similar to: How is "Canonicalized pathnames" bit handled by samba server??

Hi All, I installed OpenSUSE 11.0, updated samba server to 3.2.4 version. Added two users in linux and samba with user names "test1" and "Test2". (Please note the capital letter in Test2). Created two shares in /srv/samba/shares "test1" and "Test2". The permissions of the shares are as below. drwx------ 2 test1 users 4096 2009-12-11 12:42 test1 drwx------

Revering to the message: [Samba] smbmount and PCShare [Helios] from Sat, 12 Jul 2003 20:41:45 +0200 still trying to get a working connection between a SuSE-Linux-System and a Helios-System [PCShare - http://www.helios.de/products/PCS/PCShare.html ] The smbmount works well, but I can't copy a file to the mounted file system. The SMB Command which fails is the "Create (0x03)". Has

Full_Name: Lyndon Drake Version: Version 0.65.0 Under development (unstable) (August 07, 1999) OS: Linux 2.2.2 Submission from: (NULL) (202.14.102.52) I was attempting to install the DSE package from a directory whose pathname includes spaces. The INSTALL script interprets spaces in the pathname as a separator. For example: [lyndon@beat Rdse]$ pwd /home/lyndon/doc/uni/415.704 Computer Systems

http://bugzilla.mindrot.org/show_bug.cgi?id=978 Summary: The supplied realpath function mangles pathnames. Product: Portable OpenSSH Version: 3.9p1 Platform: Other OS/Version: other Status: NEW Severity: minor Priority: P3 Component: Miscellaneous AssignedTo: openssh-bugs at mindrot.org

I've written a GUI for using rsync on windows machines for third-party use. The project has reached a beta state and is now being tested in live situations. However, during these tests, I've encountered some odd behavior with specific pathnames. For the record: I'm currently using a cygwin distribution which includes rsync 2.6.3 to run rsync on Windows machines. The remote server is

Hello, I replaced an NT4 server by a Samba 3.0.14a ldap server following the instructions in 'Samba by example'. Everything works fine except for some files on a W2003 SP2 server: - when the W2003 server uses the Linux PDC, from a WinXP client, some files are not writable, and ACLs can be listed but not changed. - when the NT4 server is up, the same files on the W2003 server can be

I'm trying to use a Samba (3.0.37) share on my ASUS RT-ac66R router as a destination for scans from my Kodak Office Hero 6.1 all-in-one printer. I'm currently getting an invalid User ID or Password error message. I have been unable to increase the Samba log level at all even after editing the log level in smb.conf. Before an old router running DDWRT (including Samba) failed, scanning to

We're having a problem with an application called Grade Machine running on Windows 98 workstations. The problem occurs when trying to save when overwriting the currently open class file. It works on NT 3.51, it works on the local drive, it works on Samba when short directory names are used, it doesn't work under Samba 2.0.10 or 2.2.2 when saving to a file in a folder such as "My

My clients are Windows XP SP1 and SP2, members of a Samba-PDC NT domain (tested 3.0.7 and 3.0.10, same result). Attached is ethereal output of a two packet client-server exchange that takes place when an offline files sync is done. SP1 quickly does this exchange twice - first broadcast, then unicast (as attached) and goes on its way. SP2 tries, pauses many seconds, tries again, finally

Good day time! We've noticed the following issue with Samba 3.0.1 on Linux. When SMB client asks for the first time on FID for a locking or read request on an opened file, it causes an ACK to be sent from client side. Windows server also requests ACK BUT much more rarely. Sending ACK causes additional overhead when working in heavy locking and unlocking environment. Windows server

Good day time! We've noticed the following issue with Samba 3.0.1 on Linux. When SMB client asks for the first time on FID for a locking or read request on an opened file, it causes an ACK to be sent from client side. Windows server also requests ACK BUT much more rarely. Sending ACK causes additional overhead when working in heavy locking and unlocking environment. Windows server

Hello, I have just migrated an old OS/2 file server to a Linux box with Samba 3.0.23a. Now the OS/2 clients crash from time to time. I found a way to reproduce/force the crash using PMMail and did some experiments. The popuplog.os2 on the clients (Warp4 and eComStation) always names a sys3175 in pmshell.exe / doscall1.dll. Using Ethereal and comparing the network traffic between a) a client

Hi, using too long pathnames for bitmap() crash R on WinXP. I've verified that this is the case with R version 2.4.1 Patched (2007-03-25 r40958) and R version 2.5.0 alpha (2007-03-30 r40957). I cannot reproduce it on Linux. REPRODUCIBLE EXAMPLE: % Rterm --vanilla # Tell R where Ghostscript is gsexe <- "C:/gs/gs8.54/bin/gswin32c.exe"; gsexe <- "C:/Program

Hi, using samba 2.2.5-UL (on a SuSE-SLOX-System) we have to mangle long pathnames to short ones. We need this for some of our apps which generate batch-files (*.bat) for compilation. Normally this works correctly, exept for a directory named "Only_for_generation". This directory is mangled into "Only_~%0" (it' s NULL at the end). And this is the problem. In

Dear R helpers: Thanks for the previous reply. I am using Friedman racing test. According the the book "Pratical Nonprametric Statistic" by WJ Conover, after computing the statistics, he suggested to use chi-squared or F distribution to accept or reject null hypothesis. After looking into the source code, I found that R uses chi-sqaured distribution as below: PVAL <-

Dear Rs I have two vectors A and B where A is V1 d 1 0.000100000 1.123278 2 0.002186431 1.120448 3 0.004351214 1.106661 4 0.006515998 1.107713 5 0.008680781 1.107667 6 0.013010348 1.106353 7 0.019504698 1.104077 8 0.034658181 1.103202 9 0.051976447 1.103200 10 0.073624280 1.094825 11 0.093085682 1.085123 12 0.095250465 1.087325 13 0.132051782 1.086158 14 0.168853098

Hi, I starting a new project, and have run into behavior I cannot replicate in irb, (i.e. outside of using rspec) when the directory? returns true, so I thought I''d ask here, in case any one has seen this badhavior. ruby-1.9.2-p136 rspec (2.5.0) rspec-core (2.5.1) rspec-expectations (2.5.0) rspec-mocks (2.5.0) When I try to run this spec: require Pathname(__FILE__).ascend { |d|

I don't speak on behalf of any developers, but most times it is very difficult to change the license of an open source project, as the consent of all contributors would be needed which is nearly impossible. There are ways, but I don't think it is very likely. Further information about linking gpl is widely available on the net. Afaik it is always allowed to link to any library as long as

This is a copy of the supermin binding: https://github.com/libguestfs/supermin/tree/master/src realpath* --- mllib/unix_utils-c.c | 18 ++++++++++++++++++ mllib/unix_utils.ml | 4 ++++ mllib/unix_utils.mli | 5 +++++ 3 files changed, 27 insertions(+) diff --git a/mllib/unix_utils-c.c b/mllib/unix_utils-c.c index f5aaaf6..7e90541 100644 --- a/mllib/unix_utils-c.c +++ b/mllib/unix_utils-c.c @@

Sorry about the lack of info. So in the wireshark trace, our printer smbclient 4.0.7 sends a Unicode value in Russian of ???? with a password of ????123. The server rejects the login attempt with STATUS_LOGON_FAILURE. The second attempt is with the linux version smbclient 4.0.7, and the same credentials to the same server work. The only thing I see different in the requests are the flags sent: