Linux-Development-Sys Digest #14, Volume #7 Mon, 2 Aug 99 03:14:30 EDT
Contents:
Re: Linux Journal - worth or not? (John Stracke)
Re: Linux Journal - worth or not? (Tor Arntsen)
Looking for answer on ext2fs inode mode flag question. ("Elliot Spencer")
linux porting question (i386->sparc) (Andre Couture)
Re: NTFS Status? ("Igor Zlatkovic")
File table overflow problems (Geoff Wilson)
Right all you people - my freind Joanne need to raise �1950 for mencap by November
this year - anyone going to donate? ("Matt Newey")
/proc & lm_sensor kernel Oops... feature or bug? :-) (Paolo Pedaletti)
PAM/g++ (lx)
2.0.37 cold start while booting ([EMAIL PROTECTED])
What's with this new Gnome system? (Ron House)
Re: /proc & lm_sensor kernel Oops... feature or bug? :-) (Alexander Viro)
Ramdisked X server distro (Chetan Ahuja)
/dev/hdb1 reached max mount count and the system hangs. How can I fix it. Thanks. JG
(Joseph Girona)
Re: NTFS Status? ([EMAIL PROTECTED])
Re: /dev/hdb1 reached max mount count and the system hangs. How can I (Tristan
Wibberley)
how does "top" work (Yung-Hsiang Lu)
----------------------------------------------------------------------------
From: John Stracke <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps
Subject: Re: Linux Journal - worth or not?
Date: Sun, 01 Aug 1999 13:22:39 GMT
"Christopher B. Browne" wrote:
> By going onto the newsstand, that forces the "advertising pressures"
> to a head, because it reduces the proportion of the revenues from
> sales that the magazine gets.
>
> After all, someone's paying for the costs of moving magazine from
> Seattle to your local newsstand, as well as the profits taken by both
> the wholesale distributor and the newsstand...
That's why they charge less for subscriptions (that and the predictability
factor). I would guess that you can't really say whether an individual
magazine gets more or less money for newsstand sales without examining their
books.
--
/==============================================================\
|John Stracke |http://www.thibault.org| S/MIME, HTML OK |
|Francois Thibault |=========================================|
|Bhakail, East | "It takes a special kind of guy to rig |
|[EMAIL PROTECTED]| an election and then lose."--Cecil Adams|
\==============================================================/
------------------------------
From: [EMAIL PROTECTED] (Tor Arntsen)
Subject: Re: Linux Journal - worth or not?
Crossposted-To: comp.os.linux.development.apps
Date: Sun, 01 Aug 1999 13:47:16 GMT
In article <[EMAIL PROTECTED]>,
"Steven J. Hill" <[EMAIL PROTECTED]> writes:
>I would tend to agree with Mr. Rempt's initial opinion that the Linux
>Journal is becoming more commercially oriented. I just recently started
I too tend to agree with that, and it worries me. I've been a subscriber
since #1 and I've liked it all the way, but recently I haven't found
much of interest. Unfortunately all good magazines turn bad after a
few years. Just look at what happened to Byte, the once great magazine
which used to have a lot about programming techniques etc. Same with
Personal Computer World (UK), one of the very first computer magazines.
It too started out great, then went downhill like the others. Linux Journal
is the only so-called "computer magazine" I still read, it's still ok but
I don't like the trend I'm seeing.
>to subscribe to Linux Magazine which IMHO is has good technical content
>and is not so full of advertisements. I have the first three issues and
>have read them all cover to cover. Here is the URL:
>
> http://www.linux-mag.com/
Hm, I'll check it out I think..
-Tor
------------------------------
From: "Elliot Spencer" <[EMAIL PROTECTED]>
Subject: Looking for answer on ext2fs inode mode flag question.
Date: Sun, 1 Aug 1999 15:01:17 +0100
Hi, I'm currently putting together some filesystem utilities for Linux
Ext2fs.
