https://github.com/kubernetes/kubernetes/blob/master/docs/us...
328 karma · joined February 18, 2013
https://github.com/kubernetes/kubernetes/blob/master/docs/us...
This means that if you decide to move your application to a different underlying platform - your storage specific application and infrastructure code remains the same and all that changes is a Flocker configuration.
We have seen this benefit with a number of folks who need to support multiple infrastructures. Before Flocker - there were multiple code-paths depending on the underlying storage. Using Flocker - the same code-path was used.
Because it uses the vanilla docker client, it means that orchestration frameworks (mesos, k8s, fleet, swarm) will also be able to trigger the adapters.
Please note - powerstrip adapters are NOT designed to be used for production and are just a way to prototype docker extensions and experiment with composability of extensions in the Docker ecosystem :)
because powerstrip presents a standard docker HTTP api - it can certainly be used to interact with any other tool that talks standard docker HTTP.
For example - I put together a small run-through in powerstrip-weave where fig can be used to allocate weave IP addresses:
https://github.com/binocarlos/powerstrip-weave/tree/master/e...
Fig speaks to the powerstrip HTTP api which in turn speaks to the weave adapter.
It would definitely be interesting to have some kind of cadvisor adapter (or equivalent) that was automatically feeding back metrics to some kind of scheduler.
The main point is to allow the vanilla docker client to interact with powerstrip and therefore by extension - all of the existing orchestration tools.
Capturing JSON messages on their way to the docker server is a far more civilised way to extend docker than trying to "wrap" the docker cli.
In my use case - using the cli - I was struggling to separate the docker arguments from the image name and container args. It's better for docker itself to do that and more generally - its better to use the vanilla docker client (because orchestration tools).
As soon as I was working with the JSON messages generated by the docker client - everything became much easier!
Not sure I'm looking to start a new project BUT I am technical, in Bristol and would be happy to meet and say hello - email in profile
I suppose it depends on what you are trying to do - if it is looking good to those with fashion sense - better get down them shops. If it's to like, actually build stuff - the feeling of being uncool because of what language you use is a massive distraction to ignore.
I've been learning Go because its a great language for concurrent server systems (and better than node in some areas).
That I feel slightly cooler because of that choice feels a bit like having the right daps (sneakers) at school.
Ditching everything vs Burning out - classic balancing act of a startup innit
Running full pelt at a marathon only to hit 2 miles exhausted - time for tortoise to take over - he actually makes it!
Having said - I'm still trying to work out the combination of source repo and docker repo - i.e. 1 dev makes a code change - they commit the code to github - they then commit the docker image to the registry.
Now - does developer number 2 'git pull' or 'docker pull'
What I really like is how docker does not really enforce anything upon you - leaving the above question to be answered any which way.
Docker ftw!
I'm not flaming - I'm just overworked and glad that just HTML will do :)
Can someone explain why I might want to use Lua in a browser over JS? Does it have a fundamental feature that JS does not have?
Cool name btw!