similar to: readdir() and read() errors ignored

Greetings, I'd like to propose a new option to rsync, which causes it to read device files as if they were regular files. This includes pipes, character devices and block devices (I'm not sure about sockets). The main motivation is cases where you need to synchronize a large amount of data that is not available as regular files, as in the following scenarios: * Keep a copy of a block

In short: Platform: linux with 2.4 kernel Version: rsync 2.6.6 Command line: rsync266 -av -W --bwlimit=1 /mnt/somedir/rsync-2.6.6.tar.gz ./ Destination: local disk Source: file on a smbfs mounted filesystem; share is exported on a NT 4.0 workstation over a very slow and unstable link Result: Rsync completes operation with no special message, but the resulting file is damaged, large

I'm using rsync to mirror files from a Windows XP machine mounted via smbfs. Apparently I have something configured wrong as I get a "permisson denied" error accessing some of the files on the smbfs mount with cp, od, etc. However, rsync produces no error messages on these files. It happily creates files in the target directory that are the right size, but filled with null bytes.

Attached is a patch that makes window strides constant when files are walked with a constant block size. In these cases, it completely avoids all memmoves. In my simple local test of rsyncing 57MB of 10 local files, memmoved bytes went from 18MB to zero. I haven't tested this for a big variety of file cases. I think that this will always reduce the memmoves involved with walking a large

smbclient (and smbtar) in version 2.2.7a (and prior) has problems with large files (> 4GB). The following patch (against 2.2.7a) fixes all known problems with this. This code has been checked into the CVS tree in all branches as well. -- ====================================================================== Herb Lewis Silicon Graphics Networking Engineer

In looking at source, I started at fileio and found write routines but no read routines. I found a 'WRITE_SIZE' (32K), but no 'READ_SIZE' -- is that' what the MAX_MAP_SIZE (256K)? I would like to make so that rsync can use larger I/O sizes if (maybe a command line option?).... The map routine led me to receiver -- where it looks like it is responsible for reading the file.

(Erm, sorry about that...) I was talking about errors of the type: Got 0 bytes. The problem was sporatic, and seemed to be load/speed based. That is, the faster the Linux box, the more errors we had. A look through clitar.c led me to a solution. Currently, clitar.c is written to pad a file with zeros if a file read error occurs. This is normal. All of our errors were of the 0 byte length

I may have a deficiency in understanding the procedure for ADS authentication with samba, but most of the server setup works so far. I have bound a Red Hat Enterprise 5 server to our windows domain, it shows up in DNS and ADS, I can ping it, but I can't get samba shares to be accessible to users, or even get the smbclient to return shares properly. wbinfo -g returns the domain groups

Hi, This weekend I migrated a samba-server to new hardware and software : I'm running samba 3.2.7-11.3.2 on Opensuse 11.1. Everything works fine I believe except for an old DOS-application, written in Clipper. It's too slow to be usable. I tried to debug this but I'm stuck. I put the loglevel to 10, but I can't see anything wrong. I first added to veto oplock files :

https://bugzilla.samba.org/show_bug.cgi?id=8177 Summary: Problems with big sparsed files Product: rsync Version: 3.0.8 Platform: x64 OS/Version: Linux Status: NEW Severity: normal Priority: P5 Component: core AssignedTo: wayned at samba.org ReportedBy: joluinfante at gmail.com

I am trying to use the value of an ID variable in an if statement and not getting the results I expected. # ID values for two school districts > with(rf, tapply(DistrictID, DistrictName, min) ) Aberdeen School Dist. # 58 Buhl Joint School District 59340 53409 This creates DNAME as I expected ...

Hi all, I''ve got alpha versions of nread and nwrite now added to the win32-file class in CVS. They seem to work fine, although I''ve noticed one bug in the File#read method. Let''s say we have a text file with "hello world" in it. If I call nread with no arguments, it works fine: fh = File.nopen("C:\\test.txt") p fh.nread -> "hello

Hi, While testing the upgrade from 2.0.6 to 2.0.7 I found some kind of misbehaviour of inherit permissions. It was already known that the SGID bit on directories is only useable with "inherit permissions = yes". Most of my shares depend heavily on this (unix) feature. Today I found out I cannot change a file attribute to read-only from NT while "inherit permissions" is set

I have a bit of an oddity here: I'm using Samba 4.2.14 with AD, using Samba's internal DNS to serve to my domain (and forwarding others to Google Public DNS 8.8.8.8). However, it looks like client queries (and subsequent forwards) to a vendor's URL (www.pitneybowes.us) fail, but are successful if I query Google directly: [root at XXXX~]# nslookup -type=any www.pitneybowes.us >

[Automatic forwarding from R-bugs failed. This message has been manually forwarded.] Hi all! I'm trying to add Thomas Lumley's defmacro() function Lumley T. "Programmer's Niche: Macros in {R}", R News, 2001, Vol 1, No. 3, pp 11--13, \url{http://CRAN.R-project.org/doc/Rnews/} to the gtools package (provided that Thomas gives his OK). And I've encountered an error in

Hi all, So far, so good. Things seem to work just fine and, in the case of nread, I am getting incredible performance. I can''t tell if this is due to some kind of caching, however. Anyway, there a couple of things I want to work out yet. The first is File#nread and Really Big Files. I created a 500 mb text file and sucked it into memory using Ruby''s File#read method just

period of time (hours up to weeks sometimes), and then 'certain' pc's would no longer be able to connect to samba, but other pc's would still be working. And the pc's affected are not the same each time. this might be the case if what is REALLY going on is that the pc's that are STILL working never lost their connection to SAMBA/iX, so they already(still) have a working

(Samba 3.0.1, RedHat 9, share is a w2k3, security = ADS) Here is a snippet of debug level 3 output of an smbclient tar operation, with error at the end. Command is: # smbclient \\\\snapper\\dfs <password> -U Administrator -E -W CISWINNET -D home -d3 -Tqca /tmp/test.tar [2004/01/14 15:05:10, 3] lib/util.c:dos_clean_name(549) dos_clean_name

Here''s my comments for the preliminary ZFS Crypto review. - Dan Webrev: http://cr.opensolaris.org/~darrenm/zfs-crypto-gate/webrev/ General comments: DEA-1 - SCCS keywords need to be removed DEA-2 - Copyright updated ------------------------------------------------------------------ usr/src/lib/libcryptoutil/common/keyfile.c pkcs11_read_data() This code in pkcs11_read_data() scares

We currently trigger BUG when VIRTIO_NET_F_CTRL_VQ is not set but one of features depending on it is. That's not a friendly way to report errors to hypervisors. Let's check, and fail probe instead. Cc: Rusty Russell <rusty at rustcorp.com.au> Cc: Cornelia Huck <cornelia.huck at de.ibm.com> Cc: Wanlong Gao <gaowanlong at cn.fujitsu.com> Signed-off-by: Michael S. Tsirkin