Memory problems with django management command

[deanmalmgren]

Try profiling the django management command against a plain Python script to see if the latter is a lighter option

[deanmalmgren]

There really has to be a better way to handle this than having a bajillion cron jobs that each follow a train down the tracks. That just won't be scalable beyond one route. Using a queuing system like rq seems like a decent option, where each worker could just ping the API at a particular stop and then add another task to the queue at a particular 'time to live' (ttl) in the future. Celery is obviously an option here too.