Linux-Misc Digest #476, Volume #18 Tue, 5 Jan 99 14:13:11 EST
Contents:
Re: Besr dual boot w/ Windows? (Leonard Evens)
Re: help me choose license (Victor Danilchenko)
Re: Slackware installation questions (Dave Brown)
Re: maxtor 17gig and rh 5.1 (joey)
Re: Why is GNOME not called a window manager? (Steve Lamb)
Re: RAMDISK in RedHat 5.2, what are them for? Can remove them? (Chee Choon Cheng
(remove "removethis" to e-mail me))
Re: Whats the best *offline* usenet reader for Linux? (Ben Sykes)
Re: gcc broken after egcs upgrade (Michael Steigman)
Re: Getting the C compiler working (Michael Steigman)
Modem vs. Sound card (RH5.2) ([EMAIL PROTECTED])
Re: RAMDISK in RedHat 5.2, what are them for? Can remove them? (Villy Kruse)
Re: Help!! We are looking for a new database system. (aldev)
Re: HELP! Lost mouse support in X! (aldev)
Re: 2038 and Linux (Villy Kruse)
Netscape 4.08 crashes when running java (Slackware-3.6) (Yukinobu Kutsukake)
Re: Anti-Linux FUD (Matthew Kirkwood)
which Xserver? (Zdravko Balorda)
Re: Netscape eats up *all* the swap (Alexander Viro)
Re: Anti-Linux FUD (Iain Georgeson)
Re: Netscape eats up *all* the swap (Alexander Viro)
Re: Configure Kernel for Oracle? (ag)
----------------------------------------------------------------------------
From: Leonard Evens <[EMAIL PROTECTED]>
Subject: Re: Besr dual boot w/ Windows?
Date: Tue, 05 Jan 1999 07:52:02 -0600
Matt O'Toole wrote:
> I'm setting up a new system with Win 95/98 (undecided) and Linux. I want to
> find the simplest, cheapest, most flexible, and most convenient boot
> manager. I have Partition Magic 3, and it works great. However, it won't
> support my new Fat32 partitions, and I don't know how full its Linux support
> is. So, to get its full benefit, I'd have to upgrade to V. 4, which costs
> money. I especially like how PM's boot manager works: it boots into the
> last system used, unless you select something else. This eliminates a step,
> so I can flick my machine on, and go make coffee, or check the mail, and
> return to have it ready to go to work. Will a free, Linux boot manager do
> this? How do I deal with how Windows overwrites the MBR? It would also be
> nice to be able to resize or change partitions later, in both Windows and
> Linux. Does PM 4 do this? Should I spend money on the upgrade, or is there
> a free, Linux answer to my needs?
>
> Matt O.
The latest version of Partition Magic (which has separated out the boot loader
as
BootMagic) does work with FAT32, but I thought the ealier versions did also.
However, I recently had a lot of problems using it with Windows 98. You may
have to get version 4.01 by download to get it to work properly. I
understand
your concern about the not having to wait for the Lilo prompt, but except for
that I would suggest doing it with lilo. It is by far the most
straightforward approach.
I suspect earlier versions of Partition Magic will have even greatger problems
with
Windows 98 and with some of the newer hard drives.
If you are starting with a single disk with Windows 95/98 on it, you can use
Fips20.exe
to non destructively contract the Windows partition. I got this to work where
Partition Magic failed. It is a standard part of the Redhat 5.2
distribution. It
can handle FAT32. Then install Linux, but when you get to the last part
(assuming you are using Redhat), put lilo in the root partition for Linux
rather
than the master boot record. It will probably also work if you put lilo in
the MBR,
but later Windows upagrades may not. For example, I have found that putting
lilo in the MBR prevents NT from booting. You also have to make sure you
make the root partition for Linux the active partition for this to work. You
can do this with the Redhat installation when you do your partitioning, but
you have to be careful. In principle, alos, before you allow the system to
reboot, you
can use one of the Alt F key combinations to get a bash prompt and run fdisk
again at the end to toggle which partition is active. However, if you have
made
a boot floppy and a rescue disk as provided by Redhat 5.2, you can boot from
the floppy in rescue mode, run fdisk again and fix it that way. One warning
about the boot floppy. It may take quite a long time to load the kernel, so
don't assume the machine has hung if nothing seems to happen for a minute
or so.
