I’m getting some strange behavior on a Mongrel deployment (my first,
please be patient). I’ve set up the mongrel_cluster.yaml ok and I can
start the cluster OK. But when I try to stop it using mongrel_rails
cluster::stop then I get this
!!! PID file tmp/pids/mongrel.8000.pid does not exist. Not running?
Even though the files are definitely there and have the correct pids
But the really strange thing is that the second attempt at stopping the
cluster will work just fine. Also if I do the sequence start, status,
stop then it works. So start, stop fails, but start, stop, stop works
and start, status, stop also works.
Something odd is going on inside mongrel_rails so that it isn’t getting
the path to the pid file correctly, and somehow a stop or status request
will kick things back into order. Which implies that these operations
alter the state of something. A request for status shouldn’t alter the
state of anything.
Any ideas what might be going on? I’m getting a bit weary of the setup
issues with mongrel and might switch to passenger.