Linux-Misc Digest #305, Volume #26 Tue, 14 Nov 00 02:13:02 EST
Contents:
Re: True GTK+ will eliminate Qt in next few years? (Kjetil Torgrim Homme)
Re: configuration for mainboard-incorporated video card (Dances With Crows)
Summary (Re: [finger] wont show .plan nor .project remotely -- why?) (John Bacalle)
Re: LinkSys betrayed us! Poor prospects for Linux. (Russell Petree)
Re: Looking for "clean disk" utility (Angry Bob)
Copy-Paste between KWrite and Netscape (Vivek Narayanamurthy)
Re: Memory leak? (Michael Erskine)
Re: Memory leak? (Michael Erskine)
Re: Memory leak? (Joe Schaefer)
Re: Memory leak? (Paul Kimoto)
UPDATE: PM Error 105 Please Look: Results of `fdisk -l /dev/hda` (Brando)
Re: paging for the sake of fs cache (Paul Kimoto)
Re: UPDATE: PM Error 105 Please Look: Results of `fdisk -l /dev/hda` (Brando)
Re: Memory leak? (Paul Kimoto)
Re: True GTK+ will eliminate Qt in next few years? (Darin Johnson)
Re: Multithreaded RPC Server Question ([EMAIL PROTECTED])
Re: TV Card (Valette =?iso-8859-1?Q?Jean=2DS=E9bastien?=)
----------------------------------------------------------------------------
From: Kjetil Torgrim Homme <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.advocacy,comp.os.linux.x,comp.unix.solaris
Subject: Re: True GTK+ will eliminate Qt in next few years?
Date: 14 Nov 2000 03:56:07 +0100
[Donovan Rebbechi]
> >* No standards for name mangling, leading to backwards compatibility
> > horrors.
>
> In theory, yes. In practice, we've also seen backward
> compatibility problems with libc.
Note the crossposting. I can't recall any such problems in Solaris,
we're still using binaries from 1991.
> >* Very hard to make bindings to languages other than C++.
>
> extern "C".
That easy, huh? How do you pass, say, a QStack?
> >The Unix ABI is based on C. Live with it.
>
> If this attitude prevailed, you wouldn't be using a computer in
> the first place.
The ABI must be based on the lowest common denominator. With the wide
differences between languages, I don't see an alternative to C which
will bring any significant advantages. Perhaps Fortran90.
Kjetil T.
------------------------------
From: [EMAIL PROTECTED] (Dances With Crows)
Subject: Re: configuration for mainboard-incorporated video card
Date: 14 Nov 2000 03:25:18 GMT
Reply-To: [EMAIL PROTECTED]
On Tue, 14 Nov 2000 01:44:36 GMT, [EMAIL PROTECTED] wrote:
>I am a newbie in linux. when i setted up linux, i could not configurate
>driver for my video card, so i cannot use WindowsX.
^^^^^^^^
It's "X" or "The X Windowing System".
>my video card was incorporated into mainboard, i failed to configurate
>even if i tried to select "Untitiled Card" and other "generic" items
>in "setup" procedure.
So boot to text mode and log in as root. Enter "cat /proc/pci | more"
at the prompt, and look for anything that resembles a video card
description. There *will* be one. Write it down. Then run your
distro's X setup utility. RedHat uses Xconfigurator, SuSE uses sax,
Debian and Slack use XF86Setup (which should work on all distros,
actually.) When you are asked for the video card name, enter the one
you found.
If you're using an Intel810 motherboard, there are some additional steps
you must follow. Search comp.os.linux.hardware for "setup intel 810"
and see what you find; I'm almost positive that someone has posted a
reasonable guide to getting that thing working.
In the future, when posting here with a question like this, please
include the details of your hardware and the distro you're using. That
info can help people help you much more quickly; if you'd said, "The
video card is a SiS6226 and the distro is RedHat 5.2", you could've
gotten specific advice instead of instructions to go digging around.
HTH, good luck....
--
Matt G|There is no Darkness in Eternity/But only Light too dim for us to see
Brainbench MVP for Linux Admin / Workin' in a code mine, hittin' Ctrl-Alt
http://www.brainbench.com / Workin' in a code mine, whoops!
=============================/ I hit a seg fault....
