similar to: [patch] Make robust_rename() handle EXDEV.

Hi, How about adding now option '--backup-only' that means making backups only and don't change any destination files? (I posted similar patch a month ago, but the patch was made for nightly snapshot of 20020808, which was tooo old! Laugh at me...) I want to use rsync with LVM snapshot to make incremental backups like below: 1) Make LVM snapshot of file system and mount it.

There is a small race condition in rsync 2.5.6. When the transfer is finished, and the file is moved into place, there is a short time period where the new file is in place with the wrong permissions. When using rsync on a busy email server to replace the exim config file with a new file, exim will produce several complaints in that short period. This small patch fixes the problem, by making

Hit the "not a regular file (recv_files)" error last night having replaced a directory with a file. The really dumb part is that this was hit doing a link-dest on an empty destination so the robust_unlink would have presented no problem anyway. I can see no reason why non-directory special files should have any problem being replaced so i've changed that logic too. The new code

I came up with a patch to fix the problem of IO Errors caused by excluded files as did Eugene V. Chupriyanov below. Is there a chance that this change will show up in a future version of rsync? Is there a reason that we should not ignore IO errors when copy_links is off? Just want to make sure that I'm not missing something here that may corrupt my syncs.... Here's the version that

This patch adds the --link-by-hash=DIR option, which hard links received files in a link farm arranged by MD4 file hash. The result is that the system will only store one copy of the unique contents of each file, regardless of the file's name. (rev 5) * Fixed silly logic error. (rev 4) * Updated for committed robust_rename() patch, other changes in CVS. (rev 3) * Don't link empty

This patch adds the --link-by-hash=DIR option, which hard links received files in a link farm arranged by MD4 file hash. The result is that the system will only store one copy of the unique contents of each file, regardless of the file's name. (rev 4) * Updated for committed robust_rename() patch, other changes in CVS. (rev 3) * Don't link empty files. * Roll over to new file when

This patch adds the --link-by-hash=DIR option, which hard links received files in a link farm arranged by MD4 file hash. The result is that the system will only store one copy of the unique contents of each file, regardless of the file's name. (rev 3) * Don't link empty files. * Roll over to new file when filesystem maximum link count is reached. * If link fails for another reason, leave

I have attached a patch that supports a new "--direct-write" option. The result of using this option is to write directly to the destination files, instead of a temporary file first. The reason this patch is needed is for rsyncing to a device where the device is full or nearly full. Say that I am writing to a device that has 1 Meg free, and a 2 meg file on that device is out of date.

https://bugzilla.samba.org/show_bug.cgi?id=3752 Summary: rsync unusable with EncFS filesystem Product: rsync Version: 2.6.8 Platform: x86 OS/Version: Linux Status: NEW Severity: critical Priority: P3 Component: core AssignedTo: wayned@samba.org ReportedBy: micheala@jacey.org

Here's a patch that adds support for the creation of hard links over SFTP. Hard links are not used very often nowdays, but they do still have their uses and this is currently the most often requested improvement for SSHFS. To detect hard links the st_nlink, st_dev and st_ino attributes are usually used. I'll also post patches adding extensions for these and other attributes. Please

I am try to use rsync from my windows box (under cygwin) to back up "My Documents" on to a debian server (known on my home network as roo.home) running rsyncd from inetd where it is run as root. my rsyncd.conf file sets the gid and uid to user backup.backup thusly:- syslog facility = daemon uid = backup gid = backup hosts allow = 192.168.0.0/24 hosts deny = 0.0.0.0/0 timeout = 600

Operating System: OpenVMS ALPHA V7.3 Compiler: Compaq C T6.5 Compiler switches: /WARN=ENABLE=(LEVEL4, QUESTCODE) syscall.c is missing the "const" qualifiers for several of it's functions. This patch should supercede the previous patch I submitted. This was discovered while working on resovling the conflicts between signed and unsigned values. -John wb8tyw@qsl.network Personal

As I reported a while back rsync doesn't handle metadata (permissions and ownership) in dryrun mode. I offered to make a patch and that offer still stands. I didn't have the time for it until now and want to pick it up again. I had some ugly hack back then but I want to redo it in a clean way. I would like some input on my thoughts. IMHO, it would be ideally if the check for dry_run

I've run into a bug in the IO handling when reading a file. Suppose I have a file that lives on an NFS filesystem. That filesystem is NOT being exported with auth=0 permissions. So, if I try to access a file as root, it successfully opens the file, but subsequent reads fail with EACCES. This produces a destination file full of null bytes. I noticed this with 2.5.7, but checked 2.6.2 as

rsync (2.5.[67]) --delete fails on dirs with the w bit cleared. (example below) Rsync will sync a dir with w bit clear, but will not remove it with --delete. This is not a big problem, but it will create situations where there are 'orphaned' files. Has anyone else had this problem? It looks like a change would be needed in robust_unlink (util.c). This function would have to do a chmod

Hi, first of all, sorry for taking so long. Unfortunately, some other tasks kept coming up. Anyway, attached is the version of the flags patch, that is based on the one I'm using with 2.6.9. It is against the rsync-3.0.0pre9 release. I've included the option name change from the repository, so the option is now called --fileflags. Improved from the previously distributed version is the

I posted a patch a few days ago that adds copying of atime. At that time, it was just enabled with -t/--times. After some time, we have figured out that that choice might not have been the best. Here's a new version of the patch (relative to CVS) that adds -A/--copy-atime instead. It also includes a test case. Any feedback on this patch and/or the previous one that I posted?

https://bugzilla.samba.org/show_bug.cgi?id=5820 Summary: rsync does not replace symlink atomically Product: rsync Version: 3.0.4 Platform: All OS/Version: All Status: NEW Severity: normal Priority: P3 Component: core AssignedTo: wayned@samba.org ReportedBy: rsync@sysoev.ru

We're using rsync to update the tables on our email relays, and very occasionally we get an error in the logs saying that a table has the wrong permissions. This is because of race conditions in finish_transfer() which mean that it does not update files atomically. This makes rsync not entirely safe to use in our situation, where the files being synced are frequently opened for short periods

https://bugzilla.samba.org/show_bug.cgi?id=3594 Summary: --delete doesn't delete destination files that don't exist at the source Product: rsync Version: 2.6.7 Platform: PPC OS/Version: Mac OS X Status: NEW Severity: normal Priority: P3 Component: core AssignedTo: