Open Source Stack Transparency and Access to Build Scripts

[simonj217]

hi @lucasbak

Since you have stated on your website that your entire stack is open source, I have noticed that, unlike the Apache Bigtop project, you have not released any build and stack creation scripts so far.

I need to build all the sources myself, from Apache Ambari itself to the related services.

However, in your GitHub repositories, there are no build scripts for this stack like there are in Apache Bigtop. You only provide binary versions for users to use. This forces users to rely solely on your binaries. Isn’t this method of access—especially with tools that require registration—a policy similar to Cloudera, which eventually even restricted access to binary stacks?

If your policy is to provide an open source stack, I kindly request that you publicly release all scripts for building and creating the stack.

I would appreciate it if you could clearly state your company’s policy, as this will help all users make informed decisions about using your stack.

Thank you.

[lucasbak]

Hi @simonj217,

Thank you for your issue.

We do publish all sources that correspond to binary releases and do port back issues when needed.

The goal of Clemlab as an Open Source company is to provide the binaries freely forever with nopaywall with corresponding sources like https://github.com/clemlabprojects/hadoop-release-odp/tags nor https://github.com/clemlabprojects/tez-release-odp/tags and others.

Our website documentation still misses some pieces as we focus on adding functionalities, creating the next releases and provide overall stability.

We did notice that you opened issue on Apache Ambari project too. What do you try to achieve ? we may be able to help you.

Depending on the use case we do provide custom documentation through enterprise support.

We provide best effort on the github projects, there may be some delay.
That is why it important to close the issues when you are not active anymore.

Understand that as an Open Source company, we live from support only.

Kind Regards,

Clemlab Team

[simonj217]

hi @lucasbak
I have reviewed all these sources in your GitHub. However, the build script for this stack is not available like the Bigtop project, so that we can build individual services such as Hadoop, Hive, and other services, and then use Ambari, which has also been built from source, to install the cluster. The events and policies that companies like Cloudera have adopted in recent years have led companies to utilize such Hadoop ecosystem binaries this way.

[lucasbak]

Hi @simonj217,

You can build easily each project using maven like what is documented on per each component documentation.

No worries projects sources will stay open source and binaries available without paywall, this is our commitment ;).

Clemlab has been created to provide a ready-to-use and install Hadoop Open Source Stack alternative to other closed ecosystem like Cloudera to make live Hadoop cluster and provide continuity to existing managed Ambari cluster.

NB: Takeover from HDP, CDP, and BIGTOP is also available.

Kind regards,

Clemlab Team