Linux-Development-Sys Digest #221, Volume #6 Tue, 5 Jan 99 19:14:10 EST
Contents:
Re: Registry for Linux - Bad idea (Tristan Wibberley)
Re: Registry for Linux - Bad idea (Robert Krawitz)
Escape sequences and slang?? ([EMAIL PROTECTED])
Re: Kernel v2.2 (Piniek (aka Piotr Ingling))
Re: HOW-TO submit? (Gary Momarison)
Re: WDM Emulator, anyone? (Mark Roddy)
Linux 2.2 and current->comm (Gerd Rausch)
NFS Problem (root)
Re: Ethernet/Token Ring and Cabletron Switches ([EMAIL PROTECTED])
Canon LBP-660 Printer ("Kenneth Østerballe")
2.2.0pre4 lockups (Matt Foster)
Re: modules and unresolved symbols. (Edward Lee)
Re: Help with kernel error message... (Edward Lee)
----------------------------------------------------------------------------
From: Tristan Wibberley <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps,comp.os.linux.setup
Subject: Re: Registry for Linux - Bad idea
Date: Tue, 05 Jan 1999 19:30:02 +0000
Reply-To: [EMAIL PROTECTED]
Andrew Morton wrote:
>
> > It is *not* dated, unscalable nor unwieldy. Using tools such as
> > secure shell or rdist one can easily write a program to update
> > numerous machines automatically. It took me less than an hour to come
> > up with a perl script which updates all of our Linux servers
> > automatically. It does this using RSA encryption, and requires each
> > machine's root password to be typed in before it will execute. The
> > reason why these things are so easy to construct: they use PLAIN OLD
> > TEXT and standard I/O.
>
> - no provision for local overrides (inheritance)
I agree, this is awkward, but a well designed application makes this
easily possible (and it's not difficult to do it). One solution I've
seen is to have a number of files each inheriting from the next. Though
I agree, the key/value pair method does make this easier. I think there
is too much cruft on this argument for the discussion to continue in
this thread or in THREE newsgroups. Start a new thread (leaving out the
word registry - it's not) in comp.os.linux.development.apps and discuss
the requirements, then look at how the current system doesn't meet
those, and come up with ways to alter it (rather than scrap it).
> - apps must be restarted
A well designed application doesn't require this, it just needs to be
told via SIGUSR1 or such to reread it's config and continue gracefully.
You can do this with a simple tool that knows how to tell applications
to do this. (and no, doing it automatically as soon as a change is made
to the config is not good enough, the superuser must do it).
> - requires clients to be online
Any system that doesn't cache data requires this. Network mounted
filesystems can do this.
> - all the other things we've been saying.
>
> > As for it being expensive for "administration and training resources",
> > I have one word: linuxconf. I think you will be very surprised when
> > RedHat 6+Gnome+linuxconf hits the shelves.
>
> Linuxconf is an attempt to graft a uniform interface onto legacy chaos.
I think it's time to leave out end user administration helpers from the
argument, we all know that it's got nothing to do with it (If you need a
system more complex than the current one, you're not and end user :).
> What we're doing here is exploring the requirements and design of a
> _new_ approach which will be uniform from end-to-end. Please stop
> shouting at us - some good may come of all this.
I don't think George is trying to do that, just get a uniform interface
for the application programmer (As he keeps stressing the importance of
the application, and not the administration).
> A few random thoughts:
>
> - As one of our Microsoft friends pointed out in the halloween docs, the
> OSS community is good at following tail lights. We are now almost in
> the embarrassing position of being in the overtaking lane. It's time to
> look ahead and to innovate. This will require some evolution of the
> communication and management model.
Change for changes sake... I think you're going to ruin you own idea
thinking like that. The reason OSS plays catchup is because we don't
have to scrap stuff and replace it for marketing purposes. If you're
going to do this, do it because you have a specific problem that needs
solving - That's why Linux is good, and why Windows isn't.
> - The discussion here tends to leap ahead of itself: we're getting into
> fine grain implementation issues without having identified the
> requirements.
Exactly, specify what you need first. Exactly what you need, not
innovative crap, but just the things that will solve your problem. Then
stop and think what it limits, and decide if the limitations are worth
it, or go back and rethink to eliminate the limitations (I don't mean
what innovations does it limit, but what problems does it prevent
solutions too - exactly list problems). Then code a demo. Then you'll
have something that other people will like, use, and become standard (it
will happen only if you take the advice in this paragraph... I guarantee
it.
We're serious people in these newsgroups, we don't want to hear
'registry', 'active' or 'innovation'. We want to hear 'here's a problem,
and here's a solution', followed by 'here's a problem that your solution
may cause', etc.
The reason there's been such a big argument is that all that happened at
the start was that we got 'The current solution is not a complex system
covering every possibility, let's change it... I know, a registry - a
registry for registries sake'.
Now you have finally figured out problems for your solution to cater
with. But you need to go back and do it right. Then bring it here.
> Can we please take this from the top?
*VERY* good idea. I will join in and help.
> - ACAP http://www.ietf.org/html.charters/acap-charter.html has
> everything I want (I think) but it is heavyweight, in the sense that it
> is unsuitable for configuring many of the /etc/*.conf things.
These things can be configured just fine with the current method. It's
future component based applications you need to look at first. Then if
this method becomes widely used, add the system stuff to it.
> This problem can be resolved with a periodically invoked (or daemon)
> ACAP client which rebuilds /etc/*.conf based on the ACAP server and does
> all the necessary signalling. [ This requires that named, apache, etc
> can all seamlessly hold their current transactional state across a
> SIGHUP. This I doubt. ]
Then this is a bad solution, it needs to work without requiring
overhauls to currently working apps/utilities.
--
Tristan Wibberley Linux is a registered trademark
of Linus Torvalds.
------------------------------
From: Robert Krawitz <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps,comp.os.linux.setup
Subject: Re: Registry for Linux - Bad idea
Date: 05 Jan 1999 09:48:11 -0500
[EMAIL PROTECTED] writes:
>
> My experience with YP is limited, I do have experience with NIS+, but I have
> found that although NIS and friends are centralized, they are not really
> browseable. Niscat and nismatch are nice, but they dont give you a well
> definied hierarchy, or an intuitive organization. Personally I've never seen
> an NIS+ installation that does not leave the old text files around (in case
> of failure?) which further confused the problem.
For good reason. NIS+ is a real pain in the butt, at least for small
to medium sized installations. I can see how some of its stuff could
be of use in large installations.
> Why develop plugins if it can be changed and/or standardized so that
> configurations are all handled in a common way?
Fine, let's cut to the chase. How would inetd, sendmail (or whatever
MTA of your choice), hostname, and account information all be handled
in a common way? I'm asking for a reasonably concrete proposal
Making plugins is just
> layering more complexity onto the problem. What happens if a program changes
> but the plugin does not?
Fine, what happens if inetd is changed? For example, suppose someone
changes inetd to allow an optional specification of how long a
particular connection should be allowed to live before it's terminated
(let's not debate that particular one, it's an example of an extension
that could be logically made). How is linuxconf going to know about
it? How is it going to know how to logically present the change to
the user?
> > A central configuration database is neither necessary nor sufficient
> > to do this. Regedit on Windows (just as an example) is the ultimate
> > tool for manipulating the registry, but nobody claims that it makes it
> > any easier to use.
>
> Regedit does not offer human readable keys, I'm sure some people know what
> HKEY_LOCAL_USER\FOO\BAR means, but it's hardly more readable than an old .ini
> file.
Indeed, that's my point.
If it were MY_COMPUTER\NETWORK\DEVICES\PRINTERS and then regedit was
> able to spawn the apropriate configuration utility for each printer listed
> (if applicable) it would be easier to use.
Well, maybe, if you know where to look.
Windows is designed to deny that
> ability to users, Linux does not have to be. The central config database
> would not provide any ease-of-use in and of itself, but it provides a
> mechanism such that netscape plugins or linuxconf, or emacs or whatever can
> hook directly into the config information easily.
Ah! Now we're getting somewhere. Could you sketch out in a bit more
detail how this would work?
--
Robert Krawitz <[EMAIL PROTECTED]> http://www.tiac.net/users/rlk/
Tall Clubs International -- http://www.tall.org/ or 1-888-IM-TALL-2
Member of the League for Programming Freedom -- mail [EMAIL PROTECTED]
"Linux doesn't dictate how I work, I dictate how Linux works."
--Eric Crampton
------------------------------
From: [EMAIL PROTECTED]
Subject: Escape sequences and slang??
Date: Tue, 05 Jan 1999 21:45:32 GMT
Hi all,
I am developing a console mode application using the slang library and
need to send some specific lower level escape sequences to the console.
Specifically I need to send ESC]PHHHHHHH where the H's are hex digits.
This sequence is used to change the RGB value mapped to one of the 16
console colors. This string works fine if just sent from the console, but
does not work at all under slang.
EG: foopc:~\cat <cr>
^[]P7aaaa00
<Cntrl - C>
sets the lightgray color to be amber.
howether in my program sending this sequence using printf, before or after
initialising the library does not have the expected effect.
Anyone got any idea's?
Thanks Dan.
---
And on the evening of the first day the lord said......
... LX1, GO!, and there was light!
============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
From: [EMAIL PROTECTED] (Piniek (aka Piotr Ingling))
Subject: Re: Kernel v2.2
Date: Tue, 05 Jan 1999 22:06:32 GMT
Reply-To: [EMAIL PROTECTED]
Dnia Tue, 05 Jan 1999 12:12:43 -0800, Edward Lee <[EMAIL PROTECTED]>
napisa³(a):
>Nop, modutils is not the right stuff.
Have you installed it and it didn't help? Or is it your guess upon the name?
>Modules-X.X.X is a stealth package.
>It is still hidden somewhere out there.
>I am wondering why it is not included in the kernel sources.
>Without it, i can't build any modules at all.
You wrote earlier:
>I need the "-k" option of genksyms in the modules package.
Genksyms is part of modutils. I think it was called modules before.
Piotr Ingling
e-mail: [EMAIL PROTECTED]
------------------------------
From: Gary Momarison <[EMAIL PROTECTED]>
Subject: Re: HOW-TO submit?
Date: 05 Jan 1999 14:42:05 -0800
[EMAIL PROTECTED] writes:
> Hello people.
> I'm new one in the business of contributing something except bug reports.
[...]
> <quote>
> Thank you, if they are good new icons, they should probably made public,
> give them a free license(GPL), and should be put on web servers.
> However, I can't assist in doing so except answering questions. ;-)
> <end of quote>
>
> So could anybody enlight me what should I write in GPL license or better
> where to get the _correct_ form and where to put the icons. Which of the
First consider whether GPL is the license you really want to use.
It's usually called a "free license", but there are "free"er ones
you can use that allow more people to use your stuff. You probably
don't care if someone uses your stuff in a proprietary product.
You can find a bunch of licences using Gary's Encyclopedia at
http://www.aa.net/~swear/pedia/licencing.html [sic]
------------------------------
From: [EMAIL PROTECTED] (Mark Roddy)
Crossposted-To: comp.os.ms-windows.programmer.nt.kernel-mode
Subject: Re: WDM Emulator, anyone?
Date: Tue, 05 Jan 1999 15:14:53 GMT
On Mon, 04 Jan 1999 16:10:14 +0000, mlw <[EMAIL PROTECTED]> wrote:
>Mark Roddy wrote:
>>
>> I think it is both do-able and a good idea too, which is not the same
>> as saying I'm going to commit any of my time to it.
>>
>> This hurts microsoft right where they are vulnerable. They've defined
>> the kernel portability layer between windos98 and NT (WDM) so they
>> have to more or less stick to the definition so that the hardware
>> vendors can write drivers that actually work in both kernels. M$ wants
>> to get to one OS base so they are pushing HW vendors to use WDM.
