A zipped binary of this compilation is at:
http://www.seawall.com/rsync.zip
New arguments for $Conf{RsyncArgs}
--checksum-seed=NUM
And for
$Conf{RsyncRestoreArgs}
--checksum-seed=NUM
--chmod=Dugo+rwx,Fugo+rw
or
--chmod=Do+s,ugo+w,Fo+w,+X
for example. D=directory, F=file, each
apppended with chmod options.
Just in case someone wants to try, it passes
"make test" and works as well as any other
in my brief experienceswith it.
Andy
--- On Tue 05/04, Andy Evans < endocrantz < at > excite.com > wrote:
From: Andy Evans [mailto: endocrantz < at > excite.com]
To: cbarratt < at > users.sourceforge.net, endocrantz < at > excite.com
Cc: backuppc-users < at > lists.sourceforge.net
Date: Tue, 4 May 2004 20:54:56 -0400 (EDT)
Subject: Re: [BackupPC-users] rsynd with workgroup / file permission
I did get your --checksum-seed=NUM in just fine, however in
testing, the distribution's chmod patch remains a problem
when applied to cygwin in these environments (workgroup). It
just doesn't do anything. I think then that this is
demonstrating that the problem is with chmod as the standard
rsync chmod is failing also. The older umask/chmod patch took
a different approach and does work (in older versions). Maybe
I'll try to get more testing in with just umask, and then
just chmod, to try to pin it down more precisely. I haven't
taken the issue onto the rsync list yet as I haven't got it
well defined enough - yet. As time allows I will continue to
work on it and try to improve the whole situation. The port
of the umask patch was not so pretty as the application of
the chmod patch with the distribution, of course. Things have
changed a bit in there since last year.
Andy
--- On Tue 05/04, Craig Barratt < cbarratt < at > users.sourceforge.net > wrote:
From: Craig Barratt [mailto: cbarratt < at > users.sourceforge.net]
To: endocrantz < at > excite.com
Cc: backuppc-users < at > lists.sourceforge.net
Date: Tue, 04 May 2004 07:40:52 -0700
Subject: Re: [BackupPC-users] rsynd with workgroup / file permission
On Mon, 2004-04-19 at 18:23, Andy E wrote:
I downloaded rsync 2.6.0 soruce & 2.6.0-1 patch (from cygwin mirrors)
applied the "-1" patch, the backuppc patch from sourceforge, and then
manually patched from
http://www.mail-archive.com/rsync < at > lists.samba.org/msg06792.html
This produces a fixed-csumseed static-umask/chmod-capable
rsync binary. The result being I can create files in any mode I wish.
This satisfies the requirements of the task I was discussing here.
Cygwin in a XP workgroup doesn't want to work as a rsyncd service
in a friendly way. Cygwin acknowleges this as it is really a windows
integration issue where workgroups just aren't real clever about
rights, certainly in comparison to domains.
I can buid a diff if anyone is interested or make the binary I built
available for download. Hopefully it is stable, and I'll continue to
use it (this is day one of the hack).
Thanks for the tips, Craig. Nice work with backuppc altogether.
Kudos for your efforts.
FYI, I have decided to change the --fixed-csumseed option to
--checksum-seed=NUM option. I'm attaching an email I sent to
the rsync mail list a few days ago.
Also, rsync 2.6.2 contains an optional patch in patches/chmod-option.diff.
Will this alternative work instead of the patch above? The reason I
suggest this is that the patches included with rsync are "close" to
becoming standard features.
I'll put a new rsync 2.6.2 with the --checksum-seed=NUM patch and one of
the chmod patches, plus the latest cygwin1.dll, on backups.sf.net soon.
Craig
#######################################################################
To: jw schultz
From: Craig Barratt
Cc: rsync < at > lists.samba.org
Date: Sat, 01 May 2004 17:06:10 -0700
Subj: Re: setting checksum_seed
jw schultz writes:
There was some talk last year about adding a --fixed-checksum-seed
option, but no consensus was reached. It shouldn't hurt to make the
seed value constant for certain applications, though, so you can feel
free to proceed in that direction for what you're doing for your client.
FYI, I just checked in some changes to the checksum_seed code that will
make it easier to have other options (besi des the batch ones) specify
that a constant seed value is needed.
I would really like a --fixed-csumseed option become a standard
feature in rsync. Just using the batch value (32761) is fine.
Can I contribute a patch? The reason I want this is the next
release of BackupPC will support rsync checksum caching, so that
backups don't need to recompute block or file checksums. This
requires a fixed checksum seed on the remote rsync, hence the
need for --fixed-csumseed. I've included this feature in a
pre-built rsync for cygwin that I include on the SourceForge
BackupPC downloads.
1. Yes, you may contribute a patch. I favor the idea of
being able to supply a checksum seed.
2. Lets get the option name down to a more reasonable
length. --checksum-seed should be sufficient.
I submitted a patch in Feb 2004 to add a --fixedcsum-seed option
(which only sets checksum_seed to 32761, the batch file value):
http://lists.samba.org/archive/rsync/2004-February/008616.html
Earlier, I submitted a patch (against 2.5.6pre1 in Jan 2003)
for --checksum-seed=NUM:
http://lists.samba.org/archive/rsync/2003-January/004845.html
Since I posted both of these patches, there was an interesting thread
started by Eran Tromer about potential block checksum collisions that
could be exploited by someone to trigger first-pass failures. See:
http://lists.samba.org/archive/rsync/2004-March/008821.html
The consequence is just a performance penalty, since with very
high probabi
lity the whole-file checksum fails, triggering the
second pass with the full checksum size, which will succeed.
Eran recommended that checksum_seed be more random than time().
Join Excite! - http://www.excite.com
The most personalized portal on the Web!
Join Excite! - http://www.excite.com
The most personalized portal on the Web!
