First of all the latest release which was using Docker on public clouds (AWS, GCP and Azure) was 1.2.3. The 1.3.0 or newer versions are not using Docker to run the Hadoop services. Anyway for 1.2.3 the answers are: 1. Containers were started with net=host, thus there was one container per VM - Docker was mostly used for packaging and distribution - thus every node had one container. You needed as many nodes as the size of the cluster was. 2. You can but the container was getting the full VM resources (see #1) 3. You need to install the Cloudbreak application (anywhere, that can be an EC2 instance for example but on-prem as well). The Cloudbreak application - note it's not the cluster - is composed of several micro-services, and these are running inside containers. Can be GUI or CLI or API - every hostgroup can have different instance types, the cluster can be heteregenous. 4. http://sequenceiq.com/cloudbreak-docs/ 5. It depends on the number of nodes you'd like to provision. There are no additional costs on top of EC price thus yo ucan do a fairly easy math - multiply the number of nodes you think your cluster will have with the number of hours ... In Cloudbreak you can fully track usage costs on the Accounts tab. ... View more

Salut Constantin Currently this is tech preview due to some limitations - there are some changes undergoing within Mesos (unified containerizes, IP per container, reverse lookups, storage drivers, etc to name a few) - so expect a large change here. Nevertheless if someone is happy with the current limitations (namely running with net=host) the code is stable and works well (we and a few others have test/dev clusters). Should you want more details drop me a mail. Janos ... View more

Hi guys. This is due to an old SQL driver - as stated above you can either change the driver or wait until the next release where the driver will be updated. Sorry for the inconvenience. ... View more

The container should not be updated - you mean to say the docker version? Contiamners can be updated with docker pull but you should not be concerned as an end user or operator about this. Did you face some issues or is this an observation? ... View more

You should have these ports open (as per the documentation) - also what's in your Profile (PUBLIC_IP/PRIVATE_ID) SSH (22) Cloudbreak API (8080) Identity server (8089) Cloudbreak GUI (3000) User authentication (3001) ... View more

What do you mean local machine - physical clusters? If yes, you can't - for that you can use Ambari. Cloudbreak is for provisioning clusters in the cloud. If you use OpenStack for virtualization within your datacenter/physical boxes than you can use Cloudbreak. ... View more

Which version of CBD are you using (cbd version)? There was an issue in 1.2.2 (due to an updated systemd version) which has been fixed - download the latest 1.2.3 binary or do a cbd update. After that you should do a cbd kill && cbd regenerate && cbd start ... View more

You can give it a try now on the hosted version or do these changes in your cbd Profile: export DOCKER_TAG_CLOUDBREAK=1.2.6-rc.3 And then restart cbd with: cbd kill && cbd regenerate && cbd start ... View more

Currently it's not supported, though with the BYOS provider you might be able do that (using the Swarm orchestrator). ... View more

You don't have to disable anything. The containers are launchd with net=host which means they inherit the host's network. Azure does not support IPv6, thus there are no routable IPv6 addresses for the container. Beside Azure does not support IPv6 (this this can't be an issue) by default Docker is configured with IPv4 only. This is definitely has nothing to do with IPv6 ... ... View more