is done

cf: nginx/unit#832