If you are going to install Windows yourself, you can use the Windows fdisk
(which
is a bit hard to find, but it is on the Windows CD) and created one primary
partition in which to install Windows, and then follow up by installing Linux
on the rest of the disk. Or you could start the Linux installation, get to
the
part where you partition the disk, write out the partition, and then stop and
install Windows before doing the Linux installation from scratch again.
Partition Magic's Boot Magic is also not without problems under 4.0. It uses
the Master Boot Record, and it is possible that some Virus Protection programs
will detect this as a boot sector virus. In our case, we did end up using Boot
Magic
because we had two disks, and Norton Anit Virus wasn't bothered by what
Boot Magic did to the MBR, but other virus protection programs may not be
as clever. (If this does happen, you can use `innoculation' as a way out, but
it does created additional problems.) (At least one earlier version of
Partition
Magic boot a boot loader in a separate partition, and I think that was a more
satisfactory way to do it.)
Recent versions of the Linux kernel can read and write to FAT32 partitions
which should be mounted as type vfat.
If you end up using two disks, one with Windows and one with Linux, you
will probably have to use Partition Magic or a similar product. But another
alternative is to partition the first disk to include a small dummy partition
(of type Linux swap, say) and write your lilo.conf to put lilo there.
Otherwise,
assuming Linux is on your second disk, there is no way to boot using lilo.
--
Leonard Evens [EMAIL PROTECTED] 847-491-5537
Dept. of Mathematics, Northwestern Univ., Evanston, IL 60208
------------------------------
Date: Tue, 05 Jan 1999 10:57:00 -0500
From: Victor Danilchenko <[EMAIL PROTECTED]>
Subject: Re: help me choose license
steve mcadams wrote:
>
> [Snipped for brevity, quoted material marked with ">"]
> On Sat, 02 Jan 1999 11:37:08 GMT, [EMAIL PROTECTED] (Alex
> Butcher) wrote:
>
> >State clearly in your licensing terms that you'll assume that all
> >contributions have been assigned for you to incorporate into both the Free
> >and non-Free versions. At least that way, people know what they're getting
> >into before they contribute. And, as long as you continue to maintain the
> >Free branch (i.e. just as improvements continue to be made to the 2.0.x
> >kernel even though 2.1.x is being developed), I think you'll find people
> >more co-operative than you'd expect.
>
> This is pretty encouraging.
Also, I think a very important point would be to somehow guarantee that
the GPL license will always be available alongside with commercial one
-- to convince people that you do not intend to use their contributions
to develop version 1, then to commercial-only in the next release.
--
| Victor A. Danilchenko CSCF support |
| [EMAIL PROTECTED] A313, 5-4231 |
+--------------------------------------------+
| Quando omni flunkus, moritati. |
------------------------------
Crossposted-To: comp.os.linux.setup
Subject: Re: Slackware installation questions
From: [EMAIL PROTECTED] (Dave Brown)
Reply-To: [EMAIL PROTECTED]
Date: 29 Dec 98 18:36:52 GMT
In article <[EMAIL PROTECTED]>, [EMAIL PROTECTED] says...
>...
Slackware creates a directory called "packages" in /var/adm. In it are files
named for each package which list all the files that were extracted from the
packages installed. You might try using pkgtool (which creates/uses those
files in /var/adm/packages) to browse through the slakware/x1 tree of packages
looking for XF86Config. Or simply look for the "diskx1" file which has the
description of all the files in slakware/x1 on the cd.
--
Dave Brown Austin, TX
------------------------------
From: joey <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.hardware
Subject: Re: maxtor 17gig and rh 5.1
Date: Tue, 05 Jan 1999 09:59:58 -0500
I just thought I'd let you and the group know that this bug has not been taken
care of in win95 osr2 running fat16. I just ran a check by copying much more
than 2 gigs of stuff to the drive in question and the windows machines still
report 2 gigs of empty space along with a few very large data sets. I guess,
maybe, you have to be running fat 32 to see the true size of the samba
export. But, since there are currently no machines running fat 32, I can't
check this theory. Thanks for the information. It saved me lots of time
trying to figure out what was wrong with my Linux configuration when the
problem was "elsewhere." If anyone out there has a Samba directory larger
than 2 gigs exported to a win95/98 machine with fat32, drop me a line/email
and tell me if, when mounted as a network drive, the Samba directory is the
correct size......just curious.sendto<[EMAIL PROTECTED]>thanks
Chris wrote:
> On Mon, 04 Jan 1999 18:52:07 -0500, joey <[EMAIL PROTECTED]> wrote in
> comp.os.linux.hardware:
>
> >However, when I mount the drive
> >on a windows machine, the drive appears to be 2 gigs.
