Linux-Misc Digest #554, Volume #21 Fri, 27 Aug 99 02:13:10 EDT
Contents:
Re: Give back on ppp connect/Netscape problem (Alan Swartz)
Has anyone connected to a TCP/IP network via a Shiva LanRover D56? (Ferrari_in_1999)
Re: Communicator 4.6 kills itself ("Jim Ross")
Re: My Linux crashes more often than M$ (Robin Smith)
Re: netscape 4.51 for linux (newsseeker)
Re: mount ([EMAIL PROTECTED])
Lost permissions for sound card? ("Steve D. Perkins")
true type fonts (Klaus Leopold)
Re: It's crashed yet again! (TurboTex)
Re: Where is "glint" after RH6.0 upgrade?!? (TurboTex)
Re: No headers? ("Matthew O. Persico")
Re: The optimization debate (was: why not C++?) (David Fox)
COMPLETE system LOCKUP Mandrake 6 (Warren Bell)
Flushing Serial Port (long) (James K. Wiggs)
Upgrading to gtk+-1.2.3-2 (Andrew Commons)
The Microsoft/Linux Conspiracy ("The Ultimate OS Portal")
[dos format] in vi (Lee Kang Won)
Re: PPP, no chat (Simon Green)
Re: Telnet vs SSH (Jon Skeet)
----------------------------------------------------------------------------
From: Alan Swartz <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup,comp.windows.x.kde,linux.redhat.install
Subject: Re: Give back on ppp connect/Netscape problem
Date: Thu, 26 Aug 1999 21:48:38 -0400
Yes, RFC1855 does say to first"summarize" a message you are replying to so we don't
have to scroll down through the who lengthy exchange on each posting.....
;-)
"Matthew O. Persico" wrote:
> Son, I remember writing PETbasic programs out to cassette tapes. I actually
> OWNED a C64 and worked on a VIC-20. Been there, done that. And when I was
> there, I learned to reply to mail and news UNDERNEATH the text I was
> replying to, not above.
>
> :-)
>
> --
> Matthew O. Persico
>
> You'll have to pry my Emacs from my cold dead oversized
> control-pressing left pinky finger. -- Randal L. Schwartz
- - - - - - - - - - - - - - -
Alan Swartz
work: [EMAIL PROTECTED]
home: [EMAIL PROTECTED]
ICQ# 15001670
Abandon Windows, Try on Linux
------------------------------
From: Ferrari_in_1999 <[EMAIL PROTECTED]>
Subject: Has anyone connected to a TCP/IP network via a Shiva LanRover D56?
Date: Thu, 26 Aug 1999 21:44:04 -0400
I am trying to connect to my company network with a dial up modem
connection through a Shiva LanRover D56. I set up a PPP connection with
NetConfig, but the connection fails. I can successfully connect when I
boot under Windblows 98. Shiva claims the following OS are supported:
Compatible with Windows 95, Windows NT, Windows 3.x, Windows for
Workgroups,
IBM OS/2, DOS, Apple MAC OS and leading UNIX operating systems
Has anyone gotten this to work under Linux? I am using RedHat 5.2 with
the 2.036 kernel.
Any help would be appreciated.
Scott
------------------------------
From: "Jim Ross" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.x,aus.comuters.linux
Subject: Re: Communicator 4.6 kills itself
Date: Thu, 26 Aug 1999 23:24:58 -0400
> > But their are issues I have with IE too.
> > IE still can't fade-in a progressive jpeg.
> > IE still has a problem with the cache that click the back button often
makes
> > IE reload the page from the Internet.
>
> you forgot the big one --- there's still no internet explorer for
> linux. i fail to see how it could be more advanced than netscape.
> netscape exists for linux. IE does not. sounds like netscape beats
> IE six ways 'til sunday.
>
> --
> J o h a n K u l l s t a m
> [[EMAIL PROTECTED]]
> Don't Fear the Penguin!
Platform support and how advanced piece of software is are too different
things.
Just because it doesn't run on Linux doesn't effect that.
Sounds more like you have a beef with Microsoft's strategy of favoring
Windows platforms
than with how good IE is where it exists.
Supercomputers are not less advanced because they don't run Linux.
Jim
------------------------------
From: Robin Smith <[EMAIL PROTECTED]>
Crossposted-To: linux.redhat.misc
Subject: Re: My Linux crashes more often than M$
Date: 26 Aug 1999 13:47:04 +0100
kev <[EMAIL PROTECTED]> writes:
> Hi,
>
> Well it's just done it again! I'm using Red Hat 6, and every now and
> then, it just decides to log me out, thus losing all unsaved information
> from the apps I had open. It doesn't give any hint that anything is
> wrong, it just logs me out. This is _very_ annoying. What was that about
> Linux being the stablest OS there is?
>
> So I'm forced to appeal to you guys to give me some insight into what is
> wrong. Again. I've had so many problems with RH6, I spend far too much
> time trying to fix problems when I should be using it to be productive.
>
> Do non-Red Hat users have these problems?
>
> Thanks,
>
> - Kev
Anything of interest in the messages file? E.g. PANIC ...
Your description is unclear, do you just lose X or the whole system.
If X then which Window Manager?
If not it could probably be a hardware fault e.g. build up of heat.
Does the problem happen after five minutes or five hours of uptime?
Robin
------------------------------
From: newsseeker <[EMAIL PROTECTED]>
Crossposted-To: alt.hacking,alt.os.linux,comp.os.linux.advocacy
Subject: Re: netscape 4.51 for linux
Date: Thu, 26 Aug 1999 23:32:41 -0500
Tim Moore wrote:
> > # chkfontpath --add /usr/X11/lib/X11/fonts/75dpi
> > Netscape needs 75dpi-fonts,
>
> Eh? I switched to 100dpi because the 75's looked so crappy in
> Netscape. 4.51 & 4.61 work just fine with 100dpi. ???
The chkfontpath fix mentioned above is documented in RedHat's Knowledge
Base/FAQ as a workaround to the Netscape crash problem.
--nwskr
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: mount
Date: 26 Aug 1999 16:58:55 +0100
cedric <[EMAIL PROTECTED]> wrote:
> While booting up I get the following message,
> Mount: /dev/hdc can't read superblock failed
> Here is the entry in /etc/fstab
> /dev/hdc /mnt/LS120 ext2 default 0 0
What about /dev/hdc1? You usually mount a partition, not a raw
disc (except cdroms, of course...).
--
Alain Borel
<[EMAIL PROTECTED]>
------------------------------
From: "Steve D. Perkins" <[EMAIL PROTECTED]>
Subject: Lost permissions for sound card?
Date: Thu, 26 Aug 1999 11:17:10 -0400
Yesterday evening I finally got around to upgrading my system
from RedHat 5.2 to 6.0. So far I have been far less than
impressed... I haven't seen any really compelling new FEATURES
available to me, yet I have had to re-install KDE from scratch
and deal with a host of other problems.
One of the problems that I haven't been able to solve is my
sound card (an ESS 1688 AudioDrive). As it stands right now...
the sound works absolutely fine when I am logged in as root, but
does not work at all when logged in as an ordinary user.
Applications give me such error messages as "cannot open output
device"... all leading me to the conclusion that my system no
longer has it's permissions set for normal users to access the
sound card.
The only problem is, I have absolutely no clue whatsoever as to
where I need to go, or what changes need to be made, in order to
allow permissions of the sound card for everyone. Anyone else
know the easy answer to this problem? Thanks!
Steve
------------------------------
From: Klaus Leopold <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.x,comp.os.linux.help
Subject: true type fonts
Date: Thu, 26 Aug 1999 14:49:10 +0200
Hello!
Is ist possible to use (Windows-) True-Type-Fonts under StarOffice 5.1?
I=B4m using SuSE Linux 6.1.
Thanks in advance,
Klaus.....
-- =
Klaus Leopold
University Klagenfurt
Institute of Information Technology Tel: ++43(0)463/2700-863
Research group: Systemintegration Fax: ++43(0)463/2700-867
mailto:[EMAIL PROTECTED]
http://colossus.itec.uni-klu.ac.at/~klaus
------------------------------
From: TurboTex <[EMAIL PROTECTED]>
Crossposted-To: linux.redhat.misc
Subject: Re: It's crashed yet again!
Date: Thu, 26 Aug 1999 23:40:10 -0500
Try XFCE3. No Worries. I can crash them all from twm to kde. XFCE
never has crashed to a place that I couldn't get another term and get
outa it..
And I do Crazy Stuff.... Vmware inside of vmware??????
kev wrote:
>
> Hi,
>
> Another X lock-up.
> This time after returning to the machine, moving the mouse (may have
> pressed a button too while movig it) to get rid of screensaver, but
> instead, the screensaver carried on running, but with an error message
> in the top corner of the screen something like "Unable to grab pointer".
>
> Any idea why it does this?
> How to stop it doing this?
>
> Thanks,
>
> - Kev
--
----
M.H. Collins < LINUX: The Official OS >
****** < for the New Millennium >
Powered by TurboLinux 3.6 http://www.linuxlink.com
Driven by XFCE3 http://www.austinlug.org
------------------------------
From: TurboTex <[EMAIL PROTECTED]>
Subject: Re: Where is "glint" after RH6.0 upgrade?!?
Date: Thu, 26 Aug 1999 23:34:54 -0500
I guess you need to get used to gnorpm........ Sux I know..
Just use cli Rpm.. Works best anyway. Try to Force a package with
glint or gnorpm..
"Steve D. Perkins" wrote:
>
> I just finished upgrading my RedHat 5.2 system to 6.0 using the
> "upgrade" option... and have just noticed yet another problem
> caused by this worthless upgrade.
>
> I've experimented with "gnorpm" under Gnome and "kpackage" under
> KDE... but I always go back to good ol' "glint"... my favorite
> application for dealing with RPM packages. However, now after
> this upgrade to RedHat 6.0... "glint" is nowhere on my system!
> I've searched the entire filesystem... but it seems that the 6.0
> installation program removed it for some reason (is it considered
> obsolete these days or something?!?).
>
> I was wondering if anyone could tell me the name and/or location
> of the specific RPM package containing "glint", so that I can
> manage to get this tool back on my system again. Thanks!
>
> Steve
--
----
M.H. Collins < LINUX: The Official OS >
****** < for the New Millennium >
Powered by TurboLinux 3.6 http://www.linuxlink.com
Driven by XFCE3 http://www.austinlug.org
------------------------------
From: "Matthew O. Persico" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup
Subject: Re: No headers?
Date: Fri, 27 Aug 1999 00:50:26 -0400
Hal Burgiss wrote:
>
> On Thu, 26 Aug 1999 12:49:11 -0400, Matthew O. Persico <[EMAIL PROTECTED]> wrote:
> >Ok. Now that I have installed 6.0 with KDE, why would the standard 'C'
> >headers (stdio.h, string.h, etc) be installed in
> >/usr/src/linux-2.2.5/include/linux/include instead of /usr/include where
> >they FREAKING BELONG?!?!?!?!?!?!?!?!
> >
>
> Do you have glibc-devel-2.1.1-6 installed?
>
Yes I believe I did, but I cannot prove it. However in the interim, I
reinstalled from scratch w/o kde or gnome. All files now appear where they
are supposed to be.
> This includes the files you want, and the path you want.
>
> >Appologies for shouting. Between this and all the PPP woes, I just might
> >re-install 5.2.
> >
> >Thank you.
> >
>
> --
> Hal B
> [EMAIL PROTECTED]
> --
> Linux helps those who help themselves
--
Matthew O. Persico
You'll have to pry my Emacs from my cold dead oversized
control-pressing left pinky finger. -- Randal L. Schwartz
------------------------------
From: d s f o x @ c o g s c i . u c s d . e d u (David Fox)
Crossposted-To: comp.os.linux.development.apps,comp.os.linux.development.system
Subject: Re: The optimization debate (was: why not C++?)
Date: 26 Aug 1999 21:58:59 -0700
[EMAIL PROTECTED] (Paul D. Smith) writes:
> likely to lead to more efficient code! I, myself, quite often use
> temporary pointers to walk through arrays rather than incrementing a
> counter and using array indexing. I actually do it because I find the
> code simpler to understand that way, but it's probably faster, too.
I believe this is faster on a sparc, slower on an x86. Or perhaps its
the other way around...
--
David Fox http://hci.ucsd.edu/dsf xoF divaD
UCSD HCI Lab baL ICH DSCU
------------------------------
Date: Fri, 27 Aug 1999 05:28:45 +0000
From: Warren Bell <[EMAIL PROTECTED]>
Subject: COMPLETE system LOCKUP Mandrake 6
This is my second post on this, I appologize, but I've had a new install
of Mandrake Linux for three days and had two system lockups! What
gives? They both happened with netscape open and when it locks, it
locks hard. No mouse, no keyboard, magic sysrq doesn't even attempt to
work, can't telnet in from another machine, nothing.
I've seen other people post on this but haven't found any answers. Is
there any settings I should check in my bios? Something that might be
conflicting with Linux?
I do have some unmached memory- 2-72pin 32M EDO and 1-168pin 128M SDRAM,
I don't know if that has anything to do with it but it's done it before
on me with just the EDO ram.
As far as my hardware, it's pretty basic. Adaptec SCSI controller,
Seagate Barracuda SCSI HD, Matrox Millennium 2 video and average sound
card.
