Mike,
Maybe I'm too used to reading cryptic man pages, but this one
seems pretty clear to me...
> The groupadd command lists a "-o" option.
>
> SYNOPSIS
>groupadd [-g gid [-o]] [-r] [-f] group
> [SNIP]
>-g gid The numerical value of the group's ID. This value
> m
"Mike A. Harris" <[EMAIL PROTECTED]> writes:
> The groupadd command lists a "-o" option. The manpage talks
> about this mythical option too. It says:
>
> SYNOPSIS
>groupadd [-g gid [-o]] [-r] [-f] group
> [SNIP]
>-g gid The numerical value of the group's ID. This value
>
How do I recursively SGID a directory tree, but only hitting the
dirs, not the files?
In other words, I want:
chmod -R g+s dir/
But no files SGID, just the subdirectories?
Can I use the "X" permission somehow? I never understood it..
Or should I do something like:
find dir/ -type d -exec ch
The groupadd command lists a "-o" option. The manpage talks
about this mythical option too. It says:
SYNOPSIS
groupadd [-g gid [-o]] [-r] [-f] group
[SNIP]
-g gid The numerical value of the group's ID. This value
must be unique, unless the -o option is used. The
On Wed, 23 Aug 2000, Frank Liu wrote:
>> so what the problem ? on the emu machine is not generated for i don't
>> reason, and with the possum machine it generated, and Frank can play
>> with that (actually with version.h who manage to play with kernel.h)
>
>kernel.h is useless because on a redhat
On Thu, 24 Aug 2000, John Summerfield wrote:
>> > [summer@emu summer]$ cat /boot/kernel.h
>> > cat: /boot/kernel.h: No such file or directory
>> > [summer@emu summer]$
>> > It's part of kernel-headers. If you've downloades source from ftp.kernel.or
>> g
>> > or a mirror, the file /boot/kernel.h
On Thu, 24 Aug 2000, Rodger Donaldson wrote:
>> Please reread this slowly three times: "Info pages are supposed to be
>> read with Emacs not with info".
>
>Perhaps some people don't believe that their choice of editor should be
>dictated by the documentation for their libc?
>
>Novel idea, I know.
On Wed, 23 Aug 2000, Frank Liu wrote:
>Since redhat kernel is a bit different from the stock kernel source.
>In my C source code, I need to do things differently depending on
>what kernel the user has. How can I #ifdef test that?
That isn't much to go on. What kind of code, what is it doing,
et
On Wed, 23 Aug 2000, Matt Fahrner wrote:
>Date: Wed, 23 Aug 2000 11:22:52 -0400
>From: Matt Fahrner <[EMAIL PROTECTED]>
>To: [EMAIL PROTECTED]
>Subject: Re: basename()?
>
>Lets kill this thread.
Agreed. It has nothing at all to do with basename() anymore,
which is a C function probably written
On Wed, 23 Aug 2000 [EMAIL PROTECTED] wrote:
>> Supposed to? Says who? So I'm _supposed to_ learn emacs, which
>> I have no use for, just so I can have a better UI to info
>> documents? Sorry, no thanks. info2html is the better UI for me.
>>
>
>You are not supposedc to learn Emacs. You are
On 23 Aug 2000, Chmouel Boudjnah wrote:
> John Summerfield <[EMAIL PROTECTED]> writes:
>
> > > > [summer@emu summer]$ cat /boot/kernel.h
> > > > cat: /boot/kernel.h: No such file or directory
>
> [...]
>
> > [summer@possum summer]$ ls -l /boot/kernel.h
> > -rw-r--r-- 1 root 237 Dec
> John Summerfield <[EMAIL PROTECTED]> writes:
>
> > > > [summer@emu summer]$ cat /boot/kernel.h
> > > > cat: /boot/kernel.h: No such file or directory
>
> [...]
>
> > [summer@possum summer]$ ls -l /boot/kernel.h
> > -rw-r--r-- 1 root 237 Dec 6 1999 /boot/kernel.h
>
> so what the
Frank Liu <[EMAIL PROTECTED]> writes:
> /boot/kernel.h is generated by /etc/rc.d/rc.sysinit at
> the boot time if it isn't already there (or CPU changed):
> check that file for the detailed logic on when/how to
> generate this file.
>
> So /boot/kernel.h may show you have a redhat system, but
>
/boot/kernel.h is generated by /etc/rc.d/rc.sysinit at
the boot time if it isn't already there (or CPU changed):
check that file for the detailed logic on when/how to
generate this file.
So /boot/kernel.h may show you have a redhat system, but
doesn't mean you have a redhat kernel. I have systems
John Summerfield <[EMAIL PROTECTED]> writes:
> > > [summer@emu summer]$ cat /boot/kernel.h
> > > cat: /boot/kernel.h: No such file or directory
[...]
> [summer@possum summer]$ ls -l /boot/kernel.h
> -rw-r--r-- 1 root 237 Dec 6 1999 /boot/kernel.h
so what the problem ? on the emu m
> John Summerfield <[EMAIL PROTECTED]> writes:
>
> > [summer@emu summer]$ cat /boot/kernel.h
> > cat: /boot/kernel.h: No such file or directory
> > [summer@emu summer]$
> > It's part of kernel-headers. If you've downloades source from ftp.kernel.or
> g
> > or a mirror, the file /boot/kernel.h m
John Summerfield <[EMAIL PROTECTED]> writes:
> [summer@emu summer]$ cat /boot/kernel.h
> cat: /boot/kernel.h: No such file or directory
> [summer@emu summer]$
> It's part of kernel-headers. If you've downloades source from ftp.kernel.org
> or a mirror, the file /boot/kernel.h may not exist, and
> Frank Liu <[EMAIL PROTECTED]> writes:
>
> > This is not runtime, it is compile time.
> > (even if it is runtime, how can uname() tell if you are running
> > a kernel provided by redhat or downloaded from kernel.org?).
> >
> > Anyway, the program calls function
> > kill_fasync()
> > under Redh
Frank Liu <[EMAIL PROTECTED]> writes:
> This is not runtime, it is compile time.
> (even if it is runtime, how can uname() tell if you are running
> a kernel provided by redhat or downloaded from kernel.org?).
>
> Anyway, the program calls function
> kill_fasync()
> under Redhat's kernel, it re
> > As I recall, not all are happy with 3.0.5. I found it wouldn't build an rpm
> on
> > my newly-installed RHL 6.2 (couldn't find the files) so reverted to 3.0.4
> > which built the same rpm without any problem.
>
> Did you test this with the latest (3.0.5-9, for example) RPM v3?
[summer@pos
On Wed, Aug 23, 2000 at 02:11:38AM +0200, [EMAIL PROTECTED] wrote:
> Please reread this slowly three times: "Info pages are supposed to be
> read with Emacs not with info".
Perhaps some people don't believe that their choice of editor should be
dictated by the documentation for their libc?
Nove
On Mon, Aug 21, 2000 at 01:01:37PM +0200, [EMAIL PROTECTED] wrote:
> Last but not least man pages are no longer mantained for most of the
> fundamental Linuwx utilities or libraries including gcc and glibc.
This might be a valid argument, except that the last time most info pages
provided by the
Hi there.
I've read newt's documentation, but can't find a way to have it output
coloured text. This is for a warning, which should be rendered (you
guessed it) in red.
Quite naively, I've tried using CSI sequences in the text, as the message
will appear on the console, but the sequence gets esc
> John Summerfield <[EMAIL PROTECTED]> writes:
>
> > > initialize it, it'll work. But there are some sparkles in some cases,
> >
> > sparkle.
>
> [...]
>
> > If this is fixable by configuration, what do I do?
>
> My problem may not be the same as your problem.
>
> My problem is that once in
> As I recall, not all are happy with 3.0.5. I found it wouldn't build an rpm on
> my newly-installed RHL 6.2 (couldn't find the files) so reverted to 3.0.4
> which built the same rpm without any problem.
Did you test this with the latest (3.0.5-9, for example) RPM v3?
Certain older versions h
> Matt Nelson <[EMAIL PROTECTED]> writes:
>
> > "a hack" is not how system documentation should be accessed. this is
> > ugly, and it is genuinely difficult to figure out how to do it. it would
> > be basically impossible for a new user (the ones who need documentation
> > the most).
>
> Aw, t
John Summerfield <[EMAIL PROTECTED]> writes:
> > initialize it, it'll work. But there are some sparkles in some cases,
>
> sparkle.
[...]
> If this is fixable by configuration, what do I do?
My problem may not be the same as your problem.
My problem is that once in a while (usually around t
> Jens-Ulrik Petersen <[EMAIL PROTECTED]> writes:
>
> > Ok, I give in... this seems to be one of the better kept secrets in
> > distribution land, though I know it has been discussed here before.
> >
> > How can I upgrade smoothly from rpm-3 to rpm-4 (without downloading an
> > iso cdrom image,
> initialize it, it'll work. But there are some sparkles in some cases,
sparkle.
Good word, describes a problem I have.
Two systems; first has SiS 5597/5598, I used the onboard video. Monitor is a
5yo Sampo Alphascan 15gx. We blamed the monitor;-) It's been doing something
else odd, on two
> > 2. Ever set up a printer? Ever do it successfully?
>
> Yes. Use printtol. Trivial. Needed to map my printers to supported
> models.
That's not so simple.
< As Linux gains market share manufacturers will have to
> provide drivers or see people go elsewhere. This day we will no
> longe
This is not runtime, it is compile time.
