You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
This is a bit down the road, like when we actually have something to host, but...
Requirements
Runs python (obviously)
VPS would be nice, eg if we want to run qcumber-frontend and qcumber-api (and qcumber-scraper?) all on one server.
Options
DigitalOcean? $5/mo
Linode? Chris has qcumber.ca on this now, I think? Maybe $5 or $10?
Rackspace? $15
Heroku? Free, until we need more (likely ever?), then $35/mo...
Raspberry Pi on queen's campus somewhere? (free haha)
I'm down to chip in or cover a DigitalOcean one. I've been using them a bunch over the last months, seem pretty solid.
I lean toward that over something like Heroku because I think we've got enough dev/ops power on this team that the control would be nice. Thinking long-term though, maybe keeping up qcumber in the years to come will be more approachable if it's just a heroku dyno? I'm sure qcumber gets enough traffic that dynos idling down wouldn't be an issue..
The text was updated successfully, but these errors were encountered:
I like the idea of putting it on a heroku dyno because it would make it really easy to set up, however we would probably not be able to put the api and frontend on the same server, meaning that it may be a bit more pricey than the other options.
The other problem comes with database choice. Heroku comes with (IIRC) a small free postgres database, but has no free/cheap option for a mongo database without going to a cloud mongo provider, which will probably be another $15-20/month.
DO sounds like a good option as long as we keep puppet-like scripts set up to configure it and document how to maintain it.
This is a bit down the road, like when we actually have something to host, but...
Requirements
Options
I'm down to chip in or cover a DigitalOcean one. I've been using them a bunch over the last months, seem pretty solid.
I lean toward that over something like Heroku because I think we've got enough dev/ops power on this team that the control would be nice. Thinking long-term though, maybe keeping up qcumber in the years to come will be more approachable if it's just a heroku dyno? I'm sure qcumber gets enough traffic that dynos idling down wouldn't be an issue..
The text was updated successfully, but these errors were encountered: