Andrew Ferguson pisze:
Jakub Krzyżewski wrote:
I do some debugging and it looks like problem is related to resource forks.
Commenting out this two lines:
if Globals.resource_forks_conn and rpath.isreg():
rpath.get_resource_fork()
in rpath.py causes that process never allocates more than 50MB memory.
Of course this is not perfect soulution, because all resouce forks in
backup are lost.
Jakub,
It sounds as if one of your files has an enormous resource fork, which
is very strange (and generally discouraged on Mac OS X).
If you uncomment those two lines, and then run rdiff-backup with the the
-v5 or -v7 option, it should tell you which file it is failing on.
Andrew
Files are InDesign and EPS graphics, most of them has 3-5kb rsrc , but
some of them has bigger rsrc, up to 100kb
As far as I know, problem is how rdiff-backup handles resource fork data
or it is software error,
because memory allocation slowly grows and script does not crash on
specific file.
I think there are 3 posibilities:
1 rdiff-backup creates huge data structures containing metedata and
keeps it all in memory.
2 some objects in memory containing resource fork aren't deleted
3 python interpreter has bug in memory allocation/garbage collection
_______________________________________________
rdiff-backup-users mailing list at [email protected]
http://lists.nongnu.org/mailman/listinfo/rdiff-backup-users
Wiki URL: http://rdiff-backup.solutionsfirst.com.au/index.php/RdiffBackupWiki
