For now I run seafile inside a private network behind a vpn. But I'd rather find an alternative with a better security story. Something from a team that can justify their cryptographic constructs in detail.
I'll have to try using plain git annex from the commandline to try to understand what is going wrong (I already used it to clean some repositories that the assistant was refusing to delete)
PS: I tried it with 2 machines in the same LAN (both linux... though one of the 2 distros had a slightly older version of the git-annex package)
The default MAC algorithm to be applied on the filenames is HMACSHA1. A stronger one, for instance HMACSHA512, one can be chosen upon creation of the special remote with the option mac=HMACSHA512. The available MAC algorithms are HMACSHA1, HMACSHA224, HMACSHA256, HMACSHA384, and HMACSHA512.
In any case, while I want security in these systems to avoid mass survey style attacks, I assume anything I put over the wire to be compromised at least potentially. I keep the kind of files that need that level of encryption on encrypted flash drives.
That doesn't follow - you can solve the problem without containers, e.g. by encrypting or hashing the filenames.
Containers protect the file sizes, but they lose the ability to selectively sync certain files to certain nodes.
Regarding selective sync : you'll have to organize the files logically into containers. I think even Dropbox doesn't let you sync files selectively, you have to have content separated into folders.
I'm not saying that one is better than other, just giving an example of where de-duping at block level is needed as @lugg asked.
Sure; I was just talking about the file names. You can avoid the hierarchy part by transmitting it separately from the files in a secure way, though.
Regarding selective sync : you'll have to organize the files logically into containers. I think even Dropbox doesn't let you sync files selectively, you have to have content separated into folders.
git-annex does. You can implement rules to decide what files should be synced, which should be dropped locally, etc.
For example, an MP3 player running it might have a rule such as
filename(*.mp3) or filename(*.ogg)
Having an open source syncing service with client side encryption is great.
https://github.com/haiwen/seafile/wiki/Seafile-server-compon...
(Installed it yesterday on a debian LXC)
best part is that it does automatic, granular snapshots of my data - I've stopped using git for things like papers because I know it's backed up in seafile.
Also, anyone know how they compare? :)
EDIT: Login Log <- none for the free version?
I mainly mean it has client side encryption, the consequences of that I'd like to leave open since nobody really knows or can predict all possible vulnerabilities of all possible systems. But yes basically: your friend would have a very, very hard time doing anything useful with the data or the incoming packets.
none for the free version?
not sure what you mean, I host the thing myself (serisouly, 5$ a month for a DigitalOcean box: best way ever to spend money:) but the site seems to claim there is 1GB free storage if you try their service?
Documents and such I tend to keep in various public or private git repos; habits die hard.
Also, security track combined with the implementation language scare me to the extent of not really willing to expose the whole thing to the Internet. Just being superstitious here.
I'll be taking a look at Seafile now, looks promising.