------------------------------
From: [EMAIL PROTECTED] (John Bacalle)
Crossposted-To: linux.redhat.misc
Subject: Summary (Re: [finger] wont show .plan nor .project remotely -- why?)
Reply-To: [EMAIL PROTECTED]
Date: Tue, 14 Nov 2000 04:01:34 GMT
Well it turned out that $HOME directory permissions were the problem. On
a RHL6.1 system ~ is mode 700:
drwx------ 43 john john 8192 Nov 13 22:51 /home/john
Adding the executable bit to 'Other' fixes matters (o=x).
drwx-----x 43 john john 8192 Nov 13 22:53 /home/john
Now .plan and .project are 'remote' finger accessible. Whoop! Gratitude
to David Efflandt, and all the kind folks who contributed.
John
--
John Bacalle Voice/Fax: +1-212-894-3778 x1057
I'm selling several new Cisco, MCSE, Red Hat books at a discount. My reef
aquarium and equipment as well: <http://www.unixen.org/sale-main.html>
MM> I'm calling on Secretary General Koffi Annan of the United Nations to
MM> send election observers, from Rwanda, Burundi, Colombia, and Jimmy
MM> Carter to Florida and oversee the vote results. =) --Michael Moore
------------------------------
From: Russell Petree <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.hardware,comp.os.linux.networking
Subject: Re: LinkSys betrayed us! Poor prospects for Linux.
Date: Tue, 14 Nov 2000 04:02:19 GMT
You have to download the driver, compile it, move it to you modules directory,
run depmod -a then load it using modprobe. Then it works fine.
Yea, I found it very annoying that they didn't provide better support. Heck
they didn't even host
the driver on their website, but it is a very good 10/100 card for under $15...
Cheers!
-rp
Les Mikesell wrote:
> <[EMAIL PROTECTED]> wrote in message
> news:6BUO5.2706$[EMAIL PROTECTED]...
> > In comp.os.linux.hardware Les Mikesell <[EMAIL PROTECTED]> wrote:
> > : Wasn't the Linux driver included on a floppy in the box with the
> > : card? Or, if you already have a dial-up connection working
> > : just grab the newest before you switch.
> >
> > For most newer revisions of the card, the linux driver wasn't new enough.
>
> You mean they included a floppy with a driver that didn't work? I
> avoided the problem because I stocked up on the DEC-chip version
> of the card when I found out they were changing. Do the new ones
> work as well after you get the driver installed? And do the latest
> Linux distributions (Mandrake 7.2, RH 7.0, etc.) include the correct
> version?
>
> Les Mikesell
> [EMAIL PROTECTED]
------------------------------
From: Angry Bob <[EMAIL PROTECTED]>
Crossposted-To: comp.security.unix
Subject: Re: Looking for "clean disk" utility
Date: 14 Nov 2000 04:01:14 GMT
What would you like to read? [comp.security.unix or *?]
This is a [EMAIL PROTECTED] scroll! it says:
> Can anyone direct me to a Linux utility like (Windoze) "Clean Disk";
> essentially a utility that writes (and re-writes) unused space in the
> filesystem?
http://packetstorm.securify.com/UNIX/secure_delete/
--
Angry Bob
I often imagine, with full visuals, what playing nethack
sober would be like.
--sdpcat (from rec.games.roguelike.nethack)
------------------------------
From: Vivek Narayanamurthy <[EMAIL PROTECTED]>
Subject: Copy-Paste between KWrite and Netscape
Date: Mon, 13 Nov 2000 22:08:03 -0600
Hi
I am not able to copy anything from KWrite to Netscape. But the other
way round, things work fine. Why is this happening? Is there anything
that can be done about this?
Thanks
Vivek
------------------------------
From: Michael Erskine <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.security,comp.os.linux.setup
Subject: Re: Memory leak?
Date: Mon, 13 Nov 2000 23:21:21 -0500
> If anyone has any ideas about what could be causing this (I did turn off
> rhnd --that's not it), I'd appreciate the insight.
>
> Thanks!
>
> -Greg
I have been using Linux for a long time but I have not upgraded to that
kernel and have never run SMP on a Linux box so take this with some salt.
Just some thoughts, in no particular order and of little worth.
Since the even numbered kernels are 'supposed' to be stable upon release,
and this is 2.4.x, I expect a problem this severe would have been noticed
in the pre-release kernels, which are odd numbered, eg... 2.1.x, 2.3.x.
