linearly with the number
of files.
rdiff-backup and rsync use completely different protocols, and they
don't really share any code.
--
Ben Escoto
pgpmDZ722X1EZ.pgp
Description: PGP signature
___
rdiff-backup-users mailing list at rdiff-backup-users@n
rying to figure out why.
There aren't really any config files (partial exception: the
rdiff-backup-data/chars_to_quote file for the development versions).
For log files, see backup.log, file_statistics.*, and error_log.* in
the rdiff-backup-data directory.
--
Ben Esc
le
>From librsync.h's error code definitions:
RS_INTERNAL_ERROR = 107,/**< Probably a library bug. */
So you'd probably have better luck asking about this on the librsync
mailing list, if you haven't already.
--
Ben Escoto
pgpjifiCwqDnz.pgp
Description: PGP signature
_
e
rdiff-backup was running? That's the usual reason for UpdateErrors.
There's no reason to worry unless you are getting updateerrors every
session on the same files.
--
Ben Escoto
pgpjO4xJnY8d6.pgp
Description: PGP signature
___
rdiff-backup-u
backup
CVS at:
http://savannah.nongnu.org/cgi-bin/viewcvs/*checkout*/rdiff-backup/rdiff-backup/python-rdiff?rev=HEAD&content-type=text/plain
Note the blocksize there is fixed, so you'll have to edit the top to
test different sizes. Also I haven'
ile on a different disk, but
this would be slower, and also there would be no atomicity moving the
new file into place (this is important in case of failure).
--
Ben Escoto
pgpuE27UA5Eux.pgp
Description: PGP signature
___
rdiff-backup-users mailing list
that doesn't work it produces the error messages you're seeing.
However, those messages should be recoverable---if it can't find the
modules it just goes on without EA/ACL support.
So it probably doesn't matter if you're seeing those messages. If
you're
a, even the source remained complete unchanged for
all 1000 sessions.
--
Ben Escoto
pgpuoOOx08sXF.pgp
Description: PGP signature
___
rdiff-backup-users mailing list at rdiff-backup-users@nongnu.org
http://lists.nongnu.org/mailman/listinfo/rdiff-backup
S X then this part of
rdiff-backup obviously needs to be rewritten.
So anyone know how to fix this? Anyone actually want to fix it? :)
--
Ben Escoto
pgpGRm3lMYxIs.pgp
Description: PGP signature
___
rdiff-backup-users mailing list at rdiff-backup-
diff program works, but Rdiff.py
fails even when it uses the same version of librsync? Then it would
seem the problem is not librsync. Also you said Rdiff.py works on
files greater than 4GB (but less than 25GB)?
--
Ben Escoto
pgpQG5jurLFmR.pgp
Description: PGP signature
_
and see how it goes.
Well on my machine the mirror metadata takes about 10 bytes per file.
So I would expect 20MB of metadata for a 185MB volume only if you had
a lot of very small files.
--
Ben Escoto
pgpltx3p4ntOl.pgp
Description: PGP signature
__
e at
http://savannah.nongnu.org/cgi-bin/viewcvs/*checkout*/rdiff-backup/rdiff-backup/python-rdiff?rev=HEAD&content-type=text/plain
) and compare that to running rdiff. If rdiff works but python-rdiff
doesn't, then there is some problem with rdiff-backup
d guess the bottleneck is the large number of files and the
CPU and HD overhead rdiff-backup has on each (statting, writing them,
fsyncing, etc.). So I think Python's slowness/my programming could be
a problem (especially if you're using 0.1
upt your backups. Even in 0.6.0 deleting files may cause
data loss. (The next backup will recopy the file, but the previous
increments will be relative to the old version which was permanently
lost, not the new version just copied.)
--
Ben Escoto
___
t user (this doesn't work for root). The
rdiff-backup won't try to change uid/gid and the default uid/gid
should be what you wanted.