>
>I am not saying this should be done to hurt Microsoft. Far from it, I
>want to do this to help Linux. These two motives are not the same thing.
>
You can't help linux without hurting Microsoft. Every linux install
(that sticks) is one less sale for BillG and crew. Nature of a
monopoly.
>>
>> However, you should be aware that currently only a very few windos98
>> drivers are 'WDM' drivers and no (none zero zed zilch) NT drivers are
>> WDM drivers (for now anyway.) So this is not a panacea. It won't, for
>> example, solve the winmodem or hp printer problems. (And AFAIC the
>> lack of support for these two pieces of hardware leaves linux doa on
>> the home desktop.)
>
>Actually, WDM starts with the NT Driver model. We would probably start
>with the NT DKK.
>
You've just increased the surface area of your project by a
considerable amount.
WDM is a portable subset of NT DDK , proven to be portable by the fact
that WDM runs within two completely different operating systems, NT5.0
and WinDos98. NT DDK is so far proven to be portable outside the NT
kernel nowhere and the dependencies are likely to suck in most of the
functionality of the NT kernel itself. Any change to NT potentially
destroys your work. As I said, WDM is the defined portability layer
for kernel drivers, not NT DDK.
I say stick with WDM. It limits the scope of your effort and leverages
microsoft's efforts to ease windos out of the picture.
Mark Roddy
_
------------------------------
From: Gerd Rausch <[EMAIL PROTECTED]>
Subject: Linux 2.2 and current->comm
Date: 06 Jan 1999 00:01:31 +0100
Hi,
trying to port a module to the new kernel, I've noticed
that overwriting comm by sprintf(current->comm, "xxx")
for a process started from within a module by kernel_thread
doesn't seem enough. "ps" still shows the "insmod ..."
command line. Is there any trick, to invalidate arguments,
so /proc/xx/cmdline will stay empty, or any other way making
"ps" display current->comm ?
Another problem I've discovered is connected to signals.
The goold old way to detect in a kernel function wether
a signal was received "if(current->signal & ~current->blocked)"
has been superceeded by "if(signal_pending(current))".
Anyway, whenever a process received, let's say a SIGINT in the
read(2) function, this signal was preserved even through the close(2)
function. So the close function could decide wether it should
finish pending buffers or abort as quickly as possible by
just reevaluating "if(current->signal & ~current->blocked)"
What's the new way of detecting such conditions ?
Any pointers or answers are really appreciated.
Regards,
Gerd
--
Gerd Rausch, Von Görschenstr. 7, 52066 Aachen/Germany
voice: +49-241-9019626, email: <[EMAIL PROTECTED]>
------------------------------
From: root <[EMAIL PROTECTED]>
Subject: NFS Problem
Date: Tue, 05 Jan 1999 23:53:17 +0100
Hi everybody,
I have just one little problem: I can't start rpc.knfsd.
The error I'm getting is:
nfssvc: Function not implemented
rpc.kmountd starts and registers its rpc services, same for quota and
stat.
Using RH5.2 + updates, Kernel 2.2.0-pre4, latest knfs-Package
knfs compiled into kernel or as a module doesn't change the behavior.
When using userspace-nfs, everything's ok. (except lockd)
Anyone here who knows what's missing?
regards,
René
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: comp.os.linux.networking,comp.os.linux.misc
Subject: Re: Ethernet/Token Ring and Cabletron Switches
Date: 5 Jan 1999 16:16:08 GMT
In comp.os.linux.networking Volker Dormeyer <[EMAIL PROTECTED]> wrote:
> Hi!
The problems occurs when BRIDGING disparate layer 2-protocols and using
MTU MTU that is incompatible.
You can correct it by 2 means :
Reduce MTU to the minimum of the two networks (1500 ?) This may cause some
performance reduction on the tokenring network.
or
replace the bridge with a router (who will fragnent those ip-packets that
traverse the intersection and is to big for the destination network).
