Linux-Misc Digest #363, Volume #26 Tue, 21 Nov 00 11:13:04 EST
Contents:
Re: unresolved symbols in modules after kernel recompile (Bill Delphenich)
microemacs 4.00 port? (Martin Gregorie)
Re: simple shell script variable question (Eric)
Re: Undelete a file in Linux (Eric)
Re: Set environment variable in bsh script and effect another program. (Eric)
ftp slow but not telnet on RH 7 ("Legend")
Re: Frame Buffering and Red Hat 7.0 ([EMAIL PROTECTED])
I've Posted the answer earlier. Here it is ("Jeff")
Re: Overhead of printer triver too high to suit me. (Robert Heller)
Re: Why does linux keep crashing? (Robert Heller)
Qmail+Vpopmail problem ("Martino")
Re: microemacs 4.00 port? (Neil Cherry)
printer ("^Stinger^")
RPM's (TimCJ)
serial port (Max Lungarella)
Missing operating system? ("Eric")
ext2 fs on magneto optical disks ("Massimiliano Caovilla")
----------------------------------------------------------------------------
From: Bill Delphenich <[EMAIL PROTECTED]>
Crossposted-To: osu.sys.linux
Subject: Re: unresolved symbols in modules after kernel recompile
Date: Tue, 21 Nov 2000 08:13:38 -0500
Bill Unruh wrote:
> In <8vceuc$so9$[EMAIL PROTECTED]> [EMAIL PROTECTED] (doug reeder)
>writes:
>
> >I'm recompiling the 2.2.12 kernel to enable support for a
> >parallel port printer. Unfortunately, this breaks everything
> >that's compiled as a modules.
>
> ALL distributions that I have every heard of support parallel port
> printers out of the box. What made you think you should recompile?
> Despite what some of the ancient and outdated howtos say, you almost
> never should recompile your kernel. You will almost always make things
> worse, rather than better.
> (Yes, it is possible to recompile one's kernel and have it work.
> )
I seem to remember hearing that parallel port support only works when installed as a
module.
All distributions I have run into include the module for parallel port support out of
the
box.
I have always recompiled my kernel, being pretty much old-school. However, I have been
wrestling with RH7 lately, and I have never run into so many problems. This is not a
happy
situation for the Linux community if we lose the ability to do our own kernels.
------------------------------
From: [EMAIL PROTECTED] (Martin Gregorie)
Crossposted-To: comp.os.linux.setup
Subject: microemacs 4.00 port?
Date: Tue, 21 Nov 2000 13:17:44 GMT
I just grabbed the curretn source archive for microEmacs from its home
site. However, there is no Linux subdirectory and sure enough, a quick
hack at setting one up failed to compile.
Has anybody ported microEmacs 4.00 to Linux?
If so, can you point me to a source for the files you had to modify,
please (estruct.h and makefile at least) and stop me re-inventing the
wheel.
TIA
--
gregorie | Martin Gregorie
@logica | Logica Ltd
com | +44 020 76379111
------------------------------
From: Eric <[EMAIL PROTECTED]>
Subject: Re: simple shell script variable question
Date: Tue, 21 Nov 2000 14:47:23 +0100
Reply-To: [EMAIL PROTECTED]
[EMAIL PROTECTED] wrote:
>
> On Mon, 20 Nov 2000 23:35:42 GMT,
> Mark Post <[EMAIL PROTECTED]> wrote
>
> >>>i=`whoami`
>
> >>>This text does not show what it i
> looks like in the shell, but it is the
> >>>same key as the ~, only without the shift.
>
> >>Following that instruction I get #whoami#
>
> >I just tried his example, and it worked
> for me. I don't know if it makes
> >any difference or not, but what shell
> are you running? I'm running bash.
> >Another possibility would be to use this:
> >i=$(whoami)
> which also works under bash.
>
> The original poster's email was a .fr,
> meaning he wrote from France. Probably
> the trouble is everyone's using different
> keyboards due to the different languages.
>
> How/whether non-alphabetic
> characters entered on different
> language keyboards are handled in a
> standard way by a shell
> is a matter I know nothing about.
> . . . I'm curious how it is managed.
>
> I'll be back to see if anybody explains it.
>
You mean you haven't selected a keyboard layout?
Well you did, and you probably chose the (default) US keyboard (or
microsoft 101 key, don't recall what it was called during setup) layout.
I did the same, as I use a MS keyboard (I live in Holland though)
Eric
------------------------------
From: Eric <[EMAIL PROTECTED]>
Subject: Re: Undelete a file in Linux
Date: Tue, 21 Nov 2000 14:51:18 +0100
Reply-To: [EMAIL PROTECTED]
[EMAIL PROTECTED] wrote:
>
> One way would be to create your own version of 'rm' which cycles files
> into some other directory (junk, for instance) - and then have a cron job
> which periodically cleans out that directory.