(even if it is runtime, how can uname() tell if you are running
a kernel provided by redhat or downloaded from kernel.org?).
Anyway, the program calls function
kill_fasync()
under Redhat's kernel, it requires three arguments
(see /usr/src/linux/include/l
I'd be interested to know what you're coding that would behave
differently depending on the running kernel.
You shouldn't #ifdef this, you should use runtime checks using the
uname() function.
Matt
On Wed, Aug 23, 2000 at 02:35:23PM -0500, Frank Liu wrote:
>
> Since redhat kernel is a
Since redhat kernel is a bit different from the stock kernel source.
In my C source code, I need to do things differently depending on
what kernel the user has. How can I #ifdef test that?
Thanks!
Frank
___
Redhat-devel-list mailing list
[EMAIL PROTE
Lets kill this thread. Both sides have been beat to death and things are
starting to get unfriendly out there.
Ugh.
Matt Fahrner
Burlington Coat Factory
begin:vcard
n:Fahrner;Matt
tel;pager:(603) 639-4142
tel;cell:(603) 381-3206
tel;fax:(603) 443-6190
tel;work:(603) 448-4100 xt 5150
x-mozilla-
Matt Nelson <[EMAIL PROTECTED]> writes:
> "a hack" is not how system documentation should be accessed. this is
> ugly, and it is genuinely difficult to figure out how to do it. it would
> be basically impossible for a new user (the ones who need documentation
> the most).
Aw, too bad. Complai
On 23 Aug 2000, Alan Shutko wrote:
> Matt Nelson <[EMAIL PROTECTED]> writes:
> > Btw, even using Emacs to read the Emacs Info page, I *still* wouldn't have
> > been able to figure out the above command.
>
> Not surprising, since the above invocation is a hack to replace the
> standalone version o
Matt Nelson <[EMAIL PROTECTED]> writes:
> Btw, even using Emacs to read the Emacs Info page, I *still* wouldn't have
> been able to figure out the above command.
Not surprising, since the above invocation is a hack to replace the
standalone version of info with info running in Emacs, without the
On 23 Aug 2000, Alan Shutko wrote:
> Matt Nelson <[EMAIL PROTECTED]> writes:
>
> > How do I read Info pages with Emacs?
>
> alias info="emacs -q -f info-standalone"
Now THAT'S what I call "intuitive".
Btw, even using Emacs to read the Emacs Info page, I *still* wouldn't have
been able to f
Matt Nelson <[EMAIL PROTECTED]> writes:
> How do I read Info pages with Emacs?
alias info="emacs -q -f info-standalone"
In X, it'll even start up in its own window. If you don't like that,
add a -nw after emacs in the alias above. I've checked, and you cna
do stuff like "info libc" with thi
John Summerfield <[EMAIL PROTECTED]> writes:
> There's nothing that says
> gcc [-g] [-O2] [-o name] [-c] filename.c
> will cover 90% of you needs, and you might want to look at -W , -l and -D
> with links, of course.
That's true. I'll bet that it's assumed that you are used to other
Unix
On Wed, 23 Aug 2000 [EMAIL PROTECTED] wrote:
> >
> > On Wed, 23 Aug 2000 [EMAIL PROTECTED] wrote:
> >
> > >> In other words, take a text doc, convert it to a single HTML
> > >> page, put some links in it there and there, a TOC, and INDEX, and
> > >> a bunch of anchors and you've got basically w
>
> On Wed, 23 Aug 2000 [EMAIL PROTECTED] wrote:
>
> >> In other words, take a text doc, convert it to a single HTML
> >> page, put some links in it there and there, a TOC, and INDEX, and
> >> a bunch of anchors and you've got basically what I'm
> >> looking for... I can use LYNX easily, but i
> On Mon, 21 Aug 2000, Nitebirdz wrote:
>
> >> in Windows (use KDE 2.0 exclusively (with konsole and friends disabled,
> >> of course), run autologin (http://www.linux-easy.com/development.php) to
> >> log in as root all the time [I know that's stupid, but it's basically what
> >> Windows does, a
>
>
> >* It's too hard to find stuff in it. Usually, those complaining are
> > looking for command-line flags, which are fairly buried in most info
> > documents. Those complaining are also usually unaware of info
> > features such as the easy-to-access index and the ability to search
> >
>
> Generally, there's a top-level item like "Invoking "
>
which in the particular case of gcc leads to great gobs of text and more links
and pretty much requires that one reads through heaps of options, commonly
used and uncommonly used.
There's nothing that says
gcc [-g] [-O2
> [EMAIL PROTECTED] ("Frank Schmuck, CFO" ) writes:
>
> > Is WinXX easier to use? Need some specifics?
