Linux-Setup Digest #151, Volume #19 Thu, 13 Jul 00 17:13:13 EDT
Contents:
Second network adapter ("AJ-San")
Re: Setup Linux with lilo/grub (Oktay Altunergil)
Re: Corrupt Partition Table (Oktay Altunergil)
problem with mouse (David Punsalan)
etc/rc.d stuff doesn't start at boot time ([EMAIL PROTECTED])
Re: Linux beyond 1024 Cylinder
Re: Stupid question?? VFAT partitions (D G)
Help! Can't open term session under X (Romano)
Re: lucent winmodem in laptop (D G)
Re: ipchains (D G)
how to put named in rc scripts (RH6.2)? (Noah Green)
PPP problem - chat aborts saying the serial link is not 8-bit clean (Jonas De Vuyst)
Re: Using NT bootload to boot to linux (D G)
X problems with Dell Optiplex GX110 (Bill Bannan)
Hard drive performance (Keith Rhodes)
----------------------------------------------------------------------------
From: "AJ-San" <[EMAIL PROTECTED]>
Subject: Second network adapter
Date: Thu, 13 Jul 2000 21:03:35 +0200
I'm a newbie, and I'm also new here, so if I should post this elsewhere,
please
tell me!
I've just successfully (!) installed my Slackware 7 Distribution on my
P120/32MB, and so far it's running smoothly.
In general: Where do I go from here?
More specific:
I have one NE2000 (PCI) network adapter, which is working fine.
The second one a 3COM 509 (ISA) is not.
How do I configure things so that I can use them both (I want use it as a
gateway/proxy/firewall between my cable modem and my PIII-550)
I know I should be more specific, but I'm just a beginner and I don't know
much about Linux, so if somebody could help me out here ....
Greetz,
Arjo
------------------------------
From: Oktay Altunergil <[EMAIL PROTECTED]>
Subject: Re: Setup Linux with lilo/grub
Date: Thu, 13 Jul 2000 19:30:05 GMT
Hi,
As far as I know, every distribution allows that. You might just have to do
a more customized installation. When Mandrake asks you to create a boot
floppy, say 'yes' and create the boot floppy which you will use for booting
up. At another point it will ask you whether you want to install grub/lilo
or leave the current settings alone. Choose not to install grub/lilo. This
will allow you to not have the boot loader installed and you will be able
to boot with your boot disk.
Another way to work considering you already have a working distribution
that utilizes lilo/grub, is to get rid of lilo/grub. First of all try
booting with your boot floppy to make sure it works. (it would be too bad
to find out after removing lilo that your boot floppy is damaged). If it
does work, make another copy of it and keep it in a safe place. (Since you
won't be having another way to boot linux -though you might be able to
using CD-ROM boot- you don't want to take chances) If you never had a boot
floppy, create one using the mkbootdisk command. (or something similar, I
don't remember). Apply the above steps for trying and making another copy.
After you are absolutely sure that you CAN boot linux using the floppy you
have, start DOS and type 'fdisk /mbr' . This will create a clean boot
sector which in effect removes lilo and grub. (This is exactly what I had
to do today because of a non-working grub installation).
Hope this helps.
Oktay
Visit the Free Linux CD Project @ http://freelinuxcd.org
SY wrote:
>
>
> Hi,
>
> I was wondering, is there a way of setting up Linux on a machine which
> already has win98 without installing lilo / grub etc (i.e. the only
> way to boot into linux would be via a bootdisk?
> Which distributions allow this (I know that Mandrake 7.0 does not give
> you the option).
>
> Thanks in advance.
--
Posted via CNET Help.com
http://www.help.com/
------------------------------
From: Oktay Altunergil <[EMAIL PROTECTED]>
Subject: Re: Corrupt Partition Table
Date: Thu, 13 Jul 2000 19:30:06 GMT
Mandrake explicitly states that there might be 'data corruption' due to
hard drive optimizations. I don't know if same applies for 'corrupt
partition tables' but you might try not enabling the hard drive
optimizations. (I never had a problem with that though)
Oktay
towster wrote:
>
>
> Just to share..
> Used linux since redhat 4.x and I have never seen this..
> On my laptop i have a linux partiton that I have installed
> *MANY* flavors just to test them.. Installed mandrake 7.1
> beta (2 i think) without problems.. (cept application level)
> 7.1 final comes out and i snag it and install it only to find
> out about an hour later of playing with it my partition table
> was corrupt.
> Good friend of mine installed 7.1 on his home computer and it
> worked fine for about 3 weeks.. BAM.. partition table corrupt..
> I think that there may be some issue with 7.1 final but I am
> unsure of what it is..
>
> I have compared some notes tho.. Both my friend and I during the
> installation picked expert installation, hard drive optimizations
> and grub (instead of lilo).
>
> If *ANYONE* has had this same problem maybe you should also post
> it here so that we can all avoid these nasty corrupt partitions
> or give some clue as to where the problem lies..
>
> Robert Towster
> ps. Dont email me at the posted email address..
> my last name at technologist dot com would be the place..
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
--
Posted via CNET Help.com
http://www.help.com/
------------------------------
From: David Punsalan <[EMAIL PROTECTED]>
Subject: problem with mouse
Date: Thu, 13 Jul 2000 14:48:16 -0500
Hi,
I installed Debian 2.1 Slink onto an old packard bell 486DX 50MHz with
16MB of ram. Overall, I'm pretty happy with it. When not running X, it
works beautifully. THe installation went very smoothly except for one
thing: the mouse. I tried installing the psaux module, but it didn't
help.
Symptom: mouse is unresponsive in all circumstances.
Details:
When I start X, the screen appears frozen. I can ctl-alt- +, - and
backspace, but nothing else.
I can tell that X and the window manager seems fine other than the mouse
(i.e. the image looks fine, not fuzzy, grainy or anything like that, no
"streaks" or dots).
When I run XF86Setup, graphics mode works, keyboard input works, but not
the mouse.
I must indicate that though it is an old computer, the inputs for the
keyboard and mouse are slim and circular, like the ones popular today. (as
opposed to the keyboards with relatively larger diameter plugs and mouses
(mice?) with serial port like connections). I'm not sure what the
technical term is. The mouse is an old three-button IBM mouse.
When running xf86config, I have tried specifying all of the following to
no avail:
1. PS/2 mouse
2. Microsoft mouse (two button)
3. Bus Mouse (? I think that's what it said, not sure though)
In all cases, when asked if I want to emulate three buttons, I answer "no"
b/c it seems that if I have a 3-button mouse I shouldn't have to emulate
it.
Any suggestions on how I can get my mouse working?
I would
greatly appreciate it.
Thanks,
David
------------------------------
From: [EMAIL PROTECTED]
Subject: etc/rc.d stuff doesn't start at boot time
Date: Thu, 13 Jul 2000 20:18:46 GMT
ls -l rc*/*pcmcia*
lrwxrwxrwx 1 root root 16 Jan 15 07:27 rc0.d/K96pcmcia
-> ../init.
d/pcmcia
lrwxrwxrwx 1 root root 16 Jan 15 07:27 rc1.d/K96pcmcia
-> ../init.
d/pcmcia
lrwxrwxrwx 1 root root 16 Jan 15 07:27 rc2.d/S45pcmcia
-> ../init.
d/pcmcia
lrwxrwxrwx 1 root root 23 Jan 15 16:29 rc3.d/S09pcmcia
-> /etc/rc.
d/init.d/pcmcia
lrwxrwxrwx 1 root root 16 Jan 15 07:54 rc3.d/k96pcmcia
-> ../init.
d/pcmcia
lrwxrwxrwx 1 root root 16 Jan 15 07:27 rc4.d/S45pcmcia
-> ../init.
d/pcmcia
lrwxrwxrwx 1 root root 16 Jan 15 07:54 rc5.d/k96pcmcia
-> ../init.
d/pcmcia
lrwxrwxrwx 1 root root 16 Jan 15 07:27 rc6.d/K96pcmcia
-> ../init.
d/pcmcia
[root@tosh rc.d]# ls -l rc*/*network*
lrwxrwxrwx 1 root root 17 Jan 15 07:14 rc0.d/K90network
-> ../init
.d/network
lrwxrwxrwx 1 root root 17 Jan 15 07:14 rc1.d/K90network
-> ../init
.d/network
lrwxrwxrwx 1 root root 17 Jan 15 07:14 rc2.d/S10network
-> ../init
.d/network
lrwxrwxrwx 1 root root 17 Jan 15 16:28 rc3.d/S10network
-> ../init
.d/network
lrwxrwxrwx 1 root root 17 Jan 15 07:14 rc4.d/S10network
-> ../init
.d/network
lrwxrwxrwx 1 root root 17 Jan 15 07:54 rc5.d/S10network
-> ../init
.d/network
lrwxrwxrwx 1 root root 17 Jan 15 07:14 rc6.d/K90network
-> ../init
.d/network
But yet I still have to type:
cd init.d/
[root@tosh init.d]# ./pcmcia start
Starting PCMCIA services: modules cardmgr.