>
> If you need to 'undelete', simply go to that directory and get it.
>
Yep, I have written a script that moves stuff to trash and have it
aliased to rm
There's hardly any overhead, I can still use the original rm by either
specifying the entire path, or escape rm from the shell.
There's one big BUT, and that is that whenever I work on someone else's
system I have to be very carefull :-)
Eric
------------------------------
From: Eric <[EMAIL PROTECTED]>
Subject: Re: Set environment variable in bsh script and effect another program.
Date: Tue, 21 Nov 2000 14:58:15 +0100
Reply-To: [EMAIL PROTECTED]
Joe (mvjap3) Philbrook III wrote:
>
> Ok I'm not sure if this is what your looking for... But here goes.
>
> If you mean you want to run a script that will set some environmental
> variables <or any variable for that matter> then when that script
> terminates and you return to the command prompt, you want those variables
> to still equal what your script just set them to???
>
> Then there is a workaround... lets say your script is named fix_env
> then instead of calling it with
>
> $ fix_env
>
> try calling it with
>
> $ . fix_env
>
This is called sourcing the script
`source fix_env` will do the same
It's not a workaround, but a very useful feature.
If you eg, edit .bashrc to add some aliases, source the new .bashrc, and
the new aliases will have taken affect. So you don't have to login to
ensure an adjusted rc script is used in your current shell.
Eric
------------------------------
From: "Legend" <[EMAIL PROTECTED]>
Subject: ftp slow but not telnet on RH 7
Date: Tue, 21 Nov 2000 09:10:57 -0500
I just installed RH 7 and when I try to ftp to another site, it gave me the
login prompt like at least five minutes later. I fixed the "telnet slow"
problem by adding the client hostname in /etc/hosts. If I 'telnet to the
machineA', the connection was quick, but if I ftp to machineA, it must
waits for at least five mins to connect'. Once connected, the download was
quick.
Anyone knows why? I think there is some documented workaround for this.
Anyone knows how?
Thank you very much
Benny
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: alt.os.linux,alt.linux,linux.redhat
Subject: Re: Frame Buffering and Red Hat 7.0
Date: Tue, 21 Nov 2000 14:14:43 GMT
FYI
I have a trident video card (I can't remember the model, I am not at my
house right now). And I have a CTX 1562GM Monitor. But I don't use the
CTX Monitor defaults because it never works. After I use the probe tool
that comes with the old Xconfigurator with 6.2
I like to have the 800x600 with a 72 refresh rate but the only quality
configuration seems to be after it probes and gives a recommended
configuration is the 1020x768
According to the lilo.conf below.... I know that I am not supposed to
put that menu in there..Haha
Thanks for your help guys. Does anybody know why the Xconfigurator for
the Redhat 7.0 release doesn't do a probe on the card anymore??
> <snip>
> edit /etc/lilo.conf
> set vga=xxx to one of the following
>
> Colours 640x480 800x600 1024x768 1280x1024 1600x1200
> --------+---------------------------------------------
> 256 | 769 771 773 775 796
> 32,768 | 784 787 790 793 797
> 65,536 | 785 788 791 794 798
> 16.8M | 786 789 792 795 799
>
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: "Jeff" <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux,linux.redhat.install,alt.linux
Subject: I've Posted the answer earlier. Here it is
Date: Tue, 21 Nov 2000 09:39:57 -0500
<[EMAIL PROTECTED]> wrote in message news:8vbgf1$ebh$[EMAIL PROTECTED]...
> I am at my last thread of sanity...
>
> I am using Red Hat 7.0 and trying to get KDE 2.0 on it.
>
> I have deinstalled KDE 1.0 and installed KDE 2.0 = NOTHING
> I have renamed KDE 1.0 and reinstalled KDE 2.0 = NOTHING
> I have deinstalled KDE 1.0 and tarred the slackware KDE 2.0 (is it
> compatible with Red HAT??) = NOPE!!
>
> I have also installed the KDE rpms from www.kde.org's site and I get a
> grey screen and a KDE logo to start up with but craps out after 10
> seconds.
>
> Somebody suggested compiling the source.. but I loaded it and said that
> I didn't have the correct C compiler on it (or something like that) AND
> I checked....sure enough I have gcc on the machine.
>
> Can anybody make any sain reason on how to proceed???
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
------------------------------
From: Robert Heller <[EMAIL PROTECTED]>
Subject: Re: Overhead of printer triver too high to suit me.
