11–20 of 36 posts
If you read the internals description it could just as well be about Borg, very similar principles here, though the application is very different. By the way, both buzhash and SHA-256 are kinda poor choices for a new system, especially one that targets servers.
both buzhash and SHA-256 are kinda poor choices for a new system Why?
[1] https://github.com/advancedtelematic/meta-updater
(Full disclosure: I work for Advanced Telematic, the creators and maintainers of the meta-updater Yocto layer.)
Earlier quoted context omitted.
Yap borg is the first thing I thought of. It already does a lot of this an more: encryption, configurable encoding, a rolling hash computed by the Buzhash algorithm and so on. Maybe it wasn't geared for CDN delivery during restores but otherwise I've been impressed by borg so far (haven't deployed it in production, only played with it locally though). https://github.com/borgbackup/borg This is a description of the in…
The "latest" version of that page has seen significant additions: http://borgbackup.readthedocs.io/en/latest/internals.html
I'm not sure I buy the embedded/IoT use case; OSTree is a really good model there and is more featureful. The "well, if your filesystem image delta happens to be in the form of a lot of very small files it's not so great for CDNs" doesn't strike me as a terribly good reason to give up everything OSTree gives you (especially with stuff like the meta-updater [1] Yocto integration). [1] https://github.com/advancedtelema…
I think OSTree is great — but for embedded devices that are installed in the wild, humm, uh, I don't think so? I am pretty sure there are better options than that.
As far as I can tell, the advantages compared to Borgbackup seem to be:
* casync offers control over which FS metadata is included
* casync, the server, exposes chunks over HTTP
* casync, the library, is written in C so is more easily used by systems software.
I'm betting we'll see machinectl integration. Excellent!
Earlier quoted context omitted.
both buzhash and SHA-256 are kinda poor choices for a new system Why?
Low software performance
Great. The chunked model (inspired by Borgbackup/Tarsnap) seems preferable to Docker layering, and diff-based approaches. As far as I can tell, the advantages compared to Borgbackup seem to be: * casync offers control over which FS metadata is included * casync, the server, exposes chunks over HTTP * casync, the library, is written in C so is more easily used by systems software. I'm betting we'll see machinectl inte…
systemd-nspawn integration is what I was thinking about too, so yeah! Nice work.