Hi Mac,

Glad it is working well for you, I'm always happy to hear how people are 
using it! :)

Since backups is a vast topic and one which everyone does differently there 
hasn't been a recommended backup strategy for Mayan.  I'm hoping to change 
that and the bootstrap and app_registry apps included in the latest version 
are a big step in that direction.  In the meantime the best way to backup 
an installation of Mayan is copying the 'document_storate' directory and 
dumping the database with the tool specific to the database manager being 
used (mysqump for MySQL, or copying the mayan.sqlite for SQLite).  If you 
plan to restore to a database manager of a different kind than the one you 
used for the backup, use Django's dumpdata command instead.

./manage.py dumpdata > backup.json

on the installation where the backup will be restored do:

./manage.py flush  (answer yes)

answer also no when this message appears:

"You just installed Django's auth system, which means you don't have any 
superusers defined.
Would you like to create one now? (yes/no): "  <=== No

now restore the database from the indepentent format:

./manage.py loaddata backup.json

Disclaimer :)  I have done this (dumpdata & loaddata) on small and 
disposable databases.

Thanks!  The virtual appliance has been very well received, I plan to keep 
releasing in parallel with new release of the software.

PS: Since we are talking about backups, additional recommendations, backups 
strategies or wishlists for the future backup app are welcomed.

--Roberto


On Friday, January 11, 2013 2:00:30 AM UTC-4, Mac wrote:
>
> Hi,
>
> I just discovered the Mayan EDMS and it seems to be a great piece of 
> software and an excellent way of making my pdf archive searchable.
>
> How do I do a backup of the database and later restore everything to a new 
> installment?
>
> My scenario is that I have started to use Mayan by importing the appliance 
> image into virtualbox on my windows computer, uploading some documents, 
> creating users and tweaking metadata, index lists and other settings to my 
> liking. My plan is then to install it directly (either via the fabric file 
> or git) onto my Ubuntu file server. Now it would be nice if I could migrate 
> all settings and scanned documents from my virtual box to the file server...
>
> BTW using the Virtualbox appliance image was a great way of getting 
> started!
>

-- 



Reply via email to