similar to: [klibc:master] fcntl: Fix struct flock for 32-bit architectures

Commit-ID: 3cb3ceea23ecbf2c804bf732e8234552d097b94f Gitweb: http://git.kernel.org/?p=libs/klibc/klibc.git;a=commit;h=3cb3ceea23ecbf2c804bf732e8234552d097b94f Author: Ben Hutchings <ben at decadent.org.uk> AuthorDate: Fri, 18 Jan 2019 18:39:21 +0000 Committer: Ben Hutchings <ben at decadent.org.uk> CommitDate: Fri, 18 Jan 2019 18:46:45 +0000 [klibc] fcntl: Fix file locking

Hi, I am trying to understand file locking and share modes. I am using RedHat Enterprise Linux 4, with kernel 2.6.9-55.ELsmp, samba-3.0.10-1.4E.11 and Windows XP. In particular I am looking at a machine running Linux exporting its local filesystem using Samba, with a Windows client accessing the file share. If I use byte range locking (fcntl() on Linux, LockFile() on Windows) things work as

Commit-ID: 1ec8e824a0b0ee10e93b257ccac08f07cd90f4d8 Gitweb: http://git.kernel.org/?p=libs/klibc/klibc.git;a=commit;h=1ec8e824a0b0ee10e93b257ccac08f07cd90f4d8 Author: maximilian attems <max at stro.at> AuthorDate: Fri, 29 Jun 2012 07:46:08 +0200 Committer: maximilian attems <max at stro.at> CommitDate: Fri, 29 Jun 2012 07:48:50 +0200 [klibc] include: [sys/types.h] bury

Implement as wrapper around sys_faccessat(). Latest dash started using it, also define AT_EACCESS: usr/dash/bltin/test.c: In function ?test_file_access?: usr/dash/bltin/test.c:490: error: too many arguments to function ?faccessat? Signed-off-by: maximilian attems <max at stro.at> --- usr/include/fcntl.h | 5 +++++ usr/include/unistd.h | 2 +- usr/klibc/Kbuild | 2 +-

The parts of klibc specific to the mips architecture. Signed-off-by: H. Peter Anvin <hpa at zytor.com> --- commit 8dc79563c06020d8844b9e9b821741828039b59e tree b957c8fb1fddf486f5c26b1880726051d4f6aaad parent bc9b363b31d301ab94c115cccc2e079c0d318498 author H. Peter Anvin <hpa at zytor.com> Sun, 25 Jun 2006 16:58:31 -0700 committer H. Peter Anvin <hpa at zytor.com> Sun, 25 Jun

Hi, FYI: we had glusterfs up and running with 2.0.8 and it worked fine. we decided to upgrade to 3.0.0 and found out, that with linux 2.6.31.7 and glusterfs 3.0.0 and the kernel fuse module, the flock() does not seem to work, i.e. apache2 php session files distributed via glusterfs 3.0.0 make apache2 hang on LOCK_EX we reverted back to 2.0.8 and installed glusterfs-fuse 2.7.4 and everything is

Hi list, as I know from the README in the source files OCFS2 does not support cluster aware flock. Our web developers have to use the php flock() function. There is a second method how php can handle locking - via the fnctl() function. Will this be supportet? I assume that OCFS2 supports the c-function open() mit O_CREATE as well as O_EXCL. Am I right? Regards, Stephan

Timo, The comments in dovecot-example.conf regarding lock_method for Solaris users needs to note your change. Your changelog says: - Default lock_method changed to flock instead of the old fcntl. Solaris users will need to set it back to fcntl. This makes sure that Dovecot's indexes aren't accidentally used with NFS. But there is no warning about this in the file for Solaris users.