Don't blame NT in this case, blame your network administrator for the set-up !
Peter h
> I have a serious problem in our switched Ethernet/Token Ring
> environment.
> I can´t ping (IP) from a Linux Box (Kernel 2.0.36) in the Token Ring
> segment
> to i. e a Windows NT Workstation in the Ethernet segment.
> Only when I reduce the MTU-size on the Linux Box to Ethernetsize (1500
> bytes)
> it works together with the NT-Workstations and some IBM AIX machines.
> I tried to set up a Linux-Box in the Ethernet segment; the result was
> that
> I was unable to ping a Workstation in the Token Ring Segment. The NT and
> AIX world have no problems.
> Here is some additional information about our equipment. The switches
> are from
> Cabletron Systems (Smart Switch 6000 & 9000). Ethernet and Token Ring
> are both
> the same B-Class logical IP-net.
> Some time ago I observed the same behaviour with a XyLan OmniSwitch.
> Any sugguestions/ideas?
> Thanks !
> -Volker-
> --
> #################################
> sender: Volker Dormeyer
> eMail: [EMAIL PROTECTED]
> [EMAIL PROTECTED]
--
--
Peter Håkanson Phone +46 0708 39 23 04
Network Management AB Fax +46 031 779 7844
Email : use peter (at) gbg (dot) netman (dot) se No copy to sanford wallace!
------------------------------
From: "Kenneth Østerballe" <[EMAIL PROTECTED]>
Subject: Canon LBP-660 Printer
Date: Tue, 05 Jan 1999 23:31:47 +0100
Hello all !
I'm a newbie to Linux, and I would be very happy if you could help me
get my LBP-660 to work under Linux. BTW I use Red Hat 5.2. Perhaps some
of you have made your own driver or something else that could be a help
to me and others with this "Microsoft Windows Printing System"-printer.
Please reply to me directly also - Thanks
Kenneth
------------------------------
From: Matt Foster <[EMAIL PROTECTED]>
Subject: 2.2.0pre4 lockups
Date: 5 Jan 1999 23:43:58 GMT
Hmm,
2.2.0 pre4, all the stuff "Changes" requires
PII (233M) 96Mb RAM, couple of IDE hard disks, couple of IDE CD-ROMS
(Using SCSI emulation as one of them is a re-writer). Not SMP compiled.
Trying to create a tar file (through bzip2) of a 1GB filesystem,
the system just seems to hang. Switching between virutal consoles
is OK, but that's it. It seems anything resembling real hard work
can lock it up as well. There is an NFS mount from a 2.0.36 box if
that is relevant.
Matt
------------------------------
From: Edward Lee <[EMAIL PROTECTED]>
Subject: Re: modules and unresolved symbols.
Date: Tue, 05 Jan 1999 08:28:44 -0800
I am having the same problem. We need a newer version of genksym in
modules!
Can someone tell me where to get it? I checked the followings:
sunsite.edu, tsx-11.org, kernel.org
without any luck.
James A Simmons wrote:
> What causes unresolved symbols in modules. Thank you.
>
> James Simmons
> Linux System Admin
> EdgeNet Inc.
> [EMAIL PROTECTED]
> http://www.edgeglobal.com/~jsimmons
------------------------------
From: Edward Lee <[EMAIL PROTECTED]>
Subject: Re: Help with kernel error message...
Date: Tue, 05 Jan 1999 08:31:03 -0800
See modules and unresolved symbol posting
Brian C. Thomas wrote:
> Hi all,
>
> I am running 2.2.0-pre4. In my syslog file, I am getting this error
> intermittantly, but persistantly...
>
> darkstar kernel: Symbol table has incorrect version number
>
> Can anyone explain this to me, and perhaps a way to get it fixed?
>
> Thanks,
>
> bct
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: [EMAIL PROTECTED]
You can send mail to the entire list (and comp.os.linux.development.system) via:
Internet: [EMAIL PROTECTED]
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Development-System Digest
******************************