> >
> > 1. Windows2000 auto-detects my BP6 dual processor system. All I have to
> > do to get the same level of service with Linux is to edit certain
> > setting, recompile the kernel, copy t
The debugger from 6.2 does threads without too many problems (unless
you're running on Alpha, in which case the 6.0 gdb works much better.)
gdb 5.0 does much much better handling threads, but there are a few
breakpoint reinsertion bugs that I've run into.
On Wed, Aug 23, 2000 at 08:31:27AM -0400
PLEASE! stop this thread!
yes, man is not the best, info is not the best, we need something better.
but we haven't and until somebody do that AND convert everything to that
we have to use our current tools.
amen.
-- Levente http://petition.eurolinux.org/
Jean Francois Martinez wrote:
> Please reread this slowly three times: "Info pages are supposed to be
> read with Emacs not with info".
>
> It has been told several times in this thread but you keep bashing
> info.
Is that to imply then that when they say use the "info" pages, not "man"
pages th
hi,
is anybody who use socks with kerberos authentication on redhat linux 6.2 ???
we're able to configure socks to said it can authenticate with kerberos ie.:
in /etc/socks5.conf
auth x.y.z.w - k
permit k - x.y.y.w - - - [EMAIL PROTECTED]
is not enough. we don't would like to use socks with "norm
On 23 Aug 2000, Trond Eivind[iso-8859-1] Glomsrød wrote:
> [EMAIL PROTECTED] (Sangeeta Huria ) writes:
>
> > I want to know how can I debug a multi-threaded application in Red Hat Linux
> > 6.2? Is there any debugger available through which i can debug multiple
> > threads. gdb allows only debu
On Tue, 22 Aug 2000, Harry Putnam wrote:
>Date: Tue, 22 Aug 2000 19:34:32 -0700
>From: Harry Putnam <[EMAIL PROTECTED]>
>To: [EMAIL PROTECTED]
>Subject: Re: basename()?
>
>On Tue, Aug 22, 2000 at 08:10:31PM -0400, Alan Shutko wrote:
>>
>> info --output - --subnodes 2>/dev/null libc |less
>
>Amaz
On Tue, 22 Aug 2000, Harry Putnam wrote:
>> A linear text document has a top and a bottom, and it scrolls
>> continuously down the screen as you go down, and up when you go
>> up. _that_ is what I want. Hypertext links will be fine if they
>> stay in this doc.
>
>Some one needs to put a gag on
On 22 Aug 2000, Alan Shutko wrote:
>Date: 22 Aug 2000 20:10:31 -0400
>From: Alan Shutko <[EMAIL PROTECTED]>
>To: [EMAIL PROTECTED]
>Subject: Re: basename()?
>
>"Mike A. Harris" <[EMAIL PROTECTED]> writes:
>
>> A linear text document has a top and a bottom, and it scrolls
>> continuously down the
On Wed, 23 Aug 2000 [EMAIL PROTECTED] wrote:
>> In other words, take a text doc, convert it to a single HTML
>> page, put some links in it there and there, a TOC, and INDEX, and
>> a bunch of anchors and you've got basically what I'm
>> looking for... I can use LYNX easily, but info doesn't cut
[EMAIL PROTECTED] (Sangeeta Huria ) writes:
> I want to know how can I debug a multi-threaded application in Red Hat Linux
> 6.2? Is there any debugger available through which i can debug multiple
> threads. gdb allows only debugging a single-threaded application.
No, it doesn't.
--
Trond Eivi
On 22 Aug 2000, Alan Shutko wrote:
>Date: 22 Aug 2000 19:56:17 -0400
>From: Alan Shutko <[EMAIL PROTECTED]>
>To: [EMAIL PROTECTED]
>Subject: Re: basename()?
>
>"Mike A. Harris" <[EMAIL PROTECTED]> writes:
>
>> Cool. I'll use printf() instead of basename then.
>
>Either you're intentionally twist
>From an upgrade point of view the biggest obstacles to upgrading from
6.x seem to be rpm-4.0, glibc-2.2pre (and XF86-4.0). How hard is it
to do the upgrade "by hand" ie using rpm?
Comments and advice appreciated.
Jens
___
Redhat-devel-list mailing
Hi All,
I want to know how can I debug a multi-threaded application in Red Hat Linux
6.2? Is there any debugger available through which i can debug multiple
threads. gdb allows only debugging a single-threaded application.
Or any pointers on how to debug multi-threaded application in absence of a
>
> On Tue, Aug 22, 2000 at 10:15:12AM +0200, [EMAIL PROTECTED] wrote:
> >
> > Don't remember but apparently the guy had benchmarked it. About
> > crashes, since both gcc 2.95.2 and gcc 2.96 are crashing (and gcc 2.95
> > built fine in 6.2) I would point to glibc. If it builds for you and
> >
60 matches
Mail list logo