--
Ben Escoto
___
rdiff-backup-users mailing list at rdiff-backup-users@nongnu.org
http://lists.nongnu.org/m
rno 6] Device
not configured' is raised---perhaps the circumstances are particular
to your OS.
So I would try raising your verbosity to see which file caused the
problem. If it happens all the time on the file because it is not a
normal data file or is c
-backup deletes all the
files, and removes the directory, while NFS deletes most of the files,
and then tries remove the directory.)
I forget what the workaround was besides not using NFS.
--
Ben Escoto
___
rdiff-backup-users mailing list at rdiff
the backup restores will go fine (rdiff-backup will
just complain when it tries to restore the file(s) you deleted.)
--
Ben Escoto
___
rdiff-backup-users mailing list at rdiff-backup-users@nongnu.org
http://lists.nongnu.org/mailman/listinfo/rdiff-backup-users
Wiki URL: http://rdiff-backup.solutionsfirst.com.au/index.php/RdiffBackupWiki
'm not
sure why it's not working, perhaps a python bug or a quirk of the
setup.
But anyway, rdiff-backup already requires free space in the directory
being written to in many cases, so it would cause less confusion if it
just used this in every case.
I'm not sure wh
he previous finishes. Would this cause a problem?
Shouldn't be an issue since you are writing to separate destination
directories.
--
Ben Escoto
___
rdiff-backup-users mailing list at rdiff-backup-users@nongnu.org
http://list
w to
> write that "1/2" character. The network drive is on a Mac OS X
> Server.
Possible quick-fix: just --exclude that file and don't restore it.
There are some other quoting problems people have reported, I'll look
at those all at once later.
--
Ben Escoto
__
gt; On Sat, 30 Jul 2005, Keith Edmunds wrote:
> > Do you have something that explicitly relies upon the owner/group
> > of symlinks (eg, a find(1) script or similar)?
Yes why would you want symlink uid/gid preserved? For a script like
Keith suggests?
--
Ben Escoto
p has an order it follows when writing stuff to a backup
partition. When there is an error and it needs to regress, it knows
the order so can tell how to back everying out.
--
Ben Escoto
___
rdiff-backup-users mailing list at rdiff-backup-users@n
above
should work.
Does it still mess with the permissions when run as root? As root I
would hope it realizes it doesn't need to change anything.
--
Ben Escoto
___
rdiff-backup-users mailing list at rdiff-backup-users@nongnu.org
http://lists.nongnu.org/mailman/listinfo/rdiff-backup-users
Wiki URL: http://rdiff-backup.solutionsfirst.com.au/index.php/RdiffBackupWiki
ups do not require any free space on the source directory. If
any tempfiles were to go to the volume to be written when
restoring, it would lead to the pathological situation of being
able to backup a directory but unable to restore it.
So for
ff-backup didn't finish in 24h and the next one started up...
> and the target was really screwed up after that :)
Yep, that would do it :P
--
Ben Escoto
___
rdiff-backup-users mailing list at rdiff-backup-users@nongnu.org
http://lists.nongnu.
date.
Hi rdiff-backup is more session-based than rsync so the behavior
you're describing is normal. There's no way to get it to work like
rsync, where it doesn't care whether or not a session succeeds or
fails.
Perhaps you could backup nightly---run a script that backs up yo
; rdiff-backup fails then the process exiting will release the lock.
Is locking portable though? I guess it would be enough if we could
tell whether or not we had locking. If we have flock, and can't get a
lock, then we know another rdiff-backup process
s had a trailing
space, which probably caused some confusion also. I'm unsure if it's
a bug for rdiff-backup to interpret a space literally (especially a
trailing one) but that behavior isn't changed by this patch, so be
careful about that.
--
Ben Escoto
pgpG8PrjOjUcK.pgp
Descri
ast point me to a direction.
There was a bug in rdiff-backup that was just fixed recently, see
http://savannah.nongnu.org/cgi-bin/viewcvs/rdiff-backup/rdiff-backup/rdiff_backup/selection.py.diff?r2=1.41&r1=1.40&diff_format=u
for patch. Your example should work now.
