This question already has an answer here:
I have somewhat of a general question, and I hope this is the right place to ask.
We currently have 11 (9 Red Hat/2 Centos) servers, which have been purchased at different times over the last 5 years, and although they all have Linux OS, there are hardware differences between them, simply because they were bought at different times. However, we need a core set of bioinformatics software installed on all of the machines to allow users to have the same environment. In the past we have used a shared mounted drive, but we have run into problems when software does not work on a given machine because not all necessary libraries were installed, etc.
Long story short, I’m looking for tips and advice as to the best approach to this problem. Do you think working on a shared drive and troubleshooting issue as they occur is the best path, or is there an improved method I’m not privy to (software/ theoretical). Ultimately I feel that trying to maintain each machine independently has to be the most painful method.