Linux-Setup Digest #316, Volume #20 Sun, 31 Dec 00 10:13:10 EST
Contents:
how do I fix this; Mandrake6.5 with pctel modem (David Punsalan)
Re: Red Hat 7.0 Extremely Slow (NDQ)
Re: Upgrading php from 4.0.3pl1 to 4.0.4 (NDQ)
Re: problem with xcdroast (Bernard)
xmodmap, AltGr and Swiss keyboard (Tobias Vancura)
Re: problem with xcdroast (Bernard)
Re: Can't get X or Gnome to Work- Help!!! (Steve Martin)
Re: Sending a message to all terminals (Chiefy)
Re: Apache Web Server Help ("Peter Tselios")
Re: Best Linux Distro? ("Peter Tselios")
Re: Suse 7.0 evaluation ("Peter Tselios")
creating rpms from tar.gr ("Peter Tselios")
another question with xcdroast, previous problem solved (Bernard)
----------------------------------------------------------------------------
From: David Punsalan <[EMAIL PROTECTED]>
Subject: how do I fix this; Mandrake6.5 with pctel modem
Date: Sun, 31 Dec 2000 03:03:21 -0600
Attention all pctel modem users (hsp) and knowledgeable linux users,
The last time I tried insmod on my pctel module, I was inundated with the
following error messages (see below)
any idea on how to fix the problem. I found that most if not all the
variables are declared in a file called ksyms.ver.
Moreover when I do: ksyms -a , all these variables show up.
I something fishy here? I just want my modem to work!
Pls help!
- Regards - David
Script started on Sun Dec 31 03:05:29 2000
[root@localhost hsp56-linux-1]# insmod pctel
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol bh_mask
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol request_region
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol schedule_timeout
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol __wake_up
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol
__generic_copy_from_user
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol tty_std_termios
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol schedule
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol bh_active
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol kmalloc
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol
tty_register_driver
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol probe_irq_off
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol release_region
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol kfree_s
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol tty_get_baud_rate
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol __get_free_pages
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol boot_cpu_data
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol pcibios_present
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol free_irq
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol timer_table
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol panic
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol free_pages
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol bh_base
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol do_SAK
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol
interruptible_sleep_on
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol request_irq
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol
tty_flip_buffer_push
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol __verify_write
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol tty_hangup
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol pci_find_device
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol sprintf
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol tty_hung_up_p
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol check_region
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol jiffies
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol
tty_wait_until_sent
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol printk
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol timer_active
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol __const_udelay
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol bh_mask_count
/lib/modules/2.2.13-7mdk/net/pctel.o: unresolved symbol probe_irq_on
[root@localhost hsp56-linux-1]# exit
------------------------------
From: NDQ <[EMAIL PROTECTED]>
Subject: Re: Red Hat 7.0 Extremely Slow
Date: Sun, 31 Dec 2000 12:25:07 +0100
Reply-To: NDQ <[EMAIL PROTECTED]>
Robert Morelli wrote:
>
> I just installed Red Hat 7.0 on a Pentium 166 laptop
> with 64 Mb of ram and an 80 MB swap partition. The
> system is functioning, but it is extremely sluggish,
> as if the processor has been
> set to low speed. The slowness began already during
> installation, which took over three hours off an 8X
> CD-ROM. The installed system takes well over 5 minutes
> to boot, the mouse moves jerkily, windows redraw painfully
> slowly, etc. I even get warning dialogs popping up stating
> that this or that software is not responding (probably because
> the software is taking so long to run).
>
> One clue is that if I reboot without shutting off the power,
> the memory test on reboot runs very slowly. Of course,
> by this point, linux is gone, which suggests again that
> RH 7.0 is switching the processor into a low speed mode.
>
> The machine also has a Win95 partition and an OS/2 partition
> and these are still functioning normally. That doesn't
> completely rule out a hardware problem, but it does make it
> less likely.
>
> Is it possible that RH 7.0 is optimized for a Pentium II in such
> a way that it suffers on an older processor, or causes it to
> slow itself down? (In various places the OS reports that it is
> running on an i586.)