I have a question on the inode mode flag values. I've come across some
inodes with a directory or file name spread across the bytes usually
reserved for the data block pointers and indirectors. The mode flag is
nearly always set to x'A1FF' for these inodes. Does anyone know if these are
indicative of symbolic links ? According to the documentation I have they
should be Sockets ? What are sockets in this context ?
Any help or any pointers to interesting articles on ext2fs much appreciated.
This is not a commercial project.
--
Regards, Elliot Spencer
============================================================================
Email address is [EMAIL PROTECTED]
(remove _nospam to mail me)
Home page http://www.spnc.demon.co.uk
(covers VB, REXX, 80x86 Asm, SQL and Oracle).
============================================================================
------------------------------
From: Andre Couture <[EMAIL PROTECTED]>
Subject: linux porting question (i386->sparc)
Date: Sun, 01 Aug 1999 17:35:41 +0200
Hi all,
I'm having problems porting a Linux application from i386 to sparc (both
RH 6.0)
If anybody has any experience and/or pointers please send me a email
My problems seem to be related to byte/word alignment and possibly
packing.
Thanks
Andre
[EMAIL PROTECTED]
------------------------------
From: "Igor Zlatkovic" <[EMAIL PROTECTED]>
Subject: Re: NTFS Status?
Date: Sun, 1 Aug 1999 16:26:44 +0200
Hi there.
You can compile in the write support for ntfs with no harm. When mounting
ntfs volumes, use the mount option to mount them read-only. This way, you
will not be harmed by the experimental write support, even if present in the
driver.
Writing to ntfs volumes works well, as long you don't ask too much. As long
you want to write few files, everything will work fine. Don't expect much
when using software raids, for example.
I didn't have bad expiriences with write support. Writing a file or two,
like editing a configuration file and writing the altered version back to
disk always went good.
Once I tried to copy the whole directory structure from ext2fs to ntfs which
contained about 20000 files. The ntfs driver crashed after about 30 seconds.
Dont let this be a measure, because I have asked for it. It was a
multiprocessor machine and the target drive was a compressed ntfs5 (win 2000
beta 3) volume. Even then, no filesystem corruption of the ntfs5 volume
occured.
Shortly, ntfs basically works. Be careful, but you can use it.
Ciao
Igor
David Isaac Stclair <[EMAIL PROTECTED]> wrote in message
news:7nnch5$gr5$[EMAIL PROTECTED]...
> Anyone know the current status of the NTFS on linux? I'm trying to help
> out a friend using NT, Redhat and Suse on the same machine.
> I got read access working using kernel 2.2.10
>
> I did not compile write access because it looked dangerous. Anybody got
> any recommendations?
>
> --
> _______________________________
> David I. St.Clair
> North Carolina State University
> [EMAIL PROTECTED]
>
------------------------------
From: Geoff Wilson <[EMAIL PROTECTED]>
Subject: File table overflow problems
Date: 01 Aug 1999 15:16:58 GMT
For some reason I keep getting file table overflow problems. Is it possible to
increase the size of the file table (kernel 2.0.36) and if so will that
actually help with the problem. This is for a news server that is dealing with
32 odd 1Gb files and a few other large files too (INN 2.2 with CNFS). Any help
with this would be greatly appreciated,
Geoff
---
a paranoid is just someone with all of the facts at his disposal
- william seaward burroughs 1914-1997
Geoff Wilson <[EMAIL PROTECTED]> http://www.cs.mu.oz.au/~gmwils/
------------------------------
From: "Matt Newey" <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux.development.apps,comp.society.development,comp.software.measurement,comp.sys.mentor,corelsupport.draw8-printing_scanning_color-management,cz.soc.mensa,de.org.mensa,de.soc.menschenrechte,dk.velkommen
Subject: Right all you people - my freind Joanne need to raise �1950 for mencap by
November this year - anyone going to donate?
Date: Sun, 1 Aug 1999 17:58:43 +0100
Please help Mencap???