>
> This is a bug in the original Microsoft DOS/Windows filesystem code, which
> not only had the FAT16 filesystem size limit but also assumed that the
> network servers had the same limit.... It was not fixed until Win95-OSR2.
> Pre-OSR2 operating systems will allow you to read and write to the entire
> network filesystem, but will not report the remaining space correctly
> unless (until) there is less than 2Gb free.
------------------------------
From: [EMAIL PROTECTED] (Steve Lamb)
Subject: Re: Why is GNOME not called a window manager?
Reply-To: [EMAIL PROTECTED]
Date: 5 Jan 1999 09:22:51 +0800
On Tue, 5 Jan 1999 13:33:27 +0100, Matthias Warkus <[EMAIL PROTECTED]>
wrote:
>It's not a window manager because it's a suite of programs that altogether
>make up a desktop environment. These programs are mainly:
>- Gnome Panel (with panel applets)
>- Gnome Midnight Commander (file manager)
>- Gnome Control Center
>- ... (read: plethora of other programs)
And this is also why KDE isn't a window manager while KWM is. ;)
--
Steve C. Lamb | Opinions expressed by me are not my
http://www.calweb.com/~morpheus | employer's. They hired me for my
ICQ: 5107343 | skills and labor, not my opinions!
=======================================+=====================================
------------------------------
From: [EMAIL PROTECTED] (Chee Choon Cheng (remove "removethis" to e-mail me))
Subject: Re: RAMDISK in RedHat 5.2, what are them for? Can remove them?
Date: Tue, 05 Jan 1999 07:05:29 GMT
On 03 Jan 1999 21:19:43 +0000, James Youngman
<[EMAIL PROTECTED]> wrote:
>[EMAIL PROTECTED] (Choon-Cheng Chee (remove "removethis" in my e-mail))
>writes:
>
>> >No memory is used until you actually allocate the RAM disks (with
>> >mke2fs /dev/ram[x]).
>>
>> Ok.. Any idea what they are created for?
>
>The kernel you run is the same as the one used for the install; the
>ramdisk is used as the root device while installing Linux.
>
>--
I just install RH 5.2, and didn't touch the kernel since then. Does
that mean I need to redo the kernel to get rid of them? Do they
actually take up any memory (I didn't allocate the RAM disk)?
If they won't cause negative impact, I am just going to leave them
alone. 8)
Thx!
------------------------------
From: ben*nospam*@shell.bensykes.com (Ben Sykes)
Subject: Re: Whats the best *offline* usenet reader for Linux?
Date: 5 Jan 1999 09:10:17 GMT
Reply-To: [EMAIL PROTECTED]
On Tue, 05 Jan 1999 04:39:39 GMT, [EMAIL PROTECTED]
<[EMAIL PROTECTED]> wrote:
>None, you need to use a combination of local NNTP server + newsreader for the
>above task. If you're on a single user machine, you may try leafnode + your
>favourate newsreader (for me, I'm using slrn). Leafnode works just like
>FreeAgent.. For me, I'm getting my feeds thru' UUCP + NNTP (using 'suck') and
>running INND on my notebook and INND sucking 6% of my 64Meg RAM (did 'expire'
>everyday, otherwise it will take like 10% of my memory space).
Acutally, the latest SLRN does have a feature that you can compile in that
gives it an offline mode. I've never used it, but I'd be intersted to hear
from people that have.
--
Ben Sykes
Dallas, TX
------------------------------
From: Michael Steigman <[EMAIL PROTECTED]>
Subject: Re: gcc broken after egcs upgrade
Date: Tue, 05 Jan 1999 00:53:42 -0500
Steve Romero wrote:
>
> Michael Steigman wrote:
>
> > Chris Mauritz wrote:
> > >
> > > I think you want to reinstall:
> > >
> > > libstdc++-2.8.0-14
> > > libstdc++-devel-2.8.0-14
> > >
> > > Make sure you remove the egcs versions too. Also, if you switch to
> > > egcs, some programs like groff are hard linked to the gcc clibs and
> > > break. I had to recompile groff (so that man would work again) from
> > > the source rpm after installing egcs on Redhat 5.2 with 2.1.127 kernel.
> >
> > I had already tried reinstalling these two libs but I went ahead and
> > tried again. No luck. I've even gone back to the 5.2 libc (5.3.12.27).
> > BTW, everything worked just fine after I installed the newer libc - I
> > just wanted to be sure. As far as I can tell, all packages are stock 5.2
> > and should work. Is there a particular flag that I could invoke gcc with
> > to provide more info? I've also verified what I think are the relevant
> > rpms. Everything seems fine. Should I try to force-install ld.so again?
> > Any other things to check or try? Thanks.
>
> Just to let you know I also retried the libstdc files from the default
> install with no success. I set an e-mail (ticket) to RedHat but there
> response was that they'll only fix the original install, once installed your
> on your own. I have been able to compile again however by installing the
> GNUPro Toolkit. I know that this dosen't help you but it does indicate that
> something is shakey with the default 5.2 install.
I fixed the problem. It turns out that it had nothing to do with egcs
after all. I had found and installed a more recent version of binutils,
which was not playing nicely with gcc, I guess. I had totally forgotten
about that. A few minutes ago, I was checking out rpmfind.net
(rufus.w3.org/linux), which lays out the contents and dependencies of
each rpm. Well, when I saw that gcc required binutils that set off the
alarm. binutils actually supplies the assembler, loader and a bunch of
other tools that I had assumed the gcc package contained.
------------------------------
From: Michael Steigman <[EMAIL PROTECTED]>
Crossposted-To: linux.redhat.install
Subject: Re: Getting the C compiler working
Date: Tue, 05 Jan 1999 01:01:41 -0500
Paul Davies wrote:
>
> Hi There
>
> I've just installed RedLinux 5.2 and need to use the C compiler to rebuild
> Perl.
>
> The Configure script which runs the Perl build keeps saying the C compiler
> is not working. I've definitely installed the package because when I type
> cc or gcc at the command line it is recognised.
>
> Is there any other configuration I need to do to get it working?
>
> Any help would be much appreciated.
I just wrestled with this same problem. Say, you didn't happen to
upgrade your binutils package, did you? I upgraded egcs and a bunch of
other packages Saturday and found that I couldn't compile simple C stuff
anymore. Well, I was sure it had something to do with egcs. Nope. The
binutils package 2.9.1.0-19a does NOT work with the gcc which comes with
RH5.2. I forced the downgrade back to 2.9.1.0-15 and now everything
works again. If you've upgraded binutils, that could be your problem. If
not try rpm -V binutils. If there's no output, then you gotta keep
looking. BTW, the binutils package supplies the assembler, loader and
other stuff that I thought would be included in the gcc package.
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: comp.os.linux.setup,comp.os.linux.hardware
Subject: Modem vs. Sound card (RH5.2)
Date: 5 Jan 1999 17:30:21 GMT
[sorry for the crosspost, but the col* FAQ didn't make it clear
exactly where I should post this. Follow-ups set to col.misc]
I've been having problems getting my modem to work....
When I first installed RH5.2, I finally got my modem working by running
setserial to set the IRQ and making sure that the serial port speed in
minicom matched the speed reported by setserial.
Then my modem suddenly stopped working and continued to be dead after
repeated re-installs.
I think I know what my problem is, even if I'm not quite sure how to fix
it.
My modem (USR Sportster V.90 Data/Fax internal) is jumpered to COM3, IRQ5.
I finally thought to look in /proc/interrupts and found that my sound card
(SoundBlaster 16 PnP) had grabbed IRQ5! I'm pretty sure my modem stopped
working *after* I ran sndconfig.
So when I ran setserial to assign IRQ5 to /dev/cua2, minicom was sending
the AT commands to the sound card...
In my subsequent installation attempts, I ran sndconfig right away. In
my first attempt, I didn't do it for a couple of days.
So now... A couple of questions..
1) Where can I look to see *exactly* which resources (IRQs, IO ports and
DMA channels) linux has currently assigned to my sound card? Will
the pnpdump program do it?
I also see a strange message during boot up about a 'missing' DMA
channel for the sound card.
2) Is it possible to *UN*configure a sound card (w/o installing yet again)
to test my theory? All I really need to do is get IRQ5 back.