SUSv3 is silent on that definition, but seen several userspace software that expects flock to be defined in that file and complain that it is missing in klibc. Current declaration is in unistd.h. Signed-off-by: maximilian attems <max at stro.at> --- usr/include/sys/file.h | 9 +++++++++ usr/include/unistd.h | 1 - 2 files changed, 9 insertions(+), 1 deletions(-) create mode 100644

Howdy! Am 6/30/20 um 2:37 PM schrieb Georg.Biberger--- via samba: > We are using Samba 4.10.16 as a readonly file server on a linux box with SLES 12. All is working fine with our NAS NFS Shares. We are now trying to integrate new NFS Shares from an archive provider. We have some problems with samba not responding, when accessing files on these archive NFS shares. > We have done some

> Howdy! > > Am 6/30/20 um 2:37 PM schrieb Georg.Biberger--- via samba: >> We are using Samba 4.10.16 as a readonly file server on a linux box with SLES 12. All is working fine with our NAS NFS Shares. We are now trying to integrate new NFS Shares from an archive provider. We have some problems with samba not responding, when accessing files on these archive NFS shares. >> We

On 01/07/2020 12:59, Georg.Biberger--- via samba wrote: >> The 'idmap config' lines are borked, the default domain '*' lines are >> okay, but the 'MUC' domain lines are half correct, yes you can use the >> 'rid' backend, but you must set a range. You did set a range, but it was >> incorrect and you have commented it out. The two ranges must

>Your user has the RID 1581344 and the 'rid' backend uses this along >with the low range to calculate the users Unix ID, so from the commented >line, this would be: >79846 + 1581344 = 1661190 >This is less than the high range, so would be valid. >But if you use '100001-500000000' for the range, the ID would be: >100001 + 1581344 = 1681345 >This would

On 06/07/2020 08:28, Georg.Biberger--- via samba wrote: > How can i achieve that the user qqeda11 is mapped to the the unix id 79846? > Background: All NFS files are only accessible by unix user qqeda11 with unix id 79846! > > Georg Where does '79846' come from ? If you run this: cat /etc/passwd | grep 'qqeda11' Do you get something like this:

Hello, When I run "cat /etc/passwd | grep 'qqeda11'", on my linux box, I get: qqeda11:x:79846:65600:Project Account for EDA:/home/qqeda11:/bin/bash When I run "wmic useraccount where (name='qqeda11' and domain='MUC') get name, sid" on my windows box, I get: Name SID qqeda11 S-1-5-21-43206524-2104247658-1151357142-1581344 so the user is an AD

On 06/07/2020 15:30, Georg.Biberger--- via samba wrote: > Hello, > > When I run "cat /etc/passwd | grep 'qqeda11'", on my linux box, I get: > qqeda11:x:79846:65600:Project Account for EDA:/home/qqeda11:/bin/bash That makes 'qqeda11' a LOCAL Unix user > > When I run "wmic useraccount where (name='qqeda11' and domain='MUC') get name,

Hi, I'm writing an application with C under Linux and I'm trying to use the flock system function to lock a file of an NT4 Server over Samba. The problem is that the lock works well within the same machine but if it is a process on another machine (the same program) the one that tries to access the locked file, the process is not blocked until the lock is released but given full access

On 7/6/20 1:09 PM, Rowland penny via samba wrote: > On 06/07/2020 18:33, Christopher Cox via samba wrote: >> >> IMHO, these discussion are beneficial.? "compat" traces back to NIS (mostly), >> so usually means a "combo" search of "+/-" style NIS or other allow/deny >> sources in addition to /etc/passwd (group, etc.). >> >> This

Hello, We are using Samba 4.10.16 as a readonly file server on a linux box with SLES 12. All is working fine with our NAS NFS Shares. We are now trying to integrate new NFS Shares from an archive provider. We have some problems with samba not responding, when accessing files on these archive NFS shares. We have done some Samba-Debugging and found out, that the call to flock in

Hi, as far as I understand the current situation with locking looks like this: - o2cb supports cluster-aware flock(2). fcntl(2) locks are only local. - Pacemaker and cman also support cluster-aware fcntl(2) - There is no way for a userspace program to check if fcntl(2) is cluster- aware or not Is this correct? We would like to stick with the o2cb cluster stack but there are applications like