------------------------------
From: [EMAIL PROTECTED] (Paolo Pedaletti)
Subject: /proc & lm_sensor kernel Oops... feature or bug? :-)
Date: 1 Aug 1999 23:57:00 GMT
Reply-To: [EMAIL PROTECTED]
Ciao,
I was playing with lm_sensors modules on VT1 , when I have un-load w83* module.
On VT 3 I was in /proc/sys/dev/sensors/w83781d-isa-0290 and here's what has
happened:
tty3.niels:Fri Jul 30 - 14:35:31:root
/proc/sys/dev/sensors/w83781d-isa-0290$ cd
Unable to handle kernel NULL pointer dereference at virtual address 00000030
current->tss.cr3 = 07095000, %cr3 = 07095000
*pde = 00000000
Oops: 0000
CPU: 0
EIP: 0010:[<c805e3b7>]
EFLAGS: 00010246
eax: c805e370 ebx: 00000028 ecx: c805f510 edx: 00000000
esi: c805f520 edi: c805f4c0 ebp: c6861ed0 esp: c7097f74
ds: 0018 es: 0018 ss: 0018
Process bash (pid: 8718, process nr: 10, stackpage=c7097000)
Stack: 00000000 bffffc58 c0145be3 c6861ed0 00000000 c01d65c0 c0130d38 c6861ed0
c1948500 c6861ed0 ffffffff c012f70a c6861ed0 00000000 c1948500 c0123dbe
c1948500 c7096000 00000000 c0109c94 080a844c 080a844c 00000000 00000000
Call Trace: [<c0145be3>] [<c0130d38>] [<c012f70a>] [<c0123dbe>] [<c0109c94>]
Code: 8b 42 30 52 8b 40 38 ff d0 83 c4 04 5b 5e 5f 5d c3 90 8d b4
niels login: Unable to handle kernel NULL pointer dereference at virtual address
00000030
current->tss.cr3 = 028fb000, %cr3 = 028fb000
*pde = 00000000
Oops: 0000
CPU: 0
EIP: 0010:[<c805e3ae>]
EFLAGS: 00010202
eax: c805e370 ebx: 00000028 ecx: c805f510 edx: 00000000
esi: c805f520 edi: c805f4c0 ebp: c65a1430 esp: c2225ec8
ds: 0018 es: 0018 ss: 0018
Process cat (pid: 22137, process nr: 33, stackpage=c2225000)
Stack: c000a400 c45a24e0 c0145f35 c65a1430 00000001 c425e460 c6664990 c425e4b4
c0146523 c000a400 000010ec c425e460 fffffff4 c45a24e0 c6664990 c2225f54
c6664990 ffffffea 00000000 c012b236 c6664990 c45a24e0 c2225f54 00000000
Call Trace: [<c0145f35>] [<c0146523>] [<c012b236>] [<c012b410>] [<c012b58e>]
[<c012431c>] [<c01244b2>]
[<c0109c94>]
Code: 8b 42 30 52 8b 40 34 eb 07 8b 42 30 52 8b 40 38 ff d0 83 c4
niels login:
===============================
It'a a bug or a feature? ;-)
--
Paolo Pedaletti, Como, ITALYa < paolo . pedaletti @ flashnet . it >
------------------------------
From: lx <[EMAIL PROTECTED]>
Crossposted-To: comp.unix.programmer,linux.redhat.pam
Subject: PAM/g++
Date: 2 Aug 1999 02:28:18 GMT
I am in sincere need of a little bit of assistance with PAM for
an application that I currently putting together for the UNIX
platform. If there are any experienced UNIX programmers out there
who have any idea of how PAM works, I would really, really be in
your debt if you could e-mail me and let me know so I can ask you
a few basic questions. I'm finding the documentation on the exact
procedures of configuring PAM to be quite lacking, and I just need
to have an idea where my problems are lying.