--
Ben
something if
rdiff-backup exited with error, or just check to make sure there is
exactly one current_mirror in each rdiff-backup-data directory.
--
Ben Escoto
pgpAtPq7UY3vQ.pgp
Description: PGP signature
___
rdiff-backup-users mailing list at rdiff-
handle that :)
Workaround now in CVS, see
http://savannah.nongnu.org/cgi-bin/viewcvs/rdiff-backup/rdiff-backup/rdiff_backup/rpath.py.diff?r1=1.80&r2=1.81
--
Ben Escoto
pgp5BUUNC4T0a.pgp
Description: PGP signature
___
rdiff-backup-users mailing
3.
But I'd guess that python version 2.2.x installations are pretty
common. I'll see how hard it would be to preserve sym ownership while
avoiding a v2.3+ requirement.
--
Ben Escoto
pgp8N54LYQoEC.pgp
Description: PGP signature
___
rdiff-back
CVS rdiff-backup should now preserve symlink ownership. Thank you
for the bug report/test case.
--
Ben Escoto
pgp6uGW6JtW0Y.pgp
Description: PGP signature
___
rdiff-backup-users mailing list at rdiff-backup-users@nongnu.org
http://lists.nongnu.org/mailm
ral strings aren't marked with
>throughout?
No, just bad HTML form I think :)
Thanks for your patch and the Fink update. I'm going through some of
the documentation now and updating some obsolete bits.
--
Ben Escoto
pgpZTK8NiAE2X.pgp
Description: PGP signature
_
andable error message.
Ok, I just FAQ entry added (in CVS) about the free space and the weird
error message.
--
Ben Escoto
pgp4zj8bAqagV.pgp
Description: PGP signature
___
rdiff-backup-users mailing list at rdiff-backup-users@nongn
u need higher verbosity, 5 or
above). Anyway I can't tell what this is without more information.
--
Ben Escoto
pgpYjONDMqhSR.pgp
Description: PGP signature
___
rdiff-backup-users mailing list at rdiff-backup-users@nongnu.org
http://lists.nongnu.or
t. You may want to add it to rdiff-backup wiki
under ContribScripts at
http://rdiff-backup.solutionsfirst.com.au/index.php/RdiffBackupWiki
so it doesn't get lost. Also at least one of your suggested features
(merge increments) seemed similar to one in the Wiki's suggested
feature page.
savannah.nongnu.org/cgi-bin/viewcvs/rdiff-backup/rdiff-backup/rdiff_backup/Rdiff.py.diff?r2=1.11&r1=1.10&diff_format=u
and see if that also fixes the problem for you.
--
Ben Escoto
pgpqVV4ZbkJ9l.pgp
Description: PGP signature
___
rdiff-backup
ings
are backed up to) is a Mac OS X filesystem (UFS?)?
If the source filesystem is the same as the filesystem you're
restoring to, why can't whatever characters are in that shortcut be
created on that filesystem, just as they were created originally?
--
Ben Escoto
pgpv9BczbF
t rdiff-backup wants, or edit the file in the rdiff-backup-data
('chars-to-quote'?) which says which characters are quoted in the
destination dir.
--
Ben Escoto
pgpPUqRrZRaOk.pgp
Description: PGP signature
___
rdiff-backup-users mailing list
13: Fix for overflow error that could happen when backing up
files with dates far in the future on a 64bit machine to a 32 bit one.
Symlink ownership should be preserved now. Reported by Naoki
Takebayashi and others.
--
Ben Escoto
pgpl4W5KqvleK
on Support/Roxio/rdiff-backup.tmp.6' of
> type exceptions.OSError:
...
> 629, in chmodself.conn.os.chmod(self.path, permissions)
Looks like rdiff-backup cannot change the permissions of that file.
