On Jun 13, 2008, at 1:39 AM, Josh Nisly wrote:
I also realized that my previous patch (with the try/except)
actually adds an extra roundtrip for Windows servers, since it would
make one call for make_file_dict (which would always throw), and a
second that would actually work. Since I have clients backing up
many files over a 300ms latency link, this would be a real problem!
I vote that we always use what is now make_file_dict_old. I'll
submit a patch shortly that does this.
Josh,
While I completely agree that moving back to Python's stat functions
is where we want to go, I have one more concern. Perhaps it was not
the CPU speed that required cmodule.c, but Python? That is, Python 2.2
may be dreadfully slow at this operation (and thus leading to Ben
writing cmodule.c). Perhaps your use of a modern version of Python
made all the difference, and not the CPU.
So that we have the same comparison point, can you try it again with
Python 2.2.3?
I think dropping support for Python 2.2 at this point (right before
1.2.0) is hasty, but it probably goes without saying for rdiff-backup
1.3.x
Andrew
_______________________________________________
rdiff-backup-users mailing list at rdiff-backup-users@nongnu.org
http://lists.nongnu.org/mailman/listinfo/rdiff-backup-users
Wiki URL: http://rdiff-backup.solutionsfirst.com.au/index.php/RdiffBackupWiki
