pomadchin
commented
8 years ago Hi, Steve! We actually have description in common README's:
Differences in prod and dev that dev includes Hadoop 2.7.1, ZooKeeper 3.4.6, Accumulo 1.7.x (the specific version to use is configurable), Spark 1.5.2 (Scala 2.10 / Scala 2.11) in one container and that's only can be in use for development (you can rise cluster in a distributed fasion on one machine). Prod is the absolutely the same, but with only one process per container (so that's how it should look like in production) and with cassandra run scripts (actually we use official cassandra images there).
So I am curious where did you find Kafka? It is not hard to add necessary images with necessary tools, if that's important for you probably we can add it.
It is possible to start prod with Kubernetes, but you have to solve network problems yourself, as you can see kubernetes instructions are not provided there (probably yet). If you have docker installed on any machine, definitely you can start any docker container (accroding to your cpu architecture) on that machine. Probably we can try to solve your problems, I am curious, can't we just setup docker on openshift and to start containers? And yes, very interested in Kubernetes solution. :)
Hey there: Don't see a link to a site or doc for this project. I have some questions and not sure if there is doc on it.
It looks like your docker image puts Hadoop, Accumulo and Kafka all in the same container - did I read that right?
Any chance you would like to work with me (us/Red Hat Openshift) on putting this into Kubernetes/OpenShift format? I would love to write a demo and application using all these pieces.
Thanks Steve