Perhaps you originally ran rdiff-backup as one user, and are now
writing to
patch in because there
are already similar switches like --no-acls and so on.
> 1122, in write_carbonfile
> fsobj = FSSpec(self.path)
> MacOS.Error: (-120, 'Directory not found')
But it does look like a Mac-specific carbonfile problem. Anyone more
familiar w
and I don't know of any "side effects" other than the one I just
looked up when you mentioned it :).
--
Ben Escoto
pgpImZQGjJ3yK.pgp
Description: PGP signature
___
rdiff-backup-users mailing list at rdiff-backup-users@nongnu.org
ht
OError: CRC check failed
It looks like there are errors in your destination directory. Perhaps
you could check the filesystem it's on for disk problems. Then see
which of your rdiff-backup-data/metadata*gz files have been corrupted.
--
Ben Escoto
pgpRjk0PdKVuL.pgp
Description: PGP signatu
he FAQ workaround doesn't tell you to move _both_ current_mirror
files.
> How would I ascertain if a snapshot has been corrupted?
I think I suggested checking out your mirror_metadata files. Since
they are gzipped you could zcat them and see if which one(s) have
gotten corrupted.
--
l what file it's happening to?
Also have you tried restoring different files/directories? Which get
that error and which don't?
--
Ben Escoto
pgpOBkzc5qVds.pgp
Description: PGP signature
___
rdiff-backup-users mailing list
her or not it needs to.
Anyway I just removed it, now you don't get a warning even if you are
backing up other people's files and aren't root.
--
Ben Escoto
pgpb2eiAhAMvY.pgp
Description: PGP signature
___
rdiff-backup-users mailing
up, like that described on Dean Gaudet's
page at:
http://arctic.org/~dean/rdiff-backup/unattended.html
So instead of restricting by path, it seems safer and easier to
restrict by ssh-key and/or username.
--
Ben Escoto
pgpWYLEwWYGH7.pgp
Description: PGP signature
_
hs, Does The Right Thing).
Ok I added
rm -rf $RPM_BUILD_ROOT/usr/share/doc/*
at the end of the %install section. I don't really understand why
this would help, but it didn't seem to hurt.
--
Ben Escoto
pgp626f2sAFDk.pgp
Description: PGP signature
e unsynced mirror_metadata dir problem. For that I
think these two will work:
http://savannah.nongnu.org/cgi-bin/viewcvs/rdiff-backup/rdiff-backup/rdiff_backup/rorpiter.py.diff?r2=1.27&r1=1.26&diff_format=u
http://savannah.nongnu.org/cgi-bin/viewcvs/rdiff-backup/rdiff-backup/rdiff_backup/rpath.py.diff?r
rvers in question,
> creating multiple backup accounts (to isolate the servers from each
> other) is likewise unworkable.
What's the problem with having thousands of users? It seems that
would be the safest way. Otherwise, why not write a script that
chec
rsion 1.0 will there be any problems backing the client up to the
> server?
Sorry this presumably won't work, the "protocol" changes from branch
to branch.
--
Ben Escoto
pgpdqQkbGdqxC.pgp
Description: PGP signature
___
rdiff-
trees have other empty
> directories in them... but no files of any kind. Is this normal/expected?
Yes, that's normal, but probably should be considered a bug (we might
as well delete those to reduce clutter). I added the issue to the bug
er tested the
> changes, but I saw the release announcement and thought this might
> have been overlooked since it doesn't look like it's been changed in
> the cvsweb.
Yes I also thought I had checked that patch in. I just did it now.
But as far as I know that won't fix
I think
http://savannah.nongnu.org/cgi-bin/viewcvs/rdiff-backup/rdiff-backup/rdiff_backup/Main.py.diff?r2=1.86&r1=1.85&diff_format=u
should do what you want.
--
Ben Escoto
pgpY3Woe0JOcO.pgp
Description: PGP signature
___
rdiff-backup-us
ot;--list-changed-since" output?
