Hi Guys, We have a very large darcs repository where I work, and despite working with darcs for a year, in all honestly we have found it difficult to work with due to 'Conflict Misery' and enormous memory usage and will be moving to some other DVCS.
So... what I want to do is convert repository with history over to another DVCS, which will probably be git. The problem is that every script I have tried to use to perform the conversion has failed when darcs tries to allocate memory and there isn't enough left. Fundamentally all of the scripts work by pulling each patch into a new repository one at a time, and then importing the differences into another DVCS. I can reproduce this behaviour just by running darcs standalone and trying to pull the very first patch from our repo into a brand new repo. (the first patch is quite big ~ 150Mb). Darcs fails with the 'resource exhausted (out of memory)' error. And this is on a Linux box with 4gig of RAM and not much else running. This happens whether I use darcs or darcs version 2 rc2 as the one doing the pulling. I have tried converting the repo in advance to take advantage of as many darcs2 improvements as possible, but even converting the repo fails due to memory exhaustion. I am very worried at the moment. Unless I can convert the history to a new DVCS, I may as well kiss it all goodbye: the repo has become too large and darcs performs terribly for it to be a useful means of exploring history. darcs annotate nearly always fails, for example (again, out of memory). So my question is: does anybody know of a way I can salvage my repository? I appreciate all of the hard work done by David and his team of contributors and wish all of the best to everyone involved. Best Regards, James. -- James Sadler _______________________________________________ darcs-users mailing list [email protected] http://lists.osuosl.org/mailman/listinfo/darcs-users