>
> The software I installed was the very minimum necessary
> to have networking, internet, and GNOME 1.2. I
> understand that a linux system generally requires more
> ram and runs more slowly than a comparable Windows or
> OS/2 system. However, I don't think this alone can
> account for such a dramatic difference on this machine.
> I'd really rather not catapult myself back into the
> 1980's by stripping out GNOME and recompiling the kernel
> to remove functionality and such. A P166 with 64MB ram may
> not be a supercomputer, but it should be sufficient to run a
> basic OS, even a pokey one. (I know, I know -- go tell that
> to the Win2000 team.)
>
> Any help much appreciated.
I don't use Laptop at all. But I installed RH7 on my PC P166 MMX/64Mb
ram and I found this version runs very fast by comparing to old version
6.x (which I used before RH7).
--
NGUYEN-DAI Quy
------------------------------
From: NDQ <[EMAIL PROTECTED]>
Crossposted-To: alt.comp.lang.php
Subject: Re: Upgrading php from 4.0.3pl1 to 4.0.4
Date: Sun, 31 Dec 2000 12:31:26 +0100
Reply-To: NDQ <[EMAIL PROTECTED]>
"M�rio Gamito" wrote:
>
> Hi!
>
> I wonder if you can help me out with this one.
> I've done my "home work" before coming here, but found nothing valuable.
>
> i have compiled/installed apache 1.3.14 with php 4.0.3pl1 as a module.
> Now i want to upgrade to 4.0.4, since it has several bug fixes and addons in
> respect to MySQL, something that is very important to me.
>
> The question is, which is the best way to *cleanly* uninstall 4.0.3pl1?
> Shall i erase all 4.0.3pl1 related files and directories, erase all
> /usr/local/httpd/(...) and start all over again compiling apache with 4.0.4?
Not need at all. You configure with the same "--prefix" and install.
I does that in my server and it works fine ;-) But there is not yet Zend
Optimizer for 4.0.4 !
Q.
------------------------------
From: Bernard <[EMAIL PROTECTED]>
Subject: Re: problem with xcdroast
Date: Sun, 31 Dec 2000 11:50:57 GMT
Hi Bill,
and thanks for your reply.
I am not very keen on further testing xcdroast 0.98, since it is on alpha
version, and, besides, in the README file, the author warns that no one
should test it unless he/she is already very familiar with xcdroast. He also
warns that he has some reports about compiling problems on some configs.
Indeed, I was unable to compile 0.98 on my system. Then instead of focusing
on such problematic install, I said to myself that v. 0.96e had been working
with a number of people for about 2 years, therefore there was no reason why
it should not work on mine. 0.96e has been tested a lot more than 0.98...
Besides, I have an old system (processor Cyrix P200+, which is equivalent to
Pentium 166, Linux 2.2.5-15 (RedHat 6.0). I don't use KDE or GNOME, but only
fvwm2, which satisfies me plenty.. Please, see below my replies to your
questions :
Bill Unruh wrote:
> a) Get the latest version, which is 0.98test8. Also get the new
> cdrecord, cdda2wave, etc.
>
> b) Try burning directly with cdrecord. First run the -dummy run and then
> try buring. Make sure you use the -v option to cdrecord so you get all
> the messages.
I will try that with my old version of cdrecord, and tell you later what
happens.
>
>
> What happens when you mound the cdrom and try to read it? Does it work?
> If it works. Fine. forget about the verify.
Yes it does ! And that is what puzzles me most ! It does read perfectly
cdroms, and, as far as writing, it also does something, since the burnt cds
are no longer writeable afterwards... (but the TOC is said to be unreadable)
>
>
> In <[EMAIL PROTECTED]> Bernard <[EMAIL PROTECTED]> writes:
>
> ]Hi there,
>
> ]As anyone in here any experience with xcdroast ? I can't get it to
> ]work !
