Are you using an ACL enabled kernel ?
If not you need to either patch your existing kernel (acl.bestbits.at) or
you can use the XFS filesystem from SGI (oss.sgi.com). If you want to
reinstall the whole system then SGI actually produce an adapted RH installer
to run in conjunction with the usual
mm...
am I understanding
something really wrong or is there something that I've
missed.
samba-2.2.4
configure --with-acl-support outputs these lines along
others:
checking whether to
support ACLs... checking for acl_get_file in -lacl...
nochecking for ACL support... no
normal?
Title: ±ÍÇÏÀÇ ¸ÞÀÏÁÖ¼Ò´Â À¥¼ÇÎÁß
±ÍÇÏÀÇ
¸ÞÀÏÁÖ¼Ò´Â À¥¼ÇÎÁß, http://www..com/
¿¡¼ ¾Ë°Ô µÈ°ÍÀ̸ç, E-Mail ÁÖ¼Ò ¿Ü¿¡, ´Ù¸¥ Á¤º¸´Â °®°í ÀÖÁö ¾Ê½À´Ï´Ù.
¿øÄ¡ ¾ÊÀ¸¸é ¾Æ·¡ [¼ö½Å°ÅºÎ]¸¦ ´·
On Fri, 31 May 2002 [EMAIL PROTECTED] wrote:
> source/lib/interfaces.c conditionally include sys/time.h and
> sys/sockio.h (autoconf macros already exist)
This change broke SAMBA_2_2. The reason is that lib/interfaces.c
is used in the autoconf test so sys/so
The command from Win9x/ME client works fine to NT or Samba Servers.
The command with win2k client works fine when the server is NT 4.
The only work around appears is to use:
print /D:\\servername\printername filename
for a win2K client.
The printing problem wouldn't be so terrible if just nothin
Oops -- sorry... That changes compiles OK, but NOT if you rerun configure!!! Here's
config.log:
configure:10972: checking for iface AIX
configure:10989: gcc -o conftest -O -g
conftest.c -ldl 1>&5
In file included from configure:10986:
./lib/interfaces.c:34: config.h: No such file or directory
Michael Handler <[EMAIL PROTECTED]> writes:
> ok, here's the second revision of my patches to nmbd, smbd, and winbindd.
> to summarize:
and here's the patch that applies cleanly to HEAD. on the web at:
http://www.sub-rosa.com/handler/pub/samba-HEAD-daemontools-diff
--michael
Index: docs/docbo
Michael Handler <[EMAIL PROTECTED]> writes:
> ok, here's the second revision of my patches to nmbd, smbd, and winbindd.
> to summarize: [...]
that version patches fine against 2.2.4, but not against HEAD. i'm
generating a version of this patch against HEAD right now. --michael
On Tue, 4 Jun 2002, Terry Davis wrote:
> Am I completely wrong on this? Anyone have any ideas?
Get us a sniff, and we can look into it some more.
> On Mon, 2002-06-03 at 13:50, Terry Davis wrote:
> > Hello, I have my profiles being saved to a certain share. However, the
> > profiles are not
On Tue, 2002-06-04 at 17:57, Terry Davis wrote:
> yes, all of them except a few winxp machines...
Ok WinXP is like WinNT and has real profiles, win9x not!
Plese read smb.con manual page
logon home and logon path parameters.
ah and please do not file 3 different bugs with the same contents,
spec
yes, all of them except a few winxp machines...
On Tue, 2002-06-04 at 10:45, Simo Sorce wrote:
> Any win9x machine in the mix?
>
>
> On Tue, 2002-06-04 at 17:27, Terry Davis wrote:
> > Am I completely wrong on this? Anyone have any ideas?
> >
> >
> > On Mon, 2002-06-03 at 13:50, Terry Davis
A similar fix will be needed in lib/interfaces.c on the 3.0 branch...I just
checked the other files I changed and this is the only file that has this
issue.
PG
Richard Bollinger [mailto:[EMAIL PROTECTED]] wrote:
> Problem is that there are ifdef tests depending on the
> contents of config.h bef
Any win9x machine in the mix?
On Tue, 2002-06-04 at 17:27, Terry Davis wrote:
> Am I completely wrong on this? Anyone have any ideas?
>
>
> On Mon, 2002-06-03 at 13:50, Terry Davis wrote:
> > Hello, I have my profiles being saved to a certain share. However, the
> > profiles are not always g
Hi Andrew,
Here is the patch which only moves global_sam_sid to get_global_sam_sid().
And the patch for the Makefile.in, I see no (simple) way to change this
without changing the linking dependences.
metze
-
Stefan "
ok, here's the second revision of my patches to nmbd, smbd, and winbindd.
to summarize:
* the -F option makes each server program operate normally, but the
main process doesn't fork & disassociate.
