From: Kern S. <ke...@si...> - 2009-10-12 18:50:05
|
On Monday 12 October 2009 19:26:13 Dan Langille wrote: > Kern Sibbald wrote: > > On Sunday 11 October 2009 02:40:57 Dan Langille wrote: > >> I think there is a problem with the regression testing "Temp switch to > >> branch 3.0.3". I'm seeing this issue each time the regression tests are > >> run: > >> > >> fatal: The remote end hung up unexpectedly > >> Begin nightly-disk Sat Oct 10 10:42:01 EDT 2009 > >> fatal: The remote end hung up unexpectedly > >> > >> I think it's the git commands. > >> > >> To test this theory: > >> > >> [dan@polo:~/src/BaculaRegressionTesting-TRUNK/regress] $ git pull > >> fatal: The remote end hung up unexpectedly > >> > >> I think the cause was this line in .git/config: > >> > >> url = git://bacula.git.sourceforge.net/gitroot/bacula > >> > >> Changing that file to > >> > >> url = git://bacula.git.sourceforge.net/gitroot/bacula/bacula > >> > >> removed the error. FYI. > > > > This was probably due to the fact that you did not delete the git > > directory after your first try that had the wrong path. > > > >> The next issue is this command: > >> > >> git checkout `git branch -r | grep "3\.0\.3" | line` > >> -bash: line: command not found > >> > >> It appears line(1) is not universal. How about head(1) instead? > >> > >> $ git checkout `git branch -r | grep "3\.0\.3" | head -1` > >> Note: moving to 'origin/3.0.3' which isn't a local branch > >> If you want to create a new branch from this checkout, you may do so > >> (now or later) by using -b with the checkout command again. Example: > >> git checkout -b <new_branch_name> > >> HEAD is now at fb11f7a... Update notes and date > >> > >> $ echo $? > >> 0 > >> > >> If agreed, I'll make the change to the file. > > > > Yes, I think it is a good fix. For the moment, no one but myself an Eric > > has commit permission -- to avoid accidental errors committing since git > > is significantly more complicated than svn. This will probably change as > > developers inform us that they feel 100% confortable with git. > > > > If you can figure out how to send me a "git format-patch" output file, > > I'll apply it for you. The instructions are in the Developer's guide. > > Otherwise let me know, and I will make the fix. > > I now have the branch in place, and I've run out of time today. Please > proceed without me. :) OK. I have committed the changes. Hopefully it will get pulled down by your regression tonight and all will work. Thanks for pointing this problem out -- it is yet another benefit of user regression testing for the next release :-) Kern |