>
> ]I have downloaded xcdroast 0.96e. The installation went OK. My SCSI cd
> ]roaster (TEAC CD R55S) is being reckognized all right. Besides, it
> ]operates allright as a cd reader on my Red HAT 6.0 linux system (Kernel
> ]2.2.5-15). The "SETUP" phasis goes OK. If I choose "Master CD" with a
> ]few Mb of data to burn into a cdrom, the image goes allright... even the
>
> ]cdburning seems to go allright : no error messages or anything.
> ]However, when I try to verify the burnt cd, I am being told that :
>
> ]"The "TOC" (table of contents) of this CD reports an impossible track
> ]size. This can be a bug in the cdrom firmware, or a defect on a non
> ]standard CD. Please try another CD-ROM drive, or forget your idea to
> ]copy this CD..."
>
> ]I have, of course, tried on more than one CD, which CDs I had previously
>
> ]checked using same cd burner on Windows environment and Windows driver.
> ]I have tested CD mastering with little data (about 20 Mb), and with
> ]more data (about 200 Mb), I tested the default unix system for cdrom
> ]files, as well as the msdos system... to no avail... always the same
> ]result and error messages. As far as burning speed is concerned, I
> ]started with speed 2, then speed 1, then speed zero ! I check the
> ]image contents each time, and it seems allright...
>
> ]Thanks for any hint on what might have gone wrong.
------------------------------
From: [EMAIL PROTECTED] (Tobias Vancura)
Subject: xmodmap, AltGr and Swiss keyboard
Reply-To: [EMAIL PROTECTED]
Date: 31 Dec 2000 12:54:46 +0100
Hello everybody,
I have been playing around with xmodmap over the last couple
of days trying to make a .xmodmap file for a German/Swiss
keyboard (strange that with RH7 only the American was
installed).
The thing I could not find out was how to use the "AltGr"
key to the right of the space bar. I guess that I did not
tell xmodmap exactly how to use the modifier keys.
Example:
xmodmap -e "keycode 10 = 1 plus bar"
should put "1" on the key labeled 1, "+" when it is used
together with the Shift key and "|" when used together with
AltGr (provided the modifiers are correct). Somehow I don�t
get this to work. Instead I get the "1" and "+" and when I
press "AltGr"+"1" the shell (bash) catches the keys and I
get the
(arg: 1)
prompt. I can suppress this by removing the mod1 entry.
Could someone please tell me where I find a more complete
description of xmodmap (what are "keysym", "keycode" and
modifiers actually) or even better tell me what the solution
to the AltGr problem is.
I use bash as Shell and kde1 as window manager.
Thanks a lot and Happy New Year,
Tobias
--
**********************************************************************
Tobias Vancura Email: [EMAIL PROTECTED]
------------------------------
From: Bernard <[EMAIL PROTECTED]>
Subject: Re: problem with xcdroast
Date: Sun, 31 Dec 2000 11:57:03 GMT
Hi David,
Thanks for your reply
I am not very keen on further testing xcdroast 0.98, since it is on alpha
version, and, besides, in the README file, the author warns that no one
should test it unless he/she is already very familiar with xcdroast. He also
warns that he has some reports about compiling problems on some configs.
Indeed, I was unable to compile 0.98 on my system. Then instead of focusing
on such problematic install, I said to myself that v. 0.96e had been working
with a number of people for about 2 years, therefore there was no reason why
it should not work on mine. 0.96e has been tested a lot more than 0.98...
Besides, I have an old system (processor Cyrix P200+, which is equivalent to
Pentium 166, Linux 2.2.5-15 (RedHat 6.0). I don't use KDE or GNOME, but only
fvwm2, which satisfies me plenty.. Please, see below my replies to your
questions :
I have read the Howto, on which I dodn't find any hint for my problem.
Have a happy New Year and New Century ! (We are going to get it sooner than
you in here...)
David wrote:
> Bernard wrote:
> >
> > Hi there,
> >
> > As anyone in here any experience with xcdroast ? I can't get it to
> > work !
> >
> > I have downloaded xcdroast 0.96e. The installation went OK. My SCSI cd
> > roaster (TEAC CD R55S) is being reckognized all right. Besides, it
> > operates allright as a cd reader on my Red HAT 6.0 linux system (Kernel
> > 2.2.5-15). The "SETUP" phasis goes OK. If I choose "Master CD" with a
> > few Mb of data to burn into a cdrom, the image goes allright... even the
> >
> > cdburning seems to go allright : no error messages or anything.
