Backup Strategies and Best Practices

From DevSummit
Jump to navigation Jump to search
The printable version is no longer supported and may have rendering errors. Please update your browser bookmarks and please use the default browser print function instead.

Back up Strategies & Best Practices

Identifying a reason why you or your organization wants to backup and also make sure you're not giving away your data is important – a lot of security motivation in NSA news recently

  • many groups are not motivated to think about back-ups or security in general
  • would rather have someone do it for them
  • often motivated by a known threat

First step is to figure out What Data you have & where it lives

  • create an inventory of where you have data and where it is stored
  • something like an Online Accounts Inventory

For Data that you want to keep – what options seem reasonable?

What are safe strategies – different kinds of data have different characteristics

different characteristics of the data means that you may think about how it is backed up in different ways

  • if data set only grows
    • raw video archive
  • if data set changes
    • a database of contacts
    • snapshots of the changes


  • video
    • multiple pieces of physical media
    • different physical locations
    • hard drives (cheap drives)
    • Rsync – command line tool to backup changes
  • need to establish process around consistently backing-up
  • make it easy for the people to back-up, like a button
  • online database

Questions to ask:

  • Is the data that you want to back up all in 1 places?
  • How does the backup work?

Complete the Back-up Process - can you restore from the back up - can you test that the back-up in functional - if you use external hard drives, how to

S.M.A.R.T. Mon (monitor) Tools - smartmontools - can ask the hard drive to tell you things like temperature, when the last read error was, let me know when the disk is going to fail

  • git (revision control)

Databases

  • 2 common ones
    • mySQL
    • PostGres – postgresql.org

Incremental Backups

  • Rdiff
  • Differential
  • it is really important that you understand how the backup that you are running is working, is it sync, mirrored, iterative, just tracking changes... etc

Restoring

  • database backups
    • have a machine not on internet
    • recreate database & then push the backup to it and see if it works
    • you could also run tests to see if it works

Tools that do autobackup testing -

  • test your restores
  • you tell them the kind of thing that you think you are backing up

Everything time you bring in someone new, make sure they can restore from a back – test that someone could do it (tests if it is possible & likely)

For developers – understanding the functional dependencies that the code relys on

Backing up Social Media – what is necessary to backup?