Clarification on Backup file paths

[eswbitto]

I'm in the testing phase of trying to backup both file content and database and I need some clarity on the database. I've followed the wiki on doing hot/cold backups, but when I look at my file path I have two postgresql directories.

/opt/alfresco/alf_data/postgresql
/opt/alfresco/postgresql

which one do I need to backup? Both?

[eswbitto]

@ fcorti
ok so I have implemented your backup strategy to see how it functions for my environment. Here are the steps that I did. I have a question at the end so hang tight.

1) I used the instructions from HERE
2) I figured out that I was not creating the .pgpass file and giving the right attributes (0660) that it calls for. Got that done and the script works for me. It created the tgz file so good to go.
3) In testing…I deleted a file from a testing site that I wanted to restore.
4) Moving on to test restoring: So I followed the instructions from HERE To a degree.
   A. As a precaution I moved the alf_data directory to a different location before doing any steps.
   B. I stopped the service…skipped down to the part for extracting the tgz file.
   C. In doing so when I tried to extract using

gzip -dc 2013-07-17_12-02_alfresco_dump.tgz | tar xf -
‍‍‍

I would get two additional tar files:
2013-07-17_12-02_alfresco_postgresql.tar
2013-07-17_12-02_alfresco_data.tgz

When I extracted the data.tgz file I get the whole alf_data directory back.
When I extract the postgresql.tar file I get a bunch of dat files.

Now here is my question….are the dat files in fact the lucene index files or are they something different.

The reason I ask is I removed the dat files altogether and just relied on the first file to see if I did restore the data. In which case I was successful. Any feedback would be helpful.

The other issue I have is the backup strategy doesn't allow for incremental restoration. Meaning…If I have to create a restore point of the previous day for one file. Then any content added after that time frame would be lost.

I'm open to any ways on how to restore incremental data.

[fcorti]

Hi ESWBitto,

Reading your comment I suppose you are trying to compact your backup as much as possibile… that's correct, of course.
Incremental backup in Alfresco is possible if we understand the backup strategy.

Trying to avoid to reinvent the wheel I suggest you <a href="http://forums.alfresco.com/forum/general/non-technical-alfresco-discussion/alfrtesco-backup-size-and...">this post</a> from MRogers (Alfresco Engineer).
In my personal opinion this strategy sounds a little bit tricky compared with other ECM leaders. 😉

About .dat files, please, confirm us they are stored in other sub folders than contentstore.
If yes, as I'm quite sure, they are not your documents (the documents are stored as bin files).
To understand better the use of the alf_data subfolders I suggest you to read <a href="https://forums.alfresco.com/forum/installation-upgrades-configuration-integration/configuration/alfd...">here</a>.

Hope this help you (and thank you for the request).