3) Can I use the PnP tools under linux to coerce the various resource
settings? Under WinBlows, my sound card sits on IRQ10.
4) Just curious... Does it matter which of the alternate designations
(/dev/(modem|ttyS2|cua2) I use to access the modem with setserial
and/or minicom serial port setup? Several HOW-TOs seem to deprecate
the use of /dev/cuax and /dev/modem for reasons I don't fully
understand. After running setserial against *any* of the alternatives,
*all* of the alternatives report the IRQ I set.
I'm in the midst of reading the Sound and PnP HOW-TOs and will get to the
various PnP man pages next, but a clue (or a least a hint that I can do
what I want to do in question 3) would be appreciated.
adTHANKSvance!
Simeon
P.S. Contrary to what you might think, my address
<[EMAIL PROTECTED]> is actually replyable. I use it and
procmail to trap some of the inevitable spam I get when I post
to usenet. I am perfectly content to get answers on the newsgroup
or by e-mail, but please don't do both.
------------------------------
From: [EMAIL PROTECTED] (Villy Kruse)
Subject: Re: RAMDISK in RedHat 5.2, what are them for? Can remove them?
Date: 5 Jan 1999 19:10:37 +0100
In article <[EMAIL PROTECTED]>,
Chee Choon Cheng (remove "removethis" to e-mail me) <[EMAIL PROTECTED]>
wrote:
>On 03 Jan 1999 21:19:43 +0000, James Youngman
><[EMAIL PROTECTED]> wrote:
>
>
>I just install RH 5.2, and didn't touch the kernel since then. Does
>that mean I need to redo the kernel to get rid of them? Do they
>actually take up any memory (I didn't allocate the RAM disk)?
>
The inital ram disk is only required to load the scsi driver before
mounting the root file system. If you don't have scsi you don't need
the inital ram disk and the install procedure won't make one.
In any case the space used by the initial ram disk will be released as
soon as the real root file system is mounted.
Villy
------------------------------
From: aldev <[EMAIL PROTECTED]>
Subject: Re: Help!! We are looking for a new database system.
Date: Tue, 05 Jan 1999 18:07:48 +0000
Please see the Microsoft-Linux Analogy list (under section Databases) at
Main site is at : http://members.spree.com/technology/aldev/ Mirror sites are at
http://aldev. 8m. com and http://aldev.webjump.com
and at http://homepages.infoseek.com/~aldev1/index.html
Joey Merrell wrote:
> Gleim Publications, Inc. (www.gleim.com) is a
> publisher of accounting and aviation test prep books,
> software, and audio?s in Gainesville, Florida. Gleim
> needs a new customer database/order entry/inventory
> control office system.
>
> Currently we have an SCO OpenServer 5 using
> Unify 2000 database and an Accell/SQL 4GL front-end
> developed in-house. We would very much like to
> abandon SCO and run the new database system on either
> RedHat 5.x (preferred) or NetWare 5.x. The front end
> must be 32-bit Windows compatible.
------------------------------
From: aldev <[EMAIL PROTECTED]>
Subject: Re: HELP! Lost mouse support in X!
Date: Tue, 05 Jan 1999 18:09:27 +0000
[EMAIL PROTECTED] wrote:
> My Linux box (RH5.2) is currently stuck in X without
> mouse support, and I do not currently have anything
> onscreen except the configurator box. How can I exit
> X? I can get to another virtual console, is there an
> elegant way to stop X from there?
>
> Thanks in advance
> One-more-newbie
Press CTRL+ALT+BACKSPACE keys all together to exit X-Session
------------------------------
From: [EMAIL PROTECTED] (Villy Kruse)
Subject: Re: 2038 and Linux
Date: 5 Jan 1999 19:15:32 +0100
In article <76t9dk$5vk$[EMAIL PROTECTED]>,
J.H.M. Dassen (Ray) <[EMAIL PROTECTED]> wrote:
>
>Linux has been a 64 bit kernel on Alpha for a long time, and IIRC has become
>64 bits on Sparc prior to the availability of a 64 bit version of
>Solaris/Sparc.
>
Can you even make an OS for alpha that is not 64bit. In other words:
anything running on alpha is 64bit; you don't have a choice.
