I'm writing a console application in Python (using 'click' & 'invoke') that runs certain (long running..many minutes) tasks/test scripts and deployed on many platforms. Now there is a requirement that this application be accessible over network (i.e. from another web application running on a different server in the local network). So I reckon, the best way to achieve this is to convert the commandline app into a daemon process listening on a port for incoming requests. And rather than low level socket programming, turn it into RESTful API server (provide a 'route' for every sub-command and option)
But I'm somewhat new to python web programming and deployability is a major concern than development. This daemon needs to be deployed on various platforms within our setup that includes SunOS(5.11) and AIX(7.11) too (thankfully we recently retired a HP-UX server).
- So how is a python web application typically deployed? Is it OK(or possible) to deploy it without a separate http server like apache? If there is a trade-off between ease of develop/deployment vs performance, I am ready to choose ease of develop/deploy't because it is an internal application and number of requests daemon will need to handle are quite low (but the tasks it executes will be long running..many minutes).
- What are some of the options (framework/server) that would work on SunOS & AIX (Linux is no worry, but we do have a z/Linux too)? Point to note is that python versions available on Solaris & AIX are behind the current stable release available (but will be using python 3.x only)
cheers!
Aucun commentaire:
Enregistrer un commentaire