ANY assistance will be appreciated duly, I'm at my wit's end.
Thanks so much in advance,
.lx
--
[ [EMAIL PROTECTED] -- http://www.audiophonic.com/lx/ ]
[ "in a mirror, all the time turns counter-clockwise." ]
[ "and every time i look, i find my face has changed." ]
------------------------------
From: [EMAIL PROTECTED]
Subject: 2.0.37 cold start while booting
Date: 2 Aug 1999 04:39:10 GMT
Greetings,
I just updated by kernel to 2.0.37, recompiled, did the whole LILO bit,
and now my computer reboots while it is printing all the '...' stuff when
it is loading the kernel. Fortunately, I made a boot floppy of 2.0.36 so
nothing is hosed. I'm wondering if anyone else is experiencing this sort
of thing or is it just me?
Thanks,
Phil
------------------------------
From: Ron House <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps
Subject: What's with this new Gnome system?
Date: Mon, 02 Aug 1999 04:56:53 +0000
I have been trying this new gnome system in RedHat 6. The concept is
nice but it seems to have problems. It starts up the system inside X
instead of being run from the command line, and so the windows opened in
X are the first (i.e. logon windows) you see, but it doesn't run
.bash_profile. Then the gnome-terminal steadfastly ignores the geometry
parameter and gets the colours wrong (try IndianRed and tell me if you
think a pallid pink is the rich dark colour every non-gnome app uses for
Indian Red). The gnome colour browser displays colours that are
different again. Then when you 'save' the screen it saves the values of
export variables but not aliases. If you try copying a menu item into
the user menus with right-click "add this to personal menu" it sometimes
will and sometimes refuses to be copied. Then the menu editor doesn't
have an option to clone an existing menu item (for small mods which are
easier than setting up the whole thing again). I have to think this
system is in a very raw state. It would seem better to keep the old
system in public distributions until the new one is ready.
--
Ron House [EMAIL PROTECTED]
The evils of each age always seem self-evidently right at the time.
------------------------------
From: [EMAIL PROTECTED] (Alexander Viro)
Subject: Re: /proc & lm_sensor kernel Oops... feature or bug? :-)
Date: 2 Aug 1999 01:10:56 -0400
In article <[EMAIL PROTECTED]>,
Paolo Pedaletti <[EMAIL PROTECTED]> wrote:
>Ciao,
>I was playing with lm_sensors modules on VT1 , when I have un-load w83* module.
>On VT 3 I was in /proc/sys/dev/sensors/w83781d-isa-0290 and here's what has
>happened:
[unloading the module freed CWD of process, with obvious results + added quite
a bit of shit later]
>It'a a bug or a feature? ;-)
Bug, indeed, but you've omitted very interesting piece of data: version of the
kerenl ;-) Results of ksymoops on your kern.log would be also useful - stack
trace doesn't make much sense, since the addresses of functions depend on
the version/config/etc.
--
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid. Get yourself a better computer" - Dilbert.
------------------------------
From: [EMAIL PROTECTED] (Chetan Ahuja)
Subject: Ramdisked X server distro
Date: 2 Aug 1999 06:08:37 GMT
Hi,
I am looking to fit a small distribution, which nonetheless includes
an X server, onto a Zip disk ( 'coz floppies are too slow). I want to run
the system entirely out of Ramdisks once booted ( i.e. the boot disk
should be unmounted after the bootup process.) The idea is that I would
be using this machine to do run probes on video cards ( to find out amount
of memory, chipsets etc on the card) which may or may not hang the machine.
In case it hangs, I just want to hit the reset button and reboot. Does
such a distribution exists or will I have to build it from scratch? I know
that many small floppy based distributions exist but none of them include
X and I think I'll need at least the X libraries to be able to probe the
video card. Or am I completely wrong and I can run probing software ( I
have heard of Super Probe... never used it independently of XF86setup
or such other programs). Any information/hints etc will be appreciated..