Villy
------------------------------
From: Yukinobu Kutsukake <[EMAIL PROTECTED]>
Subject: Netscape 4.08 crashes when running java (Slackware-3.6)
Date: Wed, 06 Jan 1999 03:17:51 +0900
Netscape Communicator 4.08 crashes (Bus Error) when running Java Applet.
I've removed ~/.netscape directory, but above error still occurs.
My environments are following:
- CPU: Intel Pentium 120MHz
- Memory: RAM 48M, SWAP 96M
- slackware 3.6
- Windows Manager: fvwm95-2
- XWindow : XF86_SVGA (XFree86.3.3.2)
/usr/X11R6/lib/libXt.so.6.0
/usr/X11R6/lib/libSM.so.6.0
/usr/X11R6/lib/libICE.so.6.3
/usr/X11R6/lib/libXmu.so.6.0
/usr/X11R6/lib/libXpm.so.4.11
/usr/X11R6/lib/libXext.so.6.3
/usr/X11R6/lib/libX11.so.6.1
- /lib/libdl.so.1.9.9
- /lib/libc.so.5.4.46
- /usr/lib/libstdc++.so.2.8 (linked by libg++.so.27, libstdc++.so.27)
- /lib/libm.so.5.0.9
When running Netscape on XF86_S3 X server, it runs with java applet.
I think XF86_SVGA server is related to this probram.
Please tell me any information for running Java Applet.
Thanks
--
Yukinobu Kutsukake
email: [EMAIL PROTECTED]
------------------------------
From: Matthew Kirkwood <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.x,uk.comp.os.linux
Subject: Re: Anti-Linux FUD
Date: Tue, 5 Jan 1999 18:08:31 +0000
On 5 Jan 1999, David Damerell wrote:
> >Having sbin directories in a normal users path is not required in
> >order for a normal user to know the system's IP address.
>
> What's a better way to do it than 'ifconfig', then?
host `hostname`
?
> > However, I would question that a normal user needs to know it
> > anyway... None of the examples you give require sbin in the path.
>
> mkdosfs is in an sbin directory.
But mformat isn't.
> >>[By this, I refer to someone who obviously could have root if they wanted
> >>it, but is a housemate or family member or suchlike who doesn't want or
> >>need it; such a person may have a user account with privileges which
> >>equate to being root.]
> >But that is NOT a "normal" user
>
> On a home system - where an increasing proportion of Linux users are -
> that is a normal user. _Anyone_ with physical access to the box can give
> themselves rootly powers if they want to (no, setting the boot order to
> C,A and a BIOS password is not enough); any housemate or relative is just
> as I described above, and it may well be 'normal' for these normal users
> to have floppy access, since they must neccessarily be trusted anyway.
The "user" option in fstab (and suid-root mount in general) are root
equivalent, as far as I'm concerned.
And if your users use mtools for floppy access, then they can use mformat
to make a floppy filesystem...
In my opinion, an automounter (together with something like supermount,
perhaps) is the answer to such things..
Matthew
- not quite sure who he's disgreeing with.. :)
------------------------------
From: Zdravko Balorda <[EMAIL PROTECTED]>
Subject: which Xserver?
Date: 5 Jan 1999 08:50:00 GMT
Which X server should I use for
ATI RAGE IIC AGP
graphic card?
------------------------------
From: [EMAIL PROTECTED] (Alexander Viro)
Subject: Re: Netscape eats up *all* the swap
Date: 5 Jan 1999 13:30:22 -0500
In article <76onp1$[EMAIL PROTECTED]>,
Floyd Davidson <[EMAIL PROTECTED]> wrote:
[snip]
>>I've found all versions of Netscape to be pigs. :(
>
>A total of 46Mb of virtual memory is far too little. Doubling
>both the RAM an the swap would be a good idea. If nothing else,
>at least up the swap to provide something more than 70-80Mb of
>virtual memory.
Hmm... If that doesn't shout 'pig'... BTW, what are reasons to
use it? I'm really curious and it's not a trolling. For what
I've seen the thing is memory-,CPU- and IO-thirsty, makes a
poorest newsreader (weaker than telnet new.foo.org nntp and
about equally efficient), allows HTML in email and news,
is buggy as hell (wake me up when lynx will die with SIGBUS)...
It's, erm, editor is dumber than even pico. It looks like
a fundamentally non-UNIXish program - monolitic, heavy-weight,
inflexible... What makes people use the sucker?