No, currently you have to parse the --list-changed-since output
(removing the changed description), and then feed it back in as a
filelist.
--
Ben Escoto
pgpM3FZ5ZWM97.pgp
Description: PGP signature
___
rdiff-backup-us
g up a pipe from one server to the other.
You could use ssh without encryption, or rsh, or whatever way you were
going to use before (netcat?). Just make sure rdiff-backup gets run
with the right uid on the server.
--
Ben Escoto
pgpe4oVIMahAy.pgp
Description: PGP signature
___
>>>>> Davy Durham <[EMAIL PROTECTED]>
>>>>> wrote the following on Wed, 17 Aug 2005 23:28:38 -0500
>
> BTW- when do you think those patches you created might be in a release?
They'll be in 1.0.1, but I don't know when exactly that wil
rking great!
Hmm, well in theory it shouldn't matter which version of rdiff-backup
you made the backups with, when restoring it should only overwrite
files that need it.
But since you seem satisfied I won't inquire further :)
--
Ben Escoto
pgpdeVFPF4UGq.pgp
Description: PGP signatu
hould be removed---it apparently doesn't clear up confusion :)
--
Ben Escoto
pgp62bIdk61by.pgp
Description: PGP signature
___
rdiff-backup-users mailing list at rdiff-backup-users@nongnu.org
http://lists.nongnu.org/mailman/listinfo/rdiff-backup-use
guess you could delete the ones you have, but I wouldn't recommend it.
(Then again I wouldn't recommend deleting a file from the mirror
directory.) Anyway, 0.6 and 0.12+ behave differently in this case.
--
Ben Escoto
pgp2it5mrnLmP.pgp
Description: PGP signature
__
st releases onto the site so we can at least see the
> old_stable versions (which work better at the moment than 1.0)
Most of the previous releases are available from Savannah from that
same directory:
http://savannah.nongnu.org/download/rdiff-backup/
--
Ben Es
ou could have it run
cd ; sudo -u Y rdiff-backup --restrict . --server
to avoid patching rdiff-backup and for an additional layer of
security. Anyway it's up to you, I'm mainly posting to say I think
the security bug is fixed in CVS. Thank you for the
er:
rdiff-backup --include /home/alice --include /home/bob \
--exclude ** /home /backup
or mkdir /backup/alice and /backup/bob and then
rdiff-backup /home/alice /backup/alice
rdiff-backup /home/bob /backup/bob
--
Ben Escoto
pgpAAu7GywGfl.pgp
Description: PGP signature
relevant code:
xattr.listxattr(rp.path)
if write:
xattr.setxattr(rp.path, "user.test", "test val")
assert xattr.getxattr(rp.path, "user.test") == "test val"
So if you want to see what is going on you could try testing x
2. If you avoided
backing up device files (--exclude-device-files) you wouldn't get the
error you got, so that may be an easy workaround.
I'll look at it in a bit, maybe I can just take the os.mknod code out
of python 2.4 and stick it in rdiff-backup.
--
Ben Escoto
pgpp0gl1G4Z78.pgp
up to a non-case-sensitive file
system. Since you are backing up to the same type of file system you
can try
--override-chars-to-quote ''
which tells it not to quote anything.
--
Ben Escoto
pgppBeYKzbRqz.pgp
Description: PGP signature
__
tored separately, and
you were willing to exclude device files anyway).
--
Ben Escoto
pgpChZoBAbMeg.pgp
Description: PGP signature
___
rdiff-backup-users mailing list at rdiff-backup-users@nongnu.org
http://lists.nongnu.org/mailman/listinfo/rdiff-backu
tectures I'd have to provide
binaries for i386, x86_64, Fedora, Suse, and various combinations and
versions of these.
So now I've just tried to make a source rpm that will work for many
people, and leave the binaries up to the distributions. If you want
ce with Mac OS X. How do people change the ownership of
symbolic links on your system without lchown?