Can anyone shed some light on this? I just don't get it. It seems like
a netscape problem but it locks the whole system.
Thanks
Warren Bell
------------------------------
From: [EMAIL PROTECTED] (James K. Wiggs)
Crossposted-To: linux.dev.c-programming,comp.os.linux.development.apps
Subject: Flushing Serial Port (long)
Date: 27 Aug 1999 04:50:15 GMT
Folks,
I've been doing some programming on a linux box to get it talk to a
device on the other end of a high-speed serial line. The devices talk
over a null-modem cable. The connection is asynchronous and the data
is not ascii, so I've got the device set in non-canonical mode. My
problem is that, while I can read data from the serial device that has
been sent to the computer, when I write to the serial device, I can not
count on the data getting to the box at the other end of the line. In
fact, it appears that the data doesn't actually get transmitted until I
*kill* the application!
I've spent hours poring over man pages, etc., trying to figure out
how to *force* data written to a serial device with the unix write()
command to actually be transmitted. At first I thought that tcflush()
might be what I was after, but it turned out to be the exact opposite
of what I wanted. It complete deleted the data that I'd written from
the output buffer *because* it hadn't yet been transmitted. Then I
tried the fdatasync() call, but it fails with the EINVAL error; no
serial devices allowed, apparently.
Here is a snippet of my code to show how the serial port was set
up (the code runs as a deamon; some of the lines here are oddly wrapped
to satisfy my fascist newsposting software):
port = open( serialdevice, O_RDWR | O_NOCTTY );
if( port < 0 ) {
fprintf( stderr, "Couldn't open serial port %s; exiting...",
serialdevice );
fflush( stderr );
exit(-1);
}
/* Set up a signal handler to catch IO interrupt signals. */
sigact_IO.sa_handler = IOSignal_handler;
sigemptyset( &sigact_IO.sa_mask );
sigact_IO.sa_flags = 0;
sigact_IO.sa_restorer = NULL;
sigaction( SIGIO, &sigact_IO, NULL );
/* Set up a signal handler to catch SIGHUP and re-read configuration
files. */
sigact_HUP.sa_handler = HUPSignal_handler;
sigemptyset( &sigact_HUP.sa_mask );
sigact_HUP.sa_flags = 0;
sigact_HUP.sa_restorer = NULL;
sigaction( SIGHUP, &sigact_HUP, NULL );
/* Enable SIGIO signals. */
fcntl( port, F_SETOWN, getpid() );
/* Make the port asynchronous. */
fcntl( port, F_SETFL, O_ASYNC );
/* Now set the port settings to their true, final value for use. */
bzero( &new_term, sizeof(new_term) );
new_term.c_cflag = BAUDRATE | CRTSCTS | CS8 | CLOCAL | CREAD;
new_term.c_iflag = IGNPAR;
new_term.c_oflag = 0;
new_term.c_lflag = 0;
new_term.c_cc[VMIN] = 1;
new_term.c_cc[VTIME] = 10;
tcflush( port, TCIOFLUSH );
tcsetattr( port, TCSANOW, &new_term );
/* Port is now ready for use. */
The code loops checking for packets sent from other processes via
UNIX sockets, which it then forwards to the device. When an IO signal
comes in, the handler sets a flag to true, and the code drops into a
subroutine to read the data from the port and then forward it on to
the appropriate process via another UNIX socket. Receiving the data
works just fine. However, when I send data out the port with the
following code, it appears that the code just buffers it somewhere in
memory and *never* sends it to the serial device. It doesn't actually
go out until I kill this daemon, which, of course, flushes its various
I/O buffers automatically as part of the exit process. I'd kinda like
to get the buffers to flush in some less destructive way...
Anyway, here's the writing code (here, the request variable is a
pointer to a variable of type request_packet):
if( (result = write( port, request, sizeof(request_packet) )) <
sizeof(request_packet) ) {
/* Couldn't write entire packet. */
fprintf( stderr, "Forward_Request_Packet: Could only write %d bytes
of request packet.\n", result );
if( result < 0 )
fprintf( stderr, "Forward_Request_Packet: errno = %d, error = %s\n",
errno, sys_errlist[errno] );
fflush( stderr );
}
I've tried using fdopen() to create a stream version of the file
descriptor, which I could then use the fflush() command on; the code
promptly started dying on startup. I've tried dup()'ing the file
descriptor, writing to the duplicate, then close()'ing the duplicate;
no joy. The fsync() and fdatasync() commands, as I've already said,
fail with an EINVAL error.