> > However, when I try to verify the burnt cd, I am being told that :
> >
> > "The "TOC" (table of contents) of this CD reports an impossible track
> > size. This can be a bug in the cdrom firmware, or a defect on a non
> > standard CD. Please try another CD-ROM drive, or forget your idea to
> > copy this CD..."
> >
> > I have, of course, tried on more than one CD, which CDs I had previously
> >
> > checked using same cd burner on Windows environment and Windows driver.
> > I have tested CD mastering with little data (about 20 Mb), and with
> > more data (about 200 Mb), I tested the default unix system for cdrom
> > files, as well as the msdos system... to no avail... always the same
> > result and error messages. As far as burning speed is concerned, I
> > started with speed 2, then speed 1, then speed zero ! I check the
> > image contents each time, and it seems allright...
> >
> > Thanks for any hint on what might have gone wrong.
>
> First thing I would say would be to go here: http://www.xcdroast.org/
> and get xcdroast-0.98 and cdrecord-1.9
>
> Then this link might help.
>
> http://www.guug.de/~winni/linux/CD-Writing/CD-Writing.sgml
>
> --
> Confucius say: He who play in root, eventually kill tree.
> Registered with the Linux Counter. http://counter.li.org
> ID # 123538
> Completed more W/U's than 98.967% of seti users. +/- 0.01%
------------------------------
From: Steve Martin <[EMAIL PROTECTED]>
Subject: Re: Can't get X or Gnome to Work- Help!!!
Date: Sun, 31 Dec 2000 12:14:08 GMT
[EMAIL PROTECTED] wrote:
>
> Hi,
> I just installed redhat 6.2 on my Dell inspiron 7500. When I run startx
> I get a very basic desktop with Xterm windows and nothing else. When I
> try to run X or Gnome, I get a screen with a gray background and a
> mouse pointer and nothing else. Does anyone know why this is happing?
There are a couple of ways around this. What's happening is that
there is nothing in your startup scripts that tell the system to
do anything other than start a very basic X desktop with Xterm.
This is a "default" desktop. One way to fix this would be to go
to /usr/X11R6/lib/X11/xinit and edit the xinitrc file (you must be
superuser to do this). That's where the stuff is that gets run
if nothing else overrides it; look down at the bottom of the file.
The other way, easier and actually more user-friendly IMHO, is
to simply come up in run level 5. This starts X on bootup and
gives you a GUI login screen. From there, once you log in, you
get thrown right into Gnome (or KDE or whatever, depending on
what you select under the "System" menu at the login screen).
You must edit /etc/inittab (again, you must be superuser).
Change the line that says "id:3:initdefault:" to read
"id:5:initdefault:" (just change the 3 to a 5). If you want
to try it first without messing with inittab, just type
"linux 5" at the LILO boot prompt (assuming your default
boot image is labelled "linux" -- see /etc/lilo.conf).
------------------------------
From: [EMAIL PROTECTED] (Chiefy)
Subject: Re: Sending a message to all terminals
Date: 31 Dec 2000 12:28:41 GMT
31 Dec 2000 01:35 UTC, Paul Hustava typed:
>man wall
Excellent...Thanks
--
Chiefy. <[EMAIL PROTECTED]>
------------------------------
From: "Peter Tselios" <[EMAIL PROTECTED]>
Subject: Re: Apache Web Server Help
Date: Sun, 31 Dec 2000 15:28:06 +0200
> Pardon my butting in here but I too have a "somehow" related question.
> This is my first acquaintance with Apache. I found that a trailing
> slash is required to make individual users' directories visible in a
> browser.
You area partially right. If you have installed your server in a DNS
you do not have to add the trailling /. If you do not have a DNS entry, then
you have to add the /.
>Additionally, not just the /public_html (per user) directory
> must be world accessible to Apache but the User's directory as well
> (how else could Apache get to the public_html ?). However the latter
> must(? [ie. the user dir] ) be true and then perhaps, its a matter of
> automatic inheritance - whatever.
