tree-reader.py is a plugin for [bareos.org]. It accepts the plugin options file and read.
file needs to point at the output of qumulo-tree-walk with columns path,type,id,size,blocks,owner,change_time (in that order).
anon.py anonymises the output of qumolo-tree-walk like this:
- replaces all filename / directories by randomly generated words
- replaces the owner id by a random number
- replaces the id by a random number
split-up.py splits the output list into four files (test.pid, test.full, test.diff, test.incr). This is needed for make-bareos-tables.sql.
The idea is that test.pid contains the path table that bareos needs (a simple map id -> path). test.full contains the file data in a
slightly modified way (for example it only stores filename + path id instead of the whole path). If a file is listed multiple times inside the output
then only the first occurence is saved inside test.full. The second occurance is stored inside test.diff and the third occurance is stored inside
of test.incr.
make-bareos-tables.sql takes the four files created with split-up.py and creates the postgresql tables & indices that bareos uses internally to
enable the building of restore trees. Be careful: this script will first delete all tables it intends to fill if they exist. Do not try to use
this with a real bareos database!
It will use the jobids 1, 2, 3 for the full, diff, incremental backup.