Any suggestions?
regards,
Jim Wiggs
[EMAIL PROTECTED]
------------------------------
Date: Fri, 27 Aug 1999 14:47:35 +1000
From: Andrew Commons <[EMAIL PROTECTED]>
Subject: Upgrading to gtk+-1.2.3-2
Hello,
This may be a dumb question, but...
Has anyone upgraded from gtk+-1.0.6-3 (and glib, from the RedHat 5.2
distribution) to gtk+-1.2.3-2 (and glib) using the rpm files sourced via
the gtk web site and still had everything (such as GIMP) working
afterwards?
I forced the upgrade using rpm, then manually rebuilt symbolic links to
make GIMP happy only to be eventually confronted with an undefined
symbol from GIMP, gtk_accelerator_table_set_mod_mask, at which point
I reverted to the RedHat 5.2 versions.
Polite suggestions welcome :-)
cheers & thanks,
andrew
------------------------------
From: "The Ultimate OS Portal" <the_Ultimate_OS@can't-spam-me.org>
Subject: The Microsoft/Linux Conspiracy
Date: Fri, 27 Aug 1999 01:44:19 -0400
http://theultimateos.com/The_Micronux_Conspiracy.htm
------------------------------
From: Lee Kang Won <[EMAIL PROTECTED]>
Subject: [dos format] in vi
Date: Fri, 27 Aug 1999 11:57:51 +0900
when I copy and edit(vi) files from cdrom,
I see a message [dos format] at the bottom of the line and
even though it is a shell script file, it refuses to run.
How can I change file property(which property?) and make it
executable without doing some trivial jobs?
Of course I tried changing file mode.
Thanx in advance.
------------------------------
From: Simon Green <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.networking
Subject: Re: PPP, no chat
Date: Fri, 27 Aug 1999 15:53:00 +1000
This is a multi-part message in MIME format.
==============4B496D279F63B42FF86F3576
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
You could try sing DIP instead. It has a "test" mode that you may be able to
use. However, I have to say I've never tried it.
"Matthew O. Persico" wrote:
> In order to log into my job, I must use a defender key. The sequence goes
> like this:
>
> 1) Dial the job. Get a terminal.
> 2) Put my name in the terminal. Put my PIN in the defender hardware on my
> desk.
> 3) Enter the number presented on the screen into the defender. It presents
> a number that I have to type back into the terminal.
>
> This cannot be chat'ed. I need to dial, interact and then start ppp with no
> dial and no chat, just ppp negotiation.
>
> I tried dialing with minicom, entering all the stuff and then firing up
> ifup. I removed the chat file for the particular interfac and modified
> ifup to not exit if the chat file was missing and to not use the 'connect'
> parameter in the PPP call. It didn't work - the negotiation timed out with
> an LCP error really quickly.
>
> Does anyone out there have scripts to do this? I seem to remember doing
> this many moons ago, but w/o minicom.
>
> --
> Matthew O. Persico
>
> You'll have to pry my Emacs from my cold dead oversized
> control-pressing left pinky finger. -- Randal L. Schwartz
==============4B496D279F63B42FF86F3576
Content-Type: text/x-vcard; charset=us-ascii;
name="simong.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for Simon Green
Content-Disposition: attachment;
filename="simong.vcf"
begin:vcard
n:Green;Simon
tel;cell:+61 419 479 908
tel;fax:+61 2 9202 8199
tel;work:+61 2 9202 8192
x-mozilla-html:FALSE
org:Logica Aldiscon;Customer Support
adr:;;5th floor, 76 Berry St ;Nth Sydney;NSW;2060;Australia
version:2.1
email;internet:[EMAIL PROTECTED]
title:Support Engineer
x-mozilla-cpt:;28288
fn:Simon Green
end:vcard
==============4B496D279F63B42FF86F3576==
------------------------------
From: [EMAIL PROTECTED] (Jon Skeet)
Subject: Re: Telnet vs SSH
Date: Thu, 26 Aug 1999 14:04:57 +0100
[EMAIL PROTECTED] wrote:
> I hear that you sould use SSH instead of Telnet for remote access. If I
> switch to SSH, will a windows machine still be able to telnet into my
> machine with their telnet client or will they have to get a SSH client?
You'll have to get an SSH client. However, there's a very nice telnet and
SSH client available at
http://www.chiark.greenend.org.uk/~sgtatham/putty.html
> Is SSH compatible with Telnet clients?
Not straight telnet clients, no.
--
Jon Skeet - [EMAIL PROTECTED]
http://www.pobox.com/~skeet/
------------------------------
** 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
******************************
