Hi Kevin, I believe I have resolved your issue.
The error number (102) returned by the C test program I sent you is an error specific to sockets. However, when you did `ls -al` and got -rwxr-xr-x 1 kwh staff 0 Apr 3 18:38 Commands* -rwxr-xr-x 1 kwh staff 0 Apr 3 18:38 Notification* ls did not indicate that those files are sockets! I installed Adobe Reader 8 on my PPC Mac and got it to create those files, so I did: $ ls -l ~/Library/Acrobat\ User\ Data/8.0_ppc/Synchronizer/ srwxr-xr-x 1 owsla owsla 0B Jul 16 13:04 Commands srwxr-xr-x 1 owsla owsla 0B Jul 16 13:04 Notification See the difference in the first character? I backed-up the directory with rdiff-backup and uncovered a bug related to extended attributes, which I have fixed in CVS for version 1.1.13. However, I was not able to trigger the bug you got because my system is correctly interpreting those files as sockets. Something is broken for you below the level of rdiff-backup, and below the level of python. What happens if you run: $ file ~/Library/Acrobat\ User\ Data/8.0_ppc/Synchronizer/ I get 'socket' listed for both of those files. I suggest three possible remedies: 1) Run 'Repair Disk Permissions' using Apple's Disk Utility 2) Delete the Synchronizer folder. Acrobat will recreate it if needed 3) Reinstall Adobe Reader 8 Since rdiff-backup isn't being told that those files are sockets, it proceeds with trying to open them with 'rb' as the mode, triggering the ListError that you observe. Best, Andrew -- Andrew Ferguson - [EMAIL PROTECTED] _______________________________________________ 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