Although, I am not certain that all the developers would be working with
SMP and I suppose there might be something unnoticed in a released kernel.
Still, there are a bunch of .edu's running SMP boxes in clusters. If it
were a problem in that area someone would have noticed most likely.
I have seen such a leak (though not as bad and not recently) in X on Sun
and HPUX. I doubt that you will see such a leak these days as I have not
noticed one in quite some time.
Make certain you are not misinterpreting what you are seeing. Write a
program that allocates memory until memory is completely exhausted
(including swap) and then kill the program with a SIGHUP. If you get
your memory back after that, the problem is most likely not a kernel
issue, but beware, you will have to reboot if you don't.
Look at swap activity, is the system busily swapping to disk... then
you probably have a memory problem or some program has allocated a bunch
of ram. How many users are hitting the server? What sort of use are
they putting it to? Is there a reason (lotsa apache processes, buncha
NFS/SAMBA activity, big Squid cache in memory, etc, etc...)?
IF this is a leak, use ps or top to chase down the leak. If the machine
is a server, stop running anything you don't need. You don't NEED X on
a server (at least not XDM or KDM). Just run startx when you want the
GUI and then log out of X when you aren't using it.
All else failing, I would be looking at turning off SMP for a while to
see if there is something else unnoticed in the kernel to this point.
Sorry, this ain't more help.
------------------------------
From: Michael Erskine <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.security,comp.os.linux.setup
Subject: Re: Memory leak?
Date: Mon, 13 Nov 2000 23:23:35 -0500
k wrote:
>
> Greg Engel <[EMAIL PROTECTED]> wrote:
> >My system is running the 2.4.0-0.37smp kernel from RedHat (same problem
> >with the shipping RedHat 2.4.0-0.26smp kernel, hence the upgrade) and
> >Redhat 7.0. The system is using autofs, nfs, and nis, and
> >unfortunately, it is acting as a server so I can't very well turn off
> >services for 72 h at a time. I'm also using tripwire extensively, and
> >I'm fairly certain that I haven't been hacked.
>
> have you tried using a _stable_ kernel?
>
Kernel versions which are even in the second digit, eg. 2.2.x, 2.4.x are
supposed to be stable at release.
Smart asses, are rarely ever given even numbers, or even even responses.
-m-
> --
> A great frustration in life is discovering that sometimes those who
> say something can't be done turn out to be right.
> -Donald Simanek
------------------------------
Crossposted-To: comp.os.linux.security,comp.os.linux.setup
Subject: Re: Memory leak?
From: Joe Schaefer <[EMAIL PROTECTED]>
Date: 13 Nov 2000 23:49:10 -0500
[EMAIL PROTECTED] (Andreas K�h�ri) writes:
> In article <[EMAIL PROTECTED]>, Greg Engel wrote:
> >Andreas K�h�ri wrote:
[...]
> >
> >top shows the following:
> >
> >> 2:56pm up 23:18, 5 users, load average: 0.10, 0.05, 0.15
> >> 77 processes: 76 sleeping, 1 running, 0 zombie, 0 stopped
> >> CPU0 states: 1.3% user, 1.2% system, 0.0% nice, 96.3% idle
> >> CPU1 states: 2.0% user, 0.4% system, 0.0% nice, 97.0% idle
> >> Mem: 255312K av, 227232K used, 28080K free, 0K shrd, 13920K
> >> buff
> >> Swap: 248996K av, 12K used, 248984K free 133448K
> >> cached
> >
[...]
> ps: Why is the shared memory at 0? It ought to be higher... Are you
> using a test/hacked kernel?
Sorry to barge in late, but the 0KB shared memory is bothering me, too.
I doubt it's a kernel problem, though. Perhaps your procps rpm
(rpm -qf `which ps`) isn't up to date with your kernel release?
I don't know much about your particular situation, but whenever you
upgrade your kernel in a *major* way (i.e. 2.2 -> 2.4), there's a
whole bunch of accompanying utilities that have to get upgraded
also. The kernel source is well-documented, and you can find
out the minimum requirements there if other stuff is acting wierd.
HTH.
--
Joe Schaefer
------------------------------
From: [EMAIL PROTECTED] (Paul Kimoto)
Crossposted-To: comp.os.linux.security,comp.os.linux.setup
Subject: Re: Memory leak?
