http://www.ohloh.net/projects/4632/enlistments shows that ohloh failed to download the source but doesn't say what the error is.
StepMania project svn enlistment failed
about 1 month ago
about 1 month ago
Hi Steve,
We're having trouble getting the log from this repository. When I try to run this command...
svn log --stop-on-copy -r 1:HEAD 'http://svn.stepmania.com/svn/trunk'
...I get no response at all. Is this something that just requires us to wait a very long time, or is there a problem?
Thanks, Robin
about 1 month ago
I'm looking into it. Even doing a log with -r 1:1000 takes a minute so it's possible that something is just going horribly wrong server side.
about 1 month ago
What happened was the cvs to svn script seems to have either copied or inserted invalid bytes into the log message (probably copied). I don't know why the server is so slow, but I'm fixing the entries one at a time using svnadmin.
about 1 month ago
I believe I fixed all of the problems. The history should be downloadable now.
about 1 month ago
Hi Steve,
I'm sorry to report that I am still unable to fetch the log -- I get no response at all.
I'm assuming it must work for you -- about how long does it take for you to fetch the log as shown above? Do I just need to have extreme patience?
about 1 month ago
$ time svn log --stop-on-copy -r 1:HEAD http://svn.stepmania.com/svn/trunk > sm-history
real 15m23.038s user 0m2.488s sys 0m1.482s
Looks like the majority of the time it is sitting around waiting for the server. The history file is 146112 lines for 27592 revisions. I have no idea why the server is so slow though.
about 1 month ago
Yesterday, the individual revisions were being downloaded (I saw it get to a few thousand) but at some point it went back to waiting in queue for step 1 and is still there this morning. Has something gone wrong?
about 1 month ago
Hi Steve,
Nothing has gone wrong. Our system is designed to only let a particular download run for 8 hours before it is "paused" and put back in the queue. This prevents individual large projects from monopolizing all of our server time.
It looks like we got about halfway through the download before the timer ran out. I'd expect that the next run should complete the download.
After this initial bulk download completes, we shouldn't run into this issue again -- day-to-day updates usually run very quickly.