Synology

Scripts related to synology backup (rsync & hubic) & data corruption detection

Script to backup a Synology NAS to hubic :

https://github.com/jbfuzier/Synology/tree/master/Hubic

Script to check data integrity between a master node and it's backup

• Compute hash of every file, keep it in a "db" (simple json file)
• Sync the json file between the nodes
• Both node will check file integrity based on this json file (in my use case = archiving, files should not change; if they do it is a sign of silent corruption)
• Master will add new files to the db, slave will not (and will report new files as an error)

####integrity_check_github.py All in one (messy) file, config inside the file

raid_scrub.sh

Simple shell script to perform periodical raid scrub (to detect raid incoherences), change RAID_ARRAY and MAIL_TO according to your needs. Require : nail to send email (installable via ipkg on synology)

Want to get more paranoid about data integrity :

http://www.zdnet.com/has-raid5-stopped-working-7000019939/ http://arstechnica.com/information-technology/2014/01/bitrot-and-atomic-cows-inside-next-gen-filesystems/ http://www.zdnet.com/blog/storage/data-corruption-is-worse-than-you-know/191

To run scripts from synology task scheduler

• Install python2.X package
• Add those two lines before calling any .py script (ensure filenames get read properly) : export LANG="en_US.utf8" export LC_ALL="en_US.utf8" Example : export LANG="en_US.utf8" export LC_ALL="en_US.utf8" python /volume2/homes/XXX/Hubic/main.py

What is my backup strategy ?

• 2 Nas in raid6 synced via Rsync
• Raid scrub every month
• Data integrity check (against a sha1) on each side every month
• • TODO : copy on hubic