Date: Tue, 21 Nov 2000 14:43:22 -0000
Jean-David Beyer <[EMAIL PROTECTED]>,
In a message on Mon, 20 Nov 2000 12:50:29 -0500, wrote :
JB> When I print (Red Hat Linux 6.0, Dual 550MHz Pentium III
JB> processors), the system time jumps up to over 50%. That is a
JB> heck of a lot of system time for an essentially 1100MHz system.
No, top is talking about 50% on ONE processor. 25% of your 1100MHz
system. Top does NOT take the number of processors into account. See what
happens when you really beat on the machine. Watch top tell you that
system utilization is at 200%! Do a kernel compile in one window and
rebuild gcc in another. Check out top in a third...
Also, how long is at 50%? The system will get very busy *for a very
brief time* when gs starts to process a file. It needs to allocate
(and initialize) memory for the printer raster. If this is a 'high
resolution' printer (1440dpi? lots of color?), this means a lot of bits.
JB>
JB> I have a Hewlett Packard DeskJet 660Cse printer connected to
JB> the parallel port. Right now, I am not printing, am running two
JB> instances of SetiAtHome and this Netscape and the system time
JB> is only around 6%. The gs process (that was filtering the
JB> output of Acrobat, but Acrobat had already exited) only takes
JB> about 14% of the machine, so the rest must be in the printer
JB> driver or something. Could I have made a configuration error or
JB> something? Why would the driver (if it is the driver) consuming
JB> so much CPU time? Can I do anything about it?
JB>
JB> --
JB> .~. Jean-David Beyer Registered Linux User 85642.
JB> /V\ Registered Machine 73926.
JB> /( )\ Shrewsbury, New Jersey
JB> ^^-^^ 12:45pm up 6 days, 3:28, 2 users, load average: 1.22, 1.75, 1.94
JB>
JB>
JB>
JB>
--
\/
Robert Heller ||InterNet: [EMAIL PROTECTED]
http://vis-www.cs.umass.edu/~heller || [EMAIL PROTECTED]
http://www.deepsoft.com /\FidoNet: 1:321/153
------------------------------
From: Robert Heller <[EMAIL PROTECTED]>
Subject: Re: Why does linux keep crashing?
Date: Tue, 21 Nov 2000 14:43:23 -0000
"Bob Jones" <[EMAIL PROTECTED]>,
In a message on Tue, 21 Nov 2000 07:10:09 GMT, wrote :
"J> It doesn't seem to be the same things. It is almost always an error about
"J> accessing virtual memory. I though it might be my swap so I created
"J> another one on a different drive but had the same problems.
"J> This is driving me bonkers as I've been bettling it for a month now. :(
Random thoughts:
Are you over-clocking your processor?
Is your processor's fan working?
(Is your processor overheated?)
Have you checked your memory? Are you running it at the proper
speed? (Ie. are you running 66mhz RAM on a 100MHZ bus or some thing?).
Have you done a intensive memory test?
(Is there something wrong with your memory?)
It *sounds* like you have some sort of *subtle* hardware problem.
Problems accessing virtual memory or handling page faults suggest a
problem with the CPU VM logic or problems with memory -- something is
trashing these addresses somehow.
"J>
"J> In article <[EMAIL PROTECTED]>, "Jean-David Beyer"
"J> <[EMAIL PROTECTED]> wrote:
"J>
"J> > Bob Jones wrote:
"J> >
"J> >> In article <8v73mj$sjc$[EMAIL PROTECTED]>, "Jerry L Kreps"
"J> >> <[EMAIL PROTECTED]> wrote:
"J> >>
"J> >> > "I'm running the latest developement development kernel"
"J> >> > If it's an odd numbered kernel that says it all. JLK
"J> >> >
"J> >>
"J> >> I should have mentioned that it does the same thing for 2.2.16
"J> >
"J> > If it continues to crash, even for a nominally stable (even numbered
"J> > second thing in the release number) version, then you should reconsider
"J> > the possibility that it is a hardware problem.
"J> >
"J> > Does it always crash when you are doing the same thing, with the same
"J> > error messages in /var/log/messages (if any)? or the same symptoms? That
"J> > could indicate some kind of configuration problem, or a bum module or
"J> > something.
"J> >
"J> > --
"J> > .~. Jean-David Beyer Registered Linux User 85642.
"J> > /V\ Registered Machine 73926.