I insist (hmm, probably a mispelling :( ) that in the apache conf there
is an example that shows how to do it with read only permitions. If you do
not have "x" then the dir is not accessible (or am I wrong?)
------------------------------
From: "Peter Tselios" <[EMAIL PROTECTED]>
Subject: Re: Best Linux Distro?
Date: Sun, 31 Dec 2000 15:31:27 +0200
You could try Suse too. I use it in my home PC and I like the organised
structured of directories. Also, it has a huge amount of applications, and a
smooth installation. Probably it's not as secure as Slackware, but it is
nice. On the other hand, if you like Gnome, go to Debian.
Peter
"It's interesting to note that one of the main reasons for writing the first
Unix kernel was to get better interactive performance from the computer to
play a game called Space Travel."
"Michael Madden" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Hello:
>
> I've been using Redhat since 5.0. However, I find with each
> release it gets more buggy, bloated, insecure, and unreliable.
> What is considered the most reliable, secure, high performance
> Linux distribution. Many of my pro-Debian friends have
> suggested that I try out Debian 2.2. Any opinions?
>
> Thanks,
>
> Mike
------------------------------
From: "Peter Tselios" <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux
Subject: Re: Suse 7.0 evaluation
Date: Sun, 31 Dec 2000 15:35:20 +0200
Full version is 6 Cds (but less in ws edition) in contrary with the 1 cd of
the evalutation. On the other hand, you can download (almost) everything, so
no big deal. You can also try mirror sites.
Peter
"It's interesting to note that one of the main reasons for writing the first
Unix kernel was to get better interactive performance from the computer to
play a game called Space Travel."
"malawueng" <[EMAIL PROTECTED]> wrote in message
news:92mdl6$lft$[EMAIL PROTECTED]...
> hi,
>
> i was thinking of downloading Suse 7.0 from the net but can only find a
> evaluation. Does anybody knows the difference betweent this version and
the
> full version? Is it fully functional?
>
> thanx
>
>
>
------------------------------
From: "Peter Tselios" <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux.suse
Subject: creating rpms from tar.gr
Date: Sun, 31 Dec 2000 17:04:35 +0200
Hallo to everyone,
I would like to create rpms from tarballs. Rpm's man says that I can use the
command:
"rpm --tarbuild tarball" which I translate it as: "rpm --tarbuild
package.tar.gz".
Sometimes I have the reply "tarbuild unknown option" but always I have a
segmentation fault.
The rpm I use is version 3.0.3 in SuSe 6.3
I do not want to use alien since alien creates rpms with no prefix install
directory and thats the main reason I want to create rpms (install -
unistall).
Any ideas?
--
Peter Tselios
[EMAIL PROTECTED]
------------------------------
From: Bernard <[EMAIL PROTECTED]>
Subject: another question with xcdroast, previous problem solved
Date: Sun, 31 Dec 2000 15:08:35 GMT
Hi there,
I did find the reason of my late error message : "The TOC of this CD
reports an impossible track size...". There is, in xcdroast, a feature
that is somewhat confusing. If you press the "verify burnt CD" right
after a simulation, it does not reckognize the CD and speaks out said
error message. Once this has happened, the CD is wasted and you have to
throw it out.
I then made a test in not making use of the verify button after
simulation ; I did physically burn the cd and verified only afterwards.
this time, it worked... but not as I would have expected. The verify
function says :
"verify failed ! Files differ on block 17 !"
However, the burnt cd is readable in either one of my two cdrom drives ;
I can find the copied files, open them etc... There must be some wrong
files somewhere, but I haven't discovered them yet.
I understand that the reason of such failure may be on the burning speed
(X2), but, what I would like to know is how to carry a check on the
simulated burnt image rather than on the burnt cd, when it is too late.
As previously said, the "verify burnt image" on a simulated burnt image
fails and wastes the CD.
Thanks for any hint
and Happy New Year to everyone !
------------------------------
** 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 by posting to comp.os.linux.setup.
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-Setup Digest
******************************