Date: 14 Nov 2000 00:10:45 -0500
Reply-To: [EMAIL PROTECTED]
In article <[EMAIL PROTECTED]>, Joe Schaefer wrote:
[2.4.0-test* kernels:]
>> ps: Why is the shared memory at 0? It ought to be higher... Are you
>> using a test/hacked kernel?
> Sorry to barge in late, but the 0KB shared memory is bothering me, too.
The procps programs just get the information from /proc/meminfo. As I
understand it, the kernel no longer adds up the shared memory for your
convenience, and just reports "0" instead. (Apparently it's considered too
expensive or something like that.)
--
Paul Kimoto
This message was originally posted on Usenet in plain text. Any images,
hyperlinks, or the like shown here have been added without my consent,
and may be a violation of international copyright law.
------------------------------
From: Brando <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux,comp.os.linux.help,comp.os.linux.questions,comp.os.linux.setup
Subject: UPDATE: PM Error 105 Please Look: Results of `fdisk -l /dev/hda`
Date: Tue, 14 Nov 2000 05:42:37 GMT
Eric asked me to post the result of `fdisk -l /dev/hda`
So, here it is:
> Disk /dev/hda: 240 heads, 63 sectors, 776 cylinders
> Units = cylinders of 15120 * 512 bytes
>
> Device Boot Start End Blocks Id System
> /dev/hda2 * 39 775 5571216 b Win95
>FAT32
I've deleted the Linux (non-dos) partition. All that is left is Win98. It boots just
fine
into Win98.
So, what should I do? PM still errors out as before.
Brando
------------------------------
From: [EMAIL PROTECTED] (Paul Kimoto)
Subject: Re: paging for the sake of fs cache
Date: 14 Nov 2000 01:18:55 -0500
Reply-To: [EMAIL PROTECTED]
In article <[EMAIL PROTECTED]>, Olaf Zaplinski wrote:
> my 2.2.17 SuSE 7.0 machine likes paging much - it puts ~20 MB into the swap
> space just to keep fs buffers in RAM. How do I turn this off?
I don't think that you do, but you could consult
/usr/src/linux/Documentation/sysctl/vm.txt anyway.
> The old 2.0.x
> kernels did not have this counterproductive behaviour. Well, for servers
> this might be okay, but how do I tune a 2.2.x kernel for workstation use?
Do you really think that the performance is worse? I find that 2.2.* is
better than 2.0.* for workstation-type use _because_ it buffers more of
the filesystem. Yes, it pages more aggressively, but these are mostly
pages that are hardly ever needed and _should_ be swapped out. How much
RAM do you have, and how much memory do you typically use (after
subtracting out the buffers and cache)?
--
Paul Kimoto
This message was originally posted on Usenet in plain text. Any images,
hyperlinks, or the like shown here have been added without my consent,
and may be a violation of international copyright law.
------------------------------
From: Brando <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux,comp.os.linux.help,comp.os.linux.questions,comp.os.linux.setup
Subject: Re: UPDATE: PM Error 105 Please Look: Results of `fdisk -l /dev/hda`
Date: Tue, 14 Nov 2000 06:19:28 GMT
On Tue, 14 Nov 2000 05:42:37 GMT, Brando <[EMAIL PROTECTED]> wrote:
CORRECTION: This is what it looks like
>> Disk /dev/hda: 240 heads, 63 sectors, 776 cylinders
>> Units = cylinders of 15120 * 512 bytes
>>
>> Device Boot Start End Blocks Id System
>> /dev/hda2 * 39 775 5564160 c Win95
>FAT32 (LBA)
>> Partition 2 does not end on cylinder boundary:
>> phys=(774, 15, 63) should be (774, 239, 63)
Brando
------------------------------
From: [EMAIL PROTECTED] (Paul Kimoto)
Crossposted-To: comp.os.linux.security,comp.os.linux.setup
Subject: Re: Memory leak?
Date: 14 Nov 2000 01:22:09 -0500
Reply-To: [EMAIL PROTECTED]
In article <[EMAIL PROTECTED]>, Michael Erskine wrote:
> k wrote:
>> Greg Engel <[EMAIL PROTECTED]> wrote:
>>> My system is running the 2.4.0-0.37smp kernel from RedHat
>> have you tried using a _stable_ kernel?
> Kernel versions which are even in the second digit, eg. 2.2.x, 2.4.x are
> supposed to be stable at release.