--
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid. Get yourself a better computer" - Dilbert.
------------------------------
From: Iain Georgeson <[EMAIL PROTECTED]>
Crossposted-To:
alt.destroy.microsoft,comp.os.linux.advocacy,comp.os.os2.advocacy,gnu.misc.discuss,uk.comp.os.linux
Subject: Re: Anti-Linux FUD
Date: Tue, 5 Jan 1999 18:30:02 +0000
In article <76btr5$c21$[EMAIL PROTECTED]>, Roger Espel Llima
<[EMAIL PROTECTED]> writes
>In article <76bj52$f0r$[EMAIL PROTECTED]>, Tim Smith <[EMAIL PROTECTED]> wrote:
>>Chris Hanson <[EMAIL PROTECTED]> wrote:
>>>What he's saying is not "X should support copy and paste
>>>functionality" -- I think we all know it already does. Rather, he's
>>>suggesting -- based on not only his experience but on the work of
>>>many, many human factors professionals -- that X-based interfaces for
>>>Unix have historically fallen down because there's no standard way of
>>>at that copy & paste functionality.
>>Worse, many (most?) X apps don't provide a way for you to select text without
>>copying it.
>
>what is "worse" about that? X's behaviour doesn't even have the notion
>of "copying" : you select with a button, and drop with another. it could
>hardly be easier, simpler or more practical.
Actually, I agree...partly. There is one situation where I've wanted to
got round this.
Find useful URL in text somewhere.
Select it.
Wipe URL currently in Netscape's location box by selecting it, and
pasting the first one on top.
Oops...
OK, there are simple ways round it, and 9 times out of 10, it D exactly
WIM, but occasionally I'd like a "select without copying" feature. Or is
there one already?
>when, after learning X's way, I first encountered a GUI that made me
>specifically copy the selected text, I wondered what on earth the point
>was. and I still do, for that matter.
<AOL>Me too.</AOL>
Iain.
--
#!/usr/bin/perl -w
$FOO="$ENV{HOME}/.signature";unless(-p$FOO){unlink$FOO;system('mknod',
$FOO,'p')&&die"mknod on fire: $!"}while(1){open(FOO,"> $FOO")||die
"open on fire $FOO: $!";print FOO`cat $0`;close FOO;sleep 2}
------------------------------
From: [EMAIL PROTECTED] (Alexander Viro)
Subject: Re: Netscape eats up *all* the swap
Date: 5 Jan 1999 13:37:56 -0500
In article <[EMAIL PROTECTED]>,
Philip Brown <[EMAIL PROTECTED]> wrote:
>On Sun, 03 Jan 1999 07:11:47 GMT, [EMAIL PROTECTED] wrote:
>>....
>>I have 16 megs of real RAM and about 30 megs on my swap partition.
>
>reguardless of what you decide to run, this is a problem. These days, you
>should always have at least 64 megs of swap space.
Really? Two big compiles in parallel, heavy number-crunching, TeX, editing.
16M RAM + 32M swap. Not even close to full. Oh, and diff between two large
trees piped to bzip2. Also not the lightest thing in this world. Processor?
K5-75. Are you sure that you are in the right group? Sounds like you are
either using 4.4-derived kernel (fine, but Linux VM differs from 4.4 and
takes less swap) or, excuse me, are running Windows.
--
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid. Get yourself a better computer" - Dilbert.
------------------------------
From: ag <[EMAIL PROTECTED]>
Subject: Re: Configure Kernel for Oracle?
Date: Tue, 05 Jan 1999 11:12:32 -0600
Reply-To: [EMAIL PROTECTED]
>
> SHMMAX 4294967295
> SHMMIN 1
> SHMMNI 100
> SHMSEG 10
> SEMMNS 200
> SEMMNI 70
> SEMMSL equal to or greater than the value of the PROCESS
> initialization parameter
You will need to reconfigure some of the kernel parms and recompile the
kernel. The main file to modify is
/usr/src/linux/include/asm-i386/shmparam.h. Set the:
SHM_IDX_BITS = 17
SHMMAX = 0x10000000
Make sure your other specifications (SHMMNI, etc, are >= values
specified). I tried recursive grepping but I wasn't able to locate the
"PROCESS" keyword. I think this parameter must be located somewhere
else - perhaps in initsid.ora.
Please post your experiences..........
Andrew
------------------------------
** 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
******************************