[root@tosh init.d]# ./network start
Bringing up interface lo [ OK ]
Bringing up interface eth0 [ OK ]
I have to type it everytime I boot my laptop.
Any ideas on a way to debug this?
I haven't had any luck at all getting /etc/rc.d/rc*.d/S*
stuff to run that I have added.
But the stuff from installation off the CD seems to work ok??
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: <[EMAIL PROTECTED]>
Subject: Re: Linux beyond 1024 Cylinder
Date: Thu, 13 Jul 2000 20:30:03 GMT
Thad Guidry wrote:
>
Redhat 6.2 dows not come with the latest lilo, but you can download it(latest lilo)
and install it fairly easily. Till you do that, just use the boot disk made while
installing...
> Bart Kooijman wrote:
> >
> >
> > Thank you for your Reply.
> >
> > What version of Red Hat Linux comes with this latest relaese of Lilo?
> Does
> > RH 6.0 have it?
> >
> > Thanks in advance,
> >
> > Bart Kooijman
> >
> > "Ujwal Sathyam" <[EMAIL PROTECTED]> wrote in message
> > news:[EMAIL PROTECTED]...
> > >
> > > As long as your BIOS supports booting from beyond the 1024 cylinder you
> > > should be OK. Older versions of the Linux loader lilo also had a
> problem
> > > with the 1024 limit. The latest version of lilo fixes that.
> > >
> > > Ujwal
> >
> >
>
> I can tell you that Red Hat 6.2 does indeed run into the 1024 cylinder
> problem. I should know since I tried to install it and only got "LI" to
> come up, which means there's a geometry mismatch problem. For more
> information on this problem visit the LILO Mini HowTo Documentation
> available at http://www.linuxdoc.org/HOWTO/mini/LILO.html
>
> Also the Linux System Administrators Guide has a section on booting at
> http://www.linuxdoc.org/LDP/sag/
>
> --
> Posted via CNET Help.com
> http://www.help.com/
--
Posted via CNET Help.com
http://www.help.com/
------------------------------
From: D G <[EMAIL PROTECTED]>
Subject: Re: Stupid question?? VFAT partitions
Date: Thu, 13 Jul 2000 13:34:04 -0700
Michael Nadler wrote:
>
> I assume that you are getting your mail via POP. Can you use IMAP? Is
> there an IMAP mailserver available?
>
> I ran into this same problem, and solved it by leaving my mail on the IMAP
> server, then being able to read it via Netscape on whatever system I
> booted. I also used Netscape's IMAP from other platforms, such as HP/UX.
This sounds like a good solution.
--
DG
e-mail is: [EMAIL PROTECTED]
(remove the Z's--they're what I do when I read SPAM!)
------------------------------
From: Romano <[EMAIL PROTECTED]>
Subject: Help! Can't open term session under X
Date: Thu, 13 Jul 2000 22:03:31 +0200
HI all!
I have a problem with my linux RedHat 6.1 system that I can't handle
If I try to open a terminal session under X-Windows and (being logged as
normal user) Gnome desktop environment I get the following error
message:
pt_chown: needs to be installed setuid 'root'
This problem do no not appear if I log in as root
PLEASE HELP ME!! I can't solve this by myself!!!!
THANK YOU ALL
------------------------------
From: D G <[EMAIL PROTECTED]>
Subject: Re: lucent winmodem in laptop
Date: Thu, 13 Jul 2000 13:40:23 -0700
Zebee Johnstone wrote:
>
> I have an Acer Travelmate laptop that appears to have a lucent
> winmodem in it.
>
> W98 manages fine.
>
> I got the binary driver via linmodem.org.
>
> I did the steps in the ltinist by hand.
>
> I'm running a 2.2.12-20 kernel.
>
> If I do /sbin/depmod -a it complains about unresolved libraries.
Normal.
>
> If I force an insmod, the driver appears to load.
'/sbin/lsmod' will tell you.
>
> I used mknod to make /dev/ttyS14, but if I then try to access it via
> minicom, minicom crashes (as in it quits) with "device does not
> exist".
Are you sure you did mknod with the appropriate numbers? Is your
/dev/modem symlinked to /dev/ttyS14?
>
> Where can I go from here?
>
> Wait till lucent gets it together a bit more? Or is there some other
> trick?