--
Ben Escoto
pgpq4A64mMam7.pgp
Description: PGP signature
___
rdiff-backup-users mailing list at rdiff-backup-users@nongnu.org
htt
kups. (That
time, 1125702911, is 2005-09-02 18:15:11 where I am, so looks
reasonable.)
--
Ben Escoto
pgpeep7B7BxT3.pgp
Description: PGP signature
___
rdiff-backup-users mailing list at rdiff-backup-users@nongnu.org
http://lists.nongn
m without lchown?
>
> I don't know.
So how popular is 10.3 and how long will people keep using it? Is it
worth writing a test to handle this case? Or can we just call this a
bug in the OS?
--
Ben Escoto
pgpeNdPGV9tDJ.pgp
Description: PGP signature
_
=u
to avoid the lchown requirement. They'll be in v1.0.1 which will be
released soon.
--
Ben Escoto
pgppFGxWhPLc5.pgp
Description: PGP signature
___
rdiff-backup-users mailing list at rdiff-backup-users@nongnu.org
http://lists.nongnu.org/m
--include '/home' --include '/usr/local/bin' --include
> '/usr/local/sbin' --include '/var/lib/dpkg/status' --include
> '/var/lib/dpkg/status-old' --exclude '/*' /
> [EMAIL PROTECTED]::/home/backupschizo/schizo
> Warning: su:
chine).
What does your metadata file say? (e.g. zless
mirror_metadata..snapshot.gz, then search for that file) If all
the files are still there you could either manually edit your metadata
file or move the directories around so they are in sync.
I don't know how this could have happ
n no longer required, which is good news for Mac OS X 10.3.
--
Ben Escoto
pgpIlpZWExrMa.pgp
Description: PGP signature
___
rdiff-backup-users mailing list at rdiff-backup-users@nongnu.org
http://lists.nongnu.org/mailman/listinfo/rdiff-backup-users
Wi
I access using ssh, that has --restrict-update-only on
> its commandline (as specified in command= in public key), how does a
> --exclude work?
If you have --restrict-update-only, someone can still run rdiff-backup
with --exclude '**'. This will delete the mirror tre
e it totally disables remote operation on Mac OS X
and other non-case-sensitive systems. But no one complained until
today so I wanted to see if I was missing something.
--
Ben Escoto
pgpe0SCq3U2n5.pgp
Description: PGP signature
___
rdiff-backup-users mai
ll getting error parsing commands. the back data and the target
> are all on the same machine.
Maybe you are using some weird shell that doesn't recognize " as a
quoting character? Use capital "D" for days. You can also try:
cd "/samba-saa/SAA Projec
ovided by the
pyxattr package. For ACL support, rdiff-backup looks for the
"posix1e" package, as provided by the pylibacl package. There are
links to these packages on the homepage.
I have no idea whether Mac OS X supports ACLs or user EAs at all, and
if it does, whether
gs
writing to the destination directory. However, if you just keep
--exclude'ing the lost+found directory, I think that should work. In
practice rdiff-backup will just read the mirror_metadata file, and
won't notice if the lost+found directory comes back.
--
Ben Escoto
pgpABnqxFPcE5.pg
to add a check and then just disable fsyncing if it's
not available, but I don't think this would be entirely safe.
--
Ben Escoto
pgp7GBvqqlZkL.pgp
Description: PGP signature
___
rdiff-backup-users mailing list at rdiff-backup-users@nongnu.o
ese
> characters or something..
...
> Traceback (most recent call last):
>File "/usr/bin/rdiff-backup", line 23, in ?
> rdiff_backup.Main.Main(sys.argv[1:])
> line 352, in next
> del self.cache_dict[self.cache_indicies[0]]
Is this a complete t
ince I
don't have any experience with that OS.
To whoever (John Goerzen?) wrote the get_carbonfile() function: maybe
it shouldn't catch the MacOS.error exception? It seems like it
catches it, and returns a bad cfile which write_carbonfile chokes on.
