Inherited Server from Bankrupt Startup So the short story goes is our startup went under. The CEO then decided to split up the assets and I received a server stack. What are some interesting/ useful learning scenarios I could perform. I have minimal experience using linux. |
In terms of learning things, if by a "stack" you have 10 or more, then you run a hadoop cluster or any of a number of other clustered type systems. If you were more devops oriented you could play around with deployment scripts and containers and what not. Setting up bitcoin miners or unikernel systems. Can be an excellent way to transcode video if you've got lots of VHS tapes you wish were MP4's :-)
Really depends on what you want to learn, bare metal does have it's advantages when it comes to learning how stuff goes together. Especially if you also got an ethernet switch out of it.
Plus depending on the number of racks you have running it will cost more in electricity each month than the cost of a few VPS accounts.
Install a hyper-visor and start playing with VMs, containers etc.
how old are these systems? servers are only usefull/have significant monetary value for other companies if age is << 5 años.
if they are not too old: sell them and play with an older standard pc - 64 bit / virtualization possible, enough ram.
cheers az
If it's system administration, one example that is reasonably useful is setting up an owncloud server. Or a private mail server. If it is in your home, setting up a DHCP server and a DNS server can be a fun excercise. In any case, there are many tutorials and HOWTOs available online. The search engine of your choice will tell you where. Most major linux distros also come with pretty good documentation or at least have some online. (Depending on the circumstances, you might consider reinstalling the system first - without physical access to the machine, it is not easy, though.)
If you want to learn about software development, there are more things you can do with a linux server than I can think of. When I try to think of a fun programming project, I often make the mistake of thinking of the technology (programming language, libraries/frameworks) first and then choosing a problem that goes with it. For learning, though, it can be a good idea - so I want to learn about, say, building web apps with Ruby on Rails or Django or whatever; what better way to learn about that than building a small toy application? Okay, now all I need is a nice toy problem, difficult enough to actually teach me something, but not difficult enough to be frustrating.
Again, there far too many tutorials out there for me to mention anything specific. (Just as an example, I taught myself Django by writing an application that is basically a glorified RSS reader with a builtin Bayes network - I could rate news as boring or interesting, and after a while the app could classify news items a "probably boring" or "probably interesting" well enough so I could make it filter out news that were "probably boring". It did not work that well, was really slow, and I basically suck at UI design, so it was really ugly, too, but it was still useful and fun to build.)
I hope this helps at least a little. I am staying fairly vague because your question is kind of vague. If you supplied more details about your interest or your skills/background, I might be able to give a more helpful answer. Generally speaking, try to think of something that would be actually useful to you - that way, motivation will be less of a problem, and it will probably be more fun, as well. (To be honest, learning about Linux / Unix can be very frustrating initially - within the first six months of using Linux on my desktop, I was very close to physically throwing my computer out of the window no less than three times. The learning curve can be steep at the beginning, but to me, in retrospect, it was time well spent. It also tends to be a great deal of fun after a while.)