Re: Stop password errors from getting reported as transfer errors

2007-11-04 Thread Matt McCutchen
On Sat, 2007-11-03 at 18:14 +, Wayne Davison committed: Stop password errors from getting reported as transfer errors. IMO, setting log_got_error back to zero after the password handling is a hack. Here are two possible better approaches: 1. Decide where the real division between setup and

DO NOT REPLY [Bug 5058] New: Rsync re-copies up-to-date files on external FAT drive after reconnect

2007-11-04 Thread samba-bugs
https://bugzilla.samba.org/show_bug.cgi?id=5058 Summary: Rsync re-copies up-to-date files on external FAT drive after reconnect Product: rsync Version: 2.6.9 Platform: x86 OS/Version: Windows XP Status: NEW

--relative simplification

2007-11-04 Thread Matt McCutchen
I observed the following about --relative: 1. From the user's perspective, its only effect is to change the suffix of a source argument path that is included in file-list paths. The suffix starts after the first ./ with --relative or the last / without it; if the marker is not present, the

DO NOT REPLY [Bug 5058] Rsync re-copies up-to-date files on external FAT drive after reconnect

2007-11-04 Thread samba-bugs
https://bugzilla.samba.org/show_bug.cgi?id=5058 --- Comment #1 from [EMAIL PROTECTED] 2007-11-04 12:07 CST --- Please reconnect the HD and then run rsync with the -i option to find out why it thinks the files need to be transferred. -- Configure bugmail:

DO NOT REPLY [Bug 4561] Add options --tweak, --no-tweak, --no-tweak-hlinked

2007-11-04 Thread samba-bugs
https://bugzilla.samba.org/show_bug.cgi?id=4561 --- Comment #1 from [EMAIL PROTECTED] 2007-11-04 14:51 CST --- Created an attachment (id=2960) -- (https://bugzilla.samba.org/attachment.cgi?id=2960action=view) Implementation of --tweak options Here, finally, is an implementation of

Keeping individual dirlinks

2007-11-04 Thread Matt McCutchen
It would be nice if rsync offered a way to specify individual dirlinks to keep without keeping all dirlinks. (Unison can do this.) For example, consider the following command: rsync -a --relative src/./ dest/ If rsync refrained from trimming a trailing /. off of source arguments (at least as

Re: Keeping individual dirlinks

2007-11-04 Thread Wayne Davison
On Sun, Nov 04, 2007 at 03:57:11PM -0500, Matt McCutchen wrote: If rsync refrained from trimming a trailing /. off of source arguments (at least as an option), one could keep a dirlink by passing it as an additional source argument with a /. and excluding the non-/. version so that rsync does

Re: Keeping individual dirlinks

2007-11-04 Thread Matt McCutchen
On Sun, 2007-11-04 at 13:34 -0800, Wayne Davison wrote: On Sun, Nov 04, 2007 at 03:57:11PM -0500, Matt McCutchen wrote: If rsync refrained from trimming a trailing /. off of source arguments (at least as an option), one could keep a dirlink by passing it as an additional source argument

DO NOT REPLY [Bug 5058] Rsync re-copies up-to-date files on external FAT drive after reconnect

2007-11-04 Thread samba-bugs
https://bugzilla.samba.org/show_bug.cgi?id=5058 [EMAIL PROTECTED] changed: What|Removed |Added Status|NEW |RESOLVED Resolution|

rsync delete

2007-11-04 Thread Jesse Thompson
Hello :) I am trying to build a backup system based upon Rsync, and I have a question about it's deleting features. I want to be able to use rsync to delete and remove a directory from the remote server, similar to rm -Rf directory The closest I seem to be able to get is the following

Re: rsync delete

2007-11-04 Thread Matt McCutchen
On Sun, 2007-11-04 at 13:50 -0800, Jesse Thompson wrote: What would be the closest approximation to something like this? # rsync -r --just-nuke-the-target remote::volume/directory Is it something that can be done with rsync? Yes, but it is a bit awkward because rsync considers deletion of

DO NOT REPLY [Bug 5058] Rsync re-copies up-to-date files on external FAT drive after reconnect

2007-11-04 Thread samba-bugs
https://bugzilla.samba.org/show_bug.cgi?id=5058 --- Comment #3 from [EMAIL PROTECTED] 2007-11-04 17:08 CST --- This may have been a time-change glitch. E.g., FAT stores times in localtime with no DST offset, so when DST changes, all times on a FAT filesystem appear to shift by an