https://www.freelists.org/post/metastore-announce/metastore-...
https://joeyh.name/blog/entry/announcing_etckeeper/
In less than half a year it was replaced with find scripts.
http://git.joeyh.name/?p=etckeeper.git;a=commitdiff;h=a9ce99...
And I cannot blame etckeeper. Using binary format for storing metadata, i.e. owner, group, perms, xattrs, mtime, makes it too opaque for many use cases, especially in VCS.
metastore v1.1.0, as you can already read, was meant to give metastore (from 2008) a proper release, with fixed bugs and small improvements only. There are plans to make metastore more useful in the future, though.
Post title has been also changed, and it's fine. I no longer remember mine, but it was maybe too clunky. But.
metastore should be written using all lowercase letters.
It was originally written as a supplement to git, which does not store
all metadata, making it unsuitable for e.g. storing /etc in a
repository.
metastore can also be helpful if you want to create a tarball of a file
tree and make sure that "everything" (e.g. xattrs, mtime, owner, group)
is stored along with the files.
You may use it to store stuff on filesystem that doesn't support perms, owner & group for instance, like FAT, yet still be able to restore it later. Well, FAT doesn't support lot of things, so it's maybe a bad example. Anyway, I did use metastore to store metadata of some ext4 filesystem tree before backuping it on NTFS. It worked quite well.https://github.com/przemoc/metastore/issues/6
https://github.com/przemoc/metastore/issues/7
Drafts described there are not finalized and my view on some matters changed a bit. I'll try to update descriptions one day. metastore needs also some modularization to provide better OS-dependent stuff like xattrs.
All in all metastore will need more or less substantial reworking, to the point where rewriting it and providing converter from binary format may be a simpler solution and leading to cleaner (and thus more maintainable) code.