>
> Smart asses, are rarely ever given even numbers, or even even responses.
That is nice, but the 2.4.0 kernel has not been released (see
ftp.us.kernel.org/pub/linux/kernel/v2.4). The latest has the awkward name
"2.4.0-test11/pre4". Run "finger @ftp.kernel.org" for more recent updates.
--
Paul Kimoto
This message was originally posted on Usenet in plain text. Any images,
hyperlinks, or the like shown here have been added without my consent,
and may be a violation of international copyright law.
------------------------------
Crossposted-To: comp.os.linux.advocacy,comp.os.linux.x,comp.unix.solaris
Subject: Re: True GTK+ will eliminate Qt in next few years?
From: Darin Johnson <[EMAIL PROTECTED]>
Date: Tue, 14 Nov 2000 06:35:58 GMT
mlw <[EMAIL PROTECTED]> writes:
> > * Very hard to make bindings to languages other than C++.
> Not true.
>
> extern "C" function(....)
True, but that may not be the point. Ie, if the interface is in C,
then it doesn't matter if the actual implementation is in C++ or
Fortran or something else. When people say "I wish this library was
in C++" they usually mean they want the library interface is in C++,
not implemented in C++ with an `extern "C"' interface.
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: Multithreaded RPC Server Question
Date: Tue, 14 Nov 2000 06:30:10 GMT
Thanks for your help. I am back to this project after a brief
vacation. In text below:
> What are you aiming at? You mean to find a fully threaded
> rpc-portmapper?
Portmapper is not my major concern, but with the kind of tests I want
to run it could end up being a bottleneck.
> I haven't seen any on FreeBSD or Linux yet. Besides a few hundred
> requests are no problem.
> A portmapper does'nt do much. It just negotiates a free port between
the
> client and "server". Now the client can send RPC requests to be runned
> from the server.
> Personally I hate RPC's I also don't know what your application does.
Basically I need to do a little performance and scalability comparison
of major RPC implementations. It is being done on multi-proc machines,
dualneted with several clients on each network. The staple test is
just making a simple call that does some record-keeping. Sun RPC does
very well while it uses less than 100% of a processor (about 4 client
threads placing RPCs over a 100Mbs link), but performs poorly after
that. I was just wondering of the best way to write a simple test that
would take advantage of multiple CPUs on the server without doing load
ballancing myself (which would defeat the purpose of the tests).
Thanks in advance for any help.
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: Valette =?iso-8859-1?Q?Jean=2DS=E9bastien?=
Crossposted-To:
linux.redhat.install,linux.redhat,linux.redhat.misc,comp.os.linux.x,comp.os.linux.hardware
Subject: Re: TV Card
Date: Tue, 14 Nov 2000 07:19:18 +0100
This is a multi-part message in MIME format.
==============4C38187A16064B1892ED63A1
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Have a look in the CARDLIST file in the
/usr/src/linux/Documentation/vid*/bttv dir.
It may cotain the name of ure card asssociated with a number. the do a
modprobe bttv CARD=X .
Also look for the tuner and sound modules.
"Ernest N. Mamikonyan" wrote:
>
> My friend have and Avermedia BTA848 TV Card and we were trying to get
> it to work under RedHat 7.0 using the bttv.o module. However, we hav
> enot been successfull yet, although very close. We either get a blank
> screen and sound or the one channel and no sound. Does anyone have a
> clue as to what we are doing wrong (or not doing)? By the way, we were
> alble to get all the modules to load without errors.
>
> --
> Thanks in advance!
> Ernie ([EMAIL PROTECTED])
==============4C38187A16064B1892ED63A1
Content-Type: text/x-vcard; charset=us-ascii;
name="jean-sebastien.valette.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for Valette Jean-S�bastien
Content-Disposition: attachment;
filename="jean-sebastien.valette.vcf"
begin:vcard
n:Valette;Jean-S�bastien
x-mozilla-html:FALSE
org:CNMSS;83
adr:;;85 all�es de cedres;Six-Fours;VAR;83140;France
version:2.1
email;internet:[EMAIL PROTECTED]
title:Homme R�seaux
fn:Jean-S�bastien Valette
end:vcard
==============4C38187A16064B1892ED63A1==
------------------------------
** 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.misc) 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-Misc Digest
******************************