>
> Zebee
> - posting from another machine, so not cutting and pasting, can do so
> if more info required
Do 'cat /proc/pci' and make sure the entry under "Lucent Technologies
Modem" (or whatever it is) has an assigned IRQ. If not, you'll have to
set "plug-n-play=no" in your BIOS setup screen--this worked for me. If
this still doesn't work, you might try the pci tools, although they
didn't work for me.
--
DG
e-mail is: [EMAIL PROTECTED]
(remove the Z's--they're what I do when I read SPAM!)
------------------------------
From: D G <[EMAIL PROTECTED]>
Subject: Re: ipchains
Date: Thu, 13 Jul 2000 13:44:42 -0700
"Jeff D. Hamann" wrote:
>
> I would like to set up a firewall on my linux box and I'm a little confused
> regarding the use of ipchains, is there a better explaination of firewalls
> and using ipchains somewhere?
Read the ipchains howto at www.linuxdoc.org.
While you're at it, you might want to read the networking, firewall, and
ip masquerade how-tos as well.
--
DG
e-mail is: [EMAIL PROTECTED]
(remove the Z's--they're what I do when I read SPAM!)
------------------------------
From: Noah Green <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.networking
Subject: how to put named in rc scripts (RH6.2)?
Date: Thu, 13 Jul 2000 16:50:56 -0400
Hi,
I am sorry to bother you all I am pretty new at this and am at my wit's
end.
When my linux box was set up, it did not have named in the rc startup
scripts. Now I have finally set up DNS, but cannot figure out how to get
the named command into the startup scripts! Every time a start the
machine
I have to manually call named myself. I know how to do this in Solaris
but
not in Linux . . . any help?
Also, this box has the RH ppp dialer on it. Whenever I run it, it takes
the
local DNS server address out of the resolv.conf file. Is there a way to
prevent this?
Thanks in advance for all help!
noah
[EMAIL PROTECTED]
------------------------------
From: Jonas De Vuyst <[EMAIL PROTECTED]>
Subject: PPP problem - chat aborts saying the serial link is not 8-bit clean
Date: Thu, 13 Jul 2000 22:52:22 +0200
Hey all,
I've been having a problem getting my modem to dial in. I'm currently using a
package called PPPKit-0.7.tar.gz. Before this I've tried to use the example
script of the LFS book, I've also tried the script described in the Belgian
HOWTO because these two seem to fit my situation. I had no luck, however, with
both of these scripts.
The package started working a bit better after I changed "ogin" to "sername" in
/usr/sbin/pppdialer. I still don't get past the complete login procedure but at
least my identity gets confirmed.
The error I get seems to be about bit 7 being 0, wrong parity (sp?) is all I
can
think off but that seems unlikely.
Can someone give me a clue? I'd like to avoid PAP (which is what my ISP
normally
uses), is this possible?
Thank you,
Jonas De Vuyst
BTW: Below are the last lines of /var/log/sys.log that show the whole
connection
procedure, of course MyUserName and MyPassword are not actually what chat
printed.
--- Begin /var/log/sys.log ---
Jul 13 22:53:16 rome pppd[105]: pppd 2.3.11 started by root, uid 0
Jul 13 22:53:17 rome chat[106]: timeout set to 3 seconds
Jul 13 22:53:17 rome chat[106]: abort on (\nBUSY\r)
Jul 13 22:53:17 rome chat[106]: abort on (\nNO ANSWER\r)
Jul 13 22:53:17 rome chat[106]: abort on (\nRINGING\r\n\r\nRINGING\r)
Jul 13 22:53:17 rome chat[106]: send (rAT^M)
Jul 13 22:53:18 rome chat[106]: expect (OK)
Jul 13 22:53:18 rome chat[106]: rAT^M^M
Jul 13 22:53:18 rome chat[106]: OK
Jul 13 22:53:18 rome chat[106]: -- got it
Jul 13 22:53:18 rome chat[106]: send (ATH0^M)
Jul 13 22:53:18 rome chat[106]: timeout set to 30 seconds
Jul 13 22:53:18 rome chat[106]: expect (OK)
