deployster
deployster copied to clipboard
bmorton
**Abandoned** A pretty clear winner is emerging in this field and Fleet is now dead -- use Kubernetes instead! ~~An opinionated Golang HTTP service for automating zero downtime deploys to a Fleet clu...
Deployster itself is pretty simple and scoped, but being able to take action after Deployster has done certain things would be pretty awesome. Some examples of things that Deployster likely...
After thinking about the higher level use-cases we have, and reading about some of the other solutions in this space, something I found in Mesosphere's Marathon is the concept of...
Currently, we assume that vulcand is installed and use that to accomplish zero downtime deploys. We should allow different solutions here, including things like Nginx or HAProxy. Both alternatives would...
CoreOS added a software network fabric for containers called `flannel` in version 554. Basically, it gives every Docker container an IP that is accessible from anywhere in the cluster. For...
If a deploy is already in flight for a given service, we should prevent subsequent deploys if that service. We could optionally allow a `force` parameter to cancel and cleanup...
Currently, you can pass any version, valid or not, to the `POST` deploys endpoint and we'll attempt to deploy that. Should we do validation first so we can fail earlier?...
Currently you need to know which host `deployster` was launched on in order to send command to it. In order to make less assumptions, we should automatically register it with...
Is it reasonable for deployster to implement functionality to ease in setting/managing environment variables that are exposed to services by convention? **Update 3/12**: In deploying a number of services with...
The asynchronous mode of deploys is good for many use-cases and makes things quite a bit simpler, but it leaves the person deploying a little unsure about what is going...