Digital Preservation - Server Side

Question

Is there any guidance/best practice on server side archiving of a webserver ? I can think of things such as documenting hardware setup/partitions/process running etc in text files, using Bag-It for important directories (/etc/, /usr/local/, /home, software in /src, etc), but this seems slightly haphazard.

I should say the disk image itself *may* be preserved but I would like to ensure the non-OS parts are preserved in a more accessible form.

And as always, time is short and the budget is small. I'd like to just do enough that whoever may come along in 10 years time doesn't curse me for not saving the one essential thing they need.

Answer 1

Hi Richard,

I am working on the same issue as part of my research at the Netherlands Institute for Sound and Vision. We have started investigating best practices but so far we have not come across a lot of concrete examples. I am mostly drawing inspiration from art institutions that have more experience in this area. We are trying to acquire some server-side materials from a couple of websites to test how it could be done.

I would love to discuss this with you further. Pease contact me at rbocyte@beeldengeluid.nl.

Rasa

Answer 2

Sometime ago I migrated a physical server machine to emulated hardware. I added some documentation of the process on an OPF blog post here. With Emulation as a Service (EaaS) now very useable you may want to consider following a similar approach. The research team at the University of Freiburg behind EaaS have also been pursing improved features for preserving acess to linked servers and front-ends using the same backend emulation infrastructure and may have some additional ideas that could help with your use case.