Extracting Hadoop tgz is slow as hell and causes high cpu utilization

[soenkeliebau]

We noticed this when installing Hadoop version 3.2.2. The agent took a couple of minutes to extract the archive in to the packages folder, whereas extracting it directly with tar finished much quicker. I started the untar in my terminal after the agent was going for a while on the same machine and it still finished before the agent was done. This would indicate that it is not an issue with the archive itself or hardware limits.

We should probably have an in depth look at how we extract archives and try to optimize this.

[soenkeliebau]

Blocked by packaging discussion