Also, please cc the reply if possible.
Thanks
Chetan Ahuja
--
------------------------------
From: [EMAIL PROTECTED] (Joseph Girona)
Subject: /dev/hdb1 reached max mount count and the system hangs. How can I fix it.
Thanks. JG
Date: Mon, 02 Aug 1999 05:47:35 GMT
Reply-To: [EMAIL PROTECTED]
I am getting /dev/hdb1 reached max mount count and the system hangs.
How can I fix it. Thanks. JG
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: NTFS Status?
Date: Mon, 02 Aug 1999 06:13:37 GMT
Hi Igor,
Do I have to do anything special other than just mounting a compressed
ntfs partition ? I'm using Redhat 5.2 and it doesn't work for the
compressed drive but works for the uncompressed one.
Thanks in advance,
hrvd
In article <7o1lf6$iuf$[EMAIL PROTECTED]>,
"Igor Zlatkovic" <[EMAIL PROTECTED]> wrote:
> Hi there.
>
> You can compile in the write support for ntfs with no harm. When
mounting
> ntfs volumes, use the mount option to mount them read-only. This way,
you
> will not be harmed by the experimental write support, even if present
in the
> driver.
>
> Writing to ntfs volumes works well, as long you don't ask too much. As
long
> you want to write few files, everything will work fine. Don't expect
much
> when using software raids, for example.
>
> I didn't have bad expiriences with write support. Writing a file or
two,
> like editing a configuration file and writing the altered version back
to
> disk always went good.
>
> Once I tried to copy the whole directory structure from ext2fs to ntfs
which
> contained about 20000 files. The ntfs driver crashed after about 30
seconds.
> Dont let this be a measure, because I have asked for it. It was a
> multiprocessor machine and the target drive was a compressed ntfs5
(win 2000
> beta 3) volume. Even then, no filesystem corruption of the ntfs5
volume
> occured.
>
> Shortly, ntfs basically works. Be careful, but you can use it.
>
> Ciao
> Igor
>
> David Isaac Stclair <[EMAIL PROTECTED]> wrote in message
> news:7nnch5$gr5$[EMAIL PROTECTED]...
> > Anyone know the current status of the NTFS on linux? I'm trying to
help
> > out a friend using NT, Redhat and Suse on the same machine.
> > I got read access working using kernel 2.2.10
> >
> > I did not compile write access because it looked dangerous. Anybody
got
> > any recommendations?
> >
> > --
> > _______________________________
> > David I. St.Clair
> > North Carolina State University
> > [EMAIL PROTECTED]
> >
>
>
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
------------------------------
From: Tristan Wibberley <[EMAIL PROTECTED]>
Subject: Re: /dev/hdb1 reached max mount count and the system hangs. How can I
Date: Sun, 01 Aug 1999 05:01:27 +0100
Reply-To: [EMAIL PROTECTED]
Joseph Girona wrote:
>
> I am getting /dev/hdb1 reached max mount count and the system hangs.
> How can I fix it. Thanks. JG
Read the documentation and you will find that nothing is wrong. Your
system hasn't hung (probably not), it's just doing the Linux version of
scandisk on the first partition of your second hard disk. Each ext2
filesystem has a maximal mount value, and each time the filesystem is
mounted the mount count is incremented, when the maximal mount count is
reached, the disk is checked and the count is reset - just wait and it
will carry on.
--
Tristan Wibberley
------------------------------
From: [EMAIL PROTECTED] (Yung-Hsiang Lu)
Subject: how does "top" work
Date: 2 Aug 1999 06:17:59 GMT
Hi, Everyone,
I am curious how "top" works. How can it find the utilization of CPU
idle time? Is there a place I can get the source code of top? I
believe it is not in /usr/src/linux; I did not find it in GNU ftp
sites either.
Thanks a lot.
--
Sincerely,
Yung-Hsiang Lu
[EMAIL PROTECTED]
------------------------------
** 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
******************************