--
Ben Escoto
pgpYomH3NhWab.pg
uids and gids) so restoring to a new system should
work. You can modify this behavior with the --user-mapping-file or
--group-mapping-file switches.
--
Ben Escoto
pgpIrZnZZq2DM.pgp
Description: PGP signature
___
rdiff-backup-users mailing list at rdif
Mirror/bin/rdiff-
> backup.tmp.43'
Can you run it with high verbosity (maybe 7, or at least 5) so we can
tell what file it's failing on? Then do an ls -al on that file on the
source side and post it.
--
Ben Escoto
pgpbF2jm2i5gH.pgp
Description: PGP signature
__
her semi-faulty
equipment (memory, etc.). Running rdiff-backup and a few other apps
simultaneously draws enough power to make the system unstable in a way
only a particular app notices.
> This may not be the fault of Ben Escoto, then again...
...most things are? :-P
--
Ben Escoto
pgpHuXgdhoQ
-backup doesn't have much C code in it, and python code should
never segfault no matter how buggy it is. So in general rdiff-backup
segfaults are problems with the python implementation.
--
Ben Escoto
pgp8LrTD1uDn0.pgp
Description: PGP signature
___
ike the bug described in the FAQ:
http://www.nongnu.org/rdiff-backup/FAQ.html#dir_not_empty
As it says, I think this is probably a bug in NFS (it doesn't seem to
perform certain operations in the right order).
--
Ben Escoto
pgpav2hZxstIg.pgp
Description: PGP signature
_
rom a case insensitive file system.
Well it's hard to determine if a file system is case sensitive unless
you can write to it (so right now the source isn't checked for case
sensitivity).
Maybe resource fork ability could be a good substitute for handling
the Mac OS case: if the source and d
= 524288
conn_bufsize = 1572864
I'd be curious to see if this makes much of a difference.
--
Ben Escoto
pgpX2a37HZcta.pgp
Description: PGP signature
___
rdiff-backup-users mailing list at rdiff-backup-users@nongnu.org
http://lists.nongnu.org/mail
g is only when doing incremental backups so it can regress if
necessary. But on an initial backup there is nothing to regress to so
fsyncing isn't necessary.
The strace stuff is interesting, but you may want to try increasing
those bufsizes first, since it woul
is.
In the meantime, you could avoid running rdiff-backup as root, at
least on the writing side. For instance, instead of
rdiff-backup / /winhome/Temp
you could run:
rdiff-backup / @localhost::/winhome/Temp
This should work because rdiff-backup only writes to add the special
per
certain messages, or
may wait a month before trying to catch up.
Hmm some people may have the ability to subscribe to certain
threads, which probably works similarly, but I don't have this
functionality.
3) Auto-responders (like vacation bots) don't spam the li
t FSSpec
import MacOS
Do you still need to add --no-carbonfile, or does that fix that
problem?
--
Ben Escoto
pgpmEn4xOZnB2.pgp
Description: PGP signature
___
rdiff-backup-users mailing list at rdiff-backup-users@nongnu.org
http://list
se).
Ok I get it. I think this feature was requested before but I never
understood the use before. Shouldn't be hard to add---I added a
support request at
https://savannah.nongnu.org/support/index.php?func=detailitem&item_id=104755
and should get to it by the next version.
--
Be
there's no way to do it then I'll just add a test to fs_abilities,
and display that warning you mentioned if the filesystem doesn't
support fsync.
--
Ben Escoto
pgpMqMV1X91LC.pgp
Description: PGP signature
___
rdiff-backup-users mailin
ule. Maybe there is a different bug in cygwin's fsync.
Anyway, for a simple script, try this:
echo testing > fsync-testfile
python -c 'import os; os.fsync(os.open("fsync-testfile", os.O_RDONLY))'
rm fsync-testfile
if it works, python w
1 - 100 of 283 matches
Mail list logo