Closed KristerV closed 6 years ago
Two questions need to be figured out:
I don't think there's a need for having dev/staging separately for this project. But dev and prod does make sense. Still how about we just start with production, get it figured out and continue from there, since looks like Helium is being released in 9 days.
On DigitalOcean I basically wrote some systemctl scripts to keep it alive seen here. It worked great since it's immune to restarts, but I just had way too much deployment logic in there, which should be a separate bash script.
Anyway @jschr can you please specify how you would deploy the server, as you seem to have more experience with Amazons scaling services.
jschr said he was going to use Amazon's Beanstalk, which doesn't support having heliumd anywhere, so we are without a solution.
Talked to chaositec about this and he offered to create a middle layer of PHP that would route requests between multiple wallets (heliumd).
Pros
Cons
He did offer to build the thing and have us deploy, but since we've already built the backend it doesn't seem reasonable. We can just implement caching and manual-scaling ourselves and keep the control of the system. While also being able to autoscale later by using DigitalOcean API.
Clearly going the self-deploy with docker route at this point :)
ok let's leave the server for now.
heliumd
andhelium-cli
to talk to eachother.