* the -S option makes each server program log to stdout, unbuffered.
the -F option is designed
Mea culpa. Thanks, Rich.
PG
Richard Bollinger [mailto:[EMAIL PROTECTED]] writes:
> Problem is that there are ifdef tests depending on the
> contents of config.h before it's included.
> This patch moves up that include to the beginning:
Am I completely wrong on this? Anyone have any ideas?
On Mon, 2002-06-03 at 13:50, Terry Davis wrote:
> Hello, I have my profiles being saved to a certain share. However, the
> profiles are not always going to this share, they are sometimes being
> saved to the user's home directory. here is
Here's the error on Solaris
Compiling lib/interfaces.c
"lib/interfaces.c", line 112: undefined symbol: SIOCGIFCONF
"lib/interfaces.c", line 123: undefined symbol: SIOCGIFADDR
"lib/interfaces.c", line 130: undefined symbol: SIOCGIFFLAGS
"lib/interfaces.c", line 138: undefined symbol: SIOCGIFNETMAS
I made a patch some time ago that set up a new parameter, a range of uid
you cannot issue the delete user script command for, by default 0-100,
but that patch was not accepted.
Simo.
On Tue, 2002-06-04 at 15:56, Gerald (Jerry) Carter wrote:
> On Tue, 4 Jun 2002, Volker Lendecke wrote:
>
> > On
Title: º¹±ÇÆ÷Å» »çÀÌÆ®
º¹±ÇÆ÷ÅÐ »çÀÌÆ® : http://www.lottolia.co.kr
¡¤µðÁöÅÐȦµù½º ´ë±¸¡¤°æºÏÁö»ç
¡¤¹®ÀÇ:´ë±¸±¤¿ª½Ã ¼ö¼º±¸ ¹ü¾î2µ¿ 45-5¹øÁö ¹ü¾îŸ¿ö 503È£
¡¤»ç¾÷ÀÚµî·Ï¹øÈ£ : 515-03-54877 ¢Ï(053)745-1599, 1799
O º» ¸ÞÀÏÀº Á¤º¸Åë½Å¸Á ÀÌ¿ëÃËÁø ¹× Á¤º¸º¸È£ µî¿¡ °
On Tue, Jun 04, 2002 at 08:56:43AM -0500, Gerald (Jerry) Carter wrote:
> Volker is right on this one I think. We do need to prevent
> the removal of root and some other systems accounts for 2.2.5,
> but the parameter is going to stay in.
I'd vote for a very simple test for 'root'. Nothing else.
On Tue, 4 Jun 2002, Volker Lendecke wrote:
> On Tue, Jun 04, 2002 at 03:40:37PM +0200, Simo Sorce wrote:
> > Just make it a stub function and document it.
>
> I don't think this is a good idea. 2.2.5 should be the *LAST* 2.2
> series. Somebody will definitely complain if we remove it. Just call
On Tue, Jun 04, 2002 at 03:40:37PM +0200, Simo Sorce wrote:
> Just make it a stub function and document it.
I don't think this is a good idea. 2.2.5 should be the *LAST* 2.2 series.
Somebody will definitely complain if we remove it. Just call it dead code and
don't care for it anymore.
Volker
Just make it a stub function and document it.
People that do not know anything about it will not be affected (most),
people that know what it is will take workarounds if it is so necessary.
2.2.5 will be out for long before 3.0 will be integrated in new
distributions better keeping it safer.
On
Simo,
I'm not willing to remove this for 2.2.5 at this point. As much trouble
as it has caused, it is hard to determine how removing it will affect
existing installations.
Granted there are other issues that need to be addressed.
jerry
On 4 Jun 2002, Simo Sorce wrote:
> KILL IT
>
>
> /s
On Tue, 4 Jun 2002, Michel Stoop (samba) wrote:
> > I think we already fixed this in the latest SAMBA_2_2
> > cvs code. Please retest and let me know.
>
> I just updated my SAMBA_2_2 and the problem is still there.
> Look in configure.in > line 2730
> > WINBIND_LTARGETS=""
>
> So no winbin
KILL IT
/simo on a hurry
On Tue, 2002-06-04 at 13:58, Andrew Bartlett wrote:
> While I know we can't just kill the option, I was wondering if we should
> make testparm issue a load warning if people are using this option.
>
> What do people think? Or should we kill it or...? Ignorent admins
While I know we can't just kill the option, I was wondering if we should
make testparm issue a load warning if people are using this option.
What do people think? Or should we kill it or...? Ignorent admins have
causes havoc with this option in the past, and we have modified it in
HEAD (it just
28 matches
Mail list logo