1 |
>> The problem with my current push-style layout is that if one of the 3 |
2 |
>> machines is compromised, the attacker can delete or alter the backup |
3 |
>> of the compromised machine on the backup server. I can rsync the |
4 |
>> backups from the backup server to another machine, but if the backups |
5 |
>> are deleted or altered on the backup server, the rsync'ed copy on the |
6 |
>> next machine will also be deleted or altered. |
7 |
>> |
8 |
>> If I run a pull-style layout and the backup server is compromised, the |
9 |
>> attacker would have root read access to each of the 3 machines, but |
10 |
>> the attacker would already have access to backups from each of the 3 |
11 |
>> machines stored on the backup server itself so that's not really an |
12 |
>> issue. I would also have the added inconvenience of using openvpn or |
13 |
>> ssh -R for my laptop so the backup server can pull from it through any |
14 |
>> router. |
15 |
> |
16 |
> If an attacker can read the entire filesystem, he'll gain full root |
17 |
> privileges quickly. |
18 |
|
19 |
So if I push, I don't really have backups because anyone who breaks |
20 |
into the backed-up system can delete all of its backups like this: |
21 |
|
22 |
rdiff-backup --remove-older-than 1s backup@12.34.56.78::/path/to/backup |
23 |
|
24 |
And if I pull, none of my backed-up systems are secure because anyone |
25 |
who breaks into the backup server has root read privileges on every |
26 |
backed-up system and will thereby "gain full root privileges quickly." |
27 |
|
28 |
- Grant |