Jul 13 22:53:18 rome chat[106]: ^M
Jul 13 22:53:18 rome chat[106]: ATH0^M^M
Jul 13 22:53:18 rome chat[106]: OK
Jul 13 22:53:18 rome chat[106]: -- got it
Jul 13 22:53:18 rome chat[106]: send (ATDT092700420^M)
Jul 13 22:53:18 rome chat[106]: expect (CONNECT)
Jul 13 22:53:18 rome chat[106]: ^M
Jul 13 22:53:44 rome chat[106]: ATDT092700420^M^M
Jul 13 22:53:44 rome chat[106]: COMPRESSION: V.42BIS^M
Jul 13 22:53:44 rome chat[106]: ^M
Jul 13 22:53:44 rome chat[106]: CONNECT
Jul 13 22:53:44 rome chat[106]: -- got it
Jul 13 22:53:44 rome chat[106]: send (^M)
Jul 13 22:53:44 rome chat[106]: expect (sername:)
Jul 13 22:53:44 rome chat[106]: 115200/ARQ^M
Jul 13 22:53:45 rome chat[106]: ^M
Jul 13 22:53:45 rome chat[106]: ^M
Jul 13 22:53:45 rome chat[106]: User Access Verification^M
Jul 13 22:53:45 rome chat[106]: ^M
Jul 13 22:53:45 rome chat[106]: Username:
Jul 13 22:53:45 rome chat[106]: -- got it
Jul 13 22:53:45 rome chat[106]: send (MyUserName^M)
Jul 13 22:53:45 rome chat[106]: expect (assword:)
Jul 13 22:53:45 rome chat[106]: MyUserName^M
Jul 13 22:53:45 rome chat[106]: Password:
Jul 13 22:53:45 rome chat[106]: -- got it
Jul 13 22:53:45 rome chat[106]: send (MyPassword^M)
Jul 13 22:53:45 rome pppd[105]: Serial connection established.
Jul 13 22:53:45 rome pppd[105]: Using interface ppp0
Jul 13 22:53:45 rome pppd[105]: Connect: ppp0 <--> /dev/ttyS1
Jul 13 22:53:46 rome pppd[105]: sent [LCP ConfReq id=0x1 <asyncmap 0x20a0000>
<magic 0xd3e2197c> <pcomp> <accomp>]
Jul 13 22:54:13 rome last message repeated 9 times
Jul 13 22:54:16 rome pppd[105]: LCP: timeout sending Config-Requests
Jul 13 22:54:16 rome pppd[105]: Connection terminated.
Jul 13 22:54:16 rome pppd[105]: Receive serial link is not 8-bit clean:
Jul 13 22:54:16 rome pppd[105]: Problem: all had bit 7 set to 0
Jul 13 22:54:16 rome pppd[105]: Hangup (SIGHUP)
Jul 13 22:54:16 rome pppd[105]: Exit.
--- End /var/log/sys.log ---
------------------------------
From: D G <[EMAIL PROTECTED]>
Subject: Re: Using NT bootload to boot to linux
Date: Thu, 13 Jul 2000 13:53:11 -0700
tvn wrote:
>
> I know this has been asked alot of time ... however, it doesn't work out in
> my situation , can't figure out what's wrong with it .. please help.
>
> My Redhat 6.2 was installed without no lilo.conf in the /etc directory
> (because I choose not to have lilo installed, I'll make it later)
>
> Linux was installed on the 1st partition of the 2nd HD , HDB1
>
> After RH was installed, I use the command rdev /boot/vmlinuz /dev/hdb1
> (just to make sure)
>
> and then created a lino.conf in /etc like this
>
> -----------------------------------
> boot = /dev/hdb1
> compact
> install = /boot/boot.b
> map = /boot/map
>
> image = /boot/vmlinuz
> label = linux
> root = /dev/hdb1
> -------------------------------
> and ran /sbin/lilo after that
You're sure that your boot and root are the same partition?
Your lilo.conf might not be setup correctly for RedHat. Here is mine (I
have RedHat 6.1 and use NT boot loader):
=========
boot=/dev/hdb1
map=/boot/map
install=/boot/boot.b
prompt
timeout=50
default=linux
image=/boot/vmlinuz-2.2.12-20
label=linux
initrd=/boot/initrd-2.2.12-20.img
read-only
root=/dev/hdb8
vga=extended
==========
You probably need that initrd and maybe the read-only in there.
--
DG
e-mail is: [EMAIL PROTECTED]
(remove the Z's--they're what I do when I read SPAM!)
------------------------------
From: Bill Bannan <[EMAIL PROTECTED]>
Subject: X problems with Dell Optiplex GX110
Date: Thu, 13 Jul 2000 14:02:59 -0400
I have been unable to get X to work on the Optiplex GX110.
The Graphics card has an Intel 82810E Chip. I tried a variety
of Generic settings and none have worked. Has anyone out there
had any success configuring X on the Optiplex GX110 ?
