Brian Gernhardt wrote:
It just so happens that I'm one of the guys working on git. So I'll
see what I can do to help you. :-D I don't know how extensive your
git experience is, so bear with me if I'm saying things you already know.
Thanks. Your info was exactly that little bit I needed to grok this. But
I still have a few questions for pushing back to the repo on my server.
On Apr 9, 2009, at 2:50 PM, N. Turnage wrote:
Submodules are useful, but are also a little bit bleeding edge. They
are not widely used, and as such, sometimes have odd problems. If you
encounter any serious problems, feel free to poke the git mailing list
<g...@vger.kernel.org>. The list is both for development and users and
we try to be friendly to everyone.
Why is it displaying this way instead of just telling me which commit
I am using?
The description is either a branch name or something in the following
format: <tag>-<#>-g<SHA>. The tag is the most recent tag that
_doesn't_ contain the commit. # is the number of commits between the
commit in question and the tag. And SHA is the beginning of the full
SHA-1 (usually 6 characters). (This is the result of running `git
describe HEAD` in each submodule.)
I had to read this a couple of times, but this really helped understand
that I am using the latest commits regardless of which commit the tag
It looks like copy/move, page attachments, and share layouts are all
on their master branch while blog, comments, page_attachments, sns,
and tags are all a commit of their own. This can happen if the origin
repo for them moved ahead of where your submodule has them. If you
want, you can (usually) update them by going into the submodule and
running `git pull`. Then you can `git add` the submodule and `git
commit` in the supermodule to update it's records.
I actually went to each of the submodules and ran 'git reset --hard
master' . I don't know how right or wrong that is, but it worked.
I'm not sure what you were expecting to see, since I don't know the
details of your setup. But if this description isn't enough, I'll try
to help you get where you need to be.
Also, because I cloned my repo, instead of pulling it, how can I get
the branch on the server? Right now when I try and git push I get the
message that everything it up to date. Any info is appreciated.
This is where I am really confused.
Cloning should have set up the appropriate tracking information, but
would not have set up any push information. You might want to run a
command like the following, assuming the clone URL is also something
you can push to (i.e. not git:// or http://)
git config remote.origin.push refs/heads/*:refs/heads/*
Note that pushing to a non-bare repository (one that has a working
directory attached to it) isn't recommended, as it will change the
respository out from under you. You might want to change the push
line to something like "refs/heads/*:refs/remotes/home/*" and then
`git merge home/master` on the server instead. Or set up a bare
repository that both sides can clone from. (git --bare clone <URL>
What I would like is for the local branch I created to become a branch
on my server so that I can use Capistrano to deploy from my new branch.
I don't really want to merge my changes back into master just in case
something goes wrong and I have to go back and deploy from the master.
My local branch is called: radiant07
My git push/clone url is: g...@pixelnate.gitrepo.com:pixelnate/goodtaste.git
What confuses me is the 'refs/heads/*:refs/heads/*' bit. Could you maybe
explain that a little bit?
Thanks again for your help, Brian.
Radiant mailing list