"J> > /( )\ Shrewsbury, New Jersey
"J> > ^^-^^ 12:10pm up 6 days, 2:53, 2 users, load average: 2.01, 1.86, 1.98
"J> >
"J> >
"J> >
"J>
--
\/
Robert Heller ||InterNet: [EMAIL PROTECTED]
http://vis-www.cs.umass.edu/~heller || [EMAIL PROTECTED]
http://www.deepsoft.com /\FidoNet: 1:321/153
------------------------------
From: "Martino" <[EMAIL PROTECTED]>
Subject: Qmail+Vpopmail problem
Date: Tue, 21 Nov 2000 15:59:00 +0100
I'm running a Server with Mandrake7.1+Qmail+Vpopmail for managing virtual
domains. All works well without problem. The only drawback is that if I
change the localhost name (or domain), vpopmail don't authenticate my
(virtual domain) users anymore, and I have to rebuild all virtual accounts.
Someone can help me (if someone can understand my terrible
explanation...)....
TIA
Max
------------------------------
From: [EMAIL PROTECTED] (Neil Cherry)
Crossposted-To: comp.os.linux.setup
Subject: Re: microemacs 4.00 port?
Reply-To: [EMAIL PROTECTED]
Date: Tue, 21 Nov 2000 15:00:04 GMT
On Tue, 21 Nov 2000 13:17:44 GMT, Martin Gregorie wrote:
>I just grabbed the curretn source archive for microEmacs from its home
>site. However, there is no Linux subdirectory and sure enough, a quick
>hack at setting one up failed to compile.
>
>Has anybody ported microEmacs 4.00 to Linux?
>
>If so, can you point me to a source for the files you had to modify,
>please (estruct.h and makefile at least) and stop me re-inventing the
>wheel.
Georgie, please don't take this as being chastised but use a search
engine and you'll find the answer. I recommend www.metacrawler.com
If you still have trouble please contact me and I'll get you the
answer. I'm just trying to teach a few users how to get info so we can
get down to the really hard questions.
I hope this helps, I did find the source so it's out there.
--
Linux Home Automation Neil Cherry [EMAIL PROTECTED]
http://members.home.net/ncherry (Text only)
http://meltingpot.fortunecity.com/lightsey/52 (Graphics)
http://linuxha.sourceforge.net/ (SourceForge)
------------------------------
From: "^Stinger^" <[EMAIL PROTECTED]>
Subject: printer
Date: Tue, 21 Nov 2000 15:26:44 GMT
Can someone telle me how I can make my HP600 printer print colors under suse
6.3
GrEEtZ
------------------------------
From: TimCJ <[EMAIL PROTECTED]>
Subject: RPM's
Date: Tue, 21 Nov 2000 15:17:31 GMT
I seem to have a bit of a problem with unsatisfied dependencies when
installing rpm's. I often get the unsatisfied dependencies libc.so.6
and libm.so.6 but both are in the lib/lib directory, and glibc-2.1.1 is
definitely installed. I am totally stumped, any help appreciated!
Thanks in advance
TCJ.
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: Max Lungarella <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.app
Subject: serial port
Date: Tue, 21 Nov 2000 16:31:03 +0100
hi there
i'm trying to use the serial port to control a device connect to it. the
code compiles without warnings, but while setting the attributes of the
port i get a funny error message: "Function not implemented"
(errno=ERRSYS). Here is the simplified code snippet:
struct termios *newtio;
port = open(portname, O_RDWR | O_NDELAY);
...
tcflush(port, TCIFLUSH);
if (tcsetattr(port, TCSANOW, &newtio)<0)
cout << "Error" << endl;
does anybody of you know what i'm doing wrong?
thx a lot in advance.
max
------------------------------
From: "Eric" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup
Subject: Missing operating system?
Date: Tue, 21 Nov 2000 10:44:19 -0500
I have installed RH 6.2. I asked used Disk Druid during setup, and
allocated 16 MB to /boot (it actually allocated 22 MB, I'm not sure why),
and the remaining free space to /. I also have a WinME partition. I
installed LILO on the first sector of the /boot partition. I was hoping
that if I set the /boot partition as active using fdisk from my dos startup
disk, I could get it to run lilo and boot (since I don't want to install
LILO on on MBR, due to paranoia). But this doesn't work. All I get is
"Missing Operating System" on the screen. Booting linux from a boot
diskette works fine, and setting my primary DOS partition as active results
in successful boot into WinME.
What am I doing wrong?
Eric
------------------------------
From: "Massimiliano Caovilla" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.system
Subject: ext2 fs on magneto optical disks
Date: Tue, 21 Nov 2000 16:41:12 +0100
Hi,
I'm trying to use ext2 fs on a magneto optical disk with block size of 2048
and a total size of 2.6 Gb: I have a problem whenever I try to mkfs a
partition larger then 1 Gb (made with fdisk): I get weird scsi errors (the
mo drive is /dev/sda), while if I mkfs the whole disk without partitioning I
have no problems at all. It feels like mke2fs doesn't like partitions larger
then 1Gb on my mo disk.
Any suggestions?
ciao & thanks
------------------------------
** 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
******************************