Please respond to via Email
Bill Bannan
[EMAIL PROTECTED]
Gloucester MA
------------------------------
From: Keith Rhodes <[EMAIL PROTECTED]>
Subject: Hard drive performance
Date: Thu, 13 Jul 2000 23:02:46 +0200
I'd like some advice on tuning my hard drive parameters.
Before I start, I know I could get a boost by putting the
two hard drives on separate controllers (they'r currently
both on ide0), but I might want ide1 for something other
than my CD-Rom drive.....
This is a dual celeron (400MHz overclocked to 540MHz, cool
and stable even when the ambient temperature is around 27=B0C
and running PovRay + Netscape + Gnome / Sawmill).
Mobo is an Abit BP6, so I've got two more ide connectors,
supposedly for UDMA66 drives, but is that reliable?
# uname -a
Linux localhost.localdomain 2.2.14-15mdksecure #1 SMP Tue
Jan 4 21:15:44 CET 2000 i686 unknown
Before modification of parameters:
# hdparm -t /dev/hda
/dev/hda:
Timing buffered disk reads: 64 MB in 19.97 seconds =3D 3.20
MB/sec
# hdparm -t /dev/hdb
/dev/hdb:
Timing buffered disk reads: 64 MB in 29.48 seconds =3D 2.17
MB/sec
Pretty s l o w ...... I agree
============================================================
# cat hda/model
ST36422A This is a Seagate 6.4GByte
# hdparm -i /dev/hda
/dev/hda:
Model=3DST36422A, FwRev=3D3.02, SerialNo=3D7AC03H0R
Config=3D{ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs
RotSpdTol>.5% }
RawCHS=3D13228/15/63, TrkSize=3D0, SectSize=3D0, ECCbytes=3D0
BuffType=3D0(?), BuffSize=3D256kB, MaxMultSect=3D16, MultSect=3Doff
DblWordIO=3Dno, maxPIO=3D2(fast), DMA=3Dyes, maxDMA=3D2(fast)
CurCHS=3D13228/15/63, CurSects=3D12500460, LBA=3Dyes,
LBAsects=3D12500460
tDMA=3D{min:120,rec:120}, DMA modes: mword0 mword1 mword2
IORDY=3Don/off, tPIO=3D{min:240,w/IORDY:120}, PIO modes: mode3
mode4
UDMA modes: mode0 mode1 *mode2
Drive Supports : ATA-1 ATA-2 ATA-3 ATA-4
==============================================
# cat hdb/model
WDC AC32500H This is a Western Digital Caviar 2.5GByte
# hdparm -i /dev/hdb
/dev/hdb:
Model=3DWDC AC32500H, FwRev=3D10.07H09,
SerialNo=3DWD-WT3600419255
Config=3D{ HardSect NotMFM HdSw>15uSec SpinMotCtl Fixed
DTR>5Mbs FmtGapReq }
RawCHS=3D4960/16/63, TrkSize=3D57600, SectSize=3D600, ECCbytes=3D22
BuffType=3D3(DualPortCache), BuffSize=3D128kB, MaxMultSect=3D16,
MultSect=3Doff
DblWordIO=3Dno, maxPIO=3D2(fast), DMA=3Dyes, maxDMA=3D0(slow)
CurCHS=3D4960/16/63, CurSects=3D4999680, LBA=3Dyes,
LBAsects=3D4999680
tDMA=3D{min:120,rec:120}, DMA modes: mword0 mword1 *mword2
IORDY=3Don/off, tPIO=3D{min:160,w/IORDY:120}, PIO modes: mode3
mode4
The man page for hdparm doesn't mention the asterisk before
mword2, and the explanations of some of the other values
returned are somewhat difficult to undertand.....
And how can I understad things like the contents of
/proc/ide/hda/settings, which tells me:
pio_mode write-only 0
255 w
======================================================
Now, I read on http://www.linux.com/ that you can get b i
g improvements like this:
~4MB/sec increased to ~18MB/sec i.e. an increase of
350%
~4MB/sec increased to ~24MB/sec i.e. an increase of
500%
by using hdparm -c1 -d1 /dev/hd?
WOW!!
I gave it a try...
# hdparm -c1 -d1 /dev/hdb
/dev/hdb:
setting 32-bit I/O support flag to 1
setting using_dma to 1 (on)
I/O support =3D 1 (32-bit)
using_dma =3D 1 (on)
[root@localhost ide]# hdparm -t /dev/hdb
/dev/hdb:
Timing buffered disk reads: 64 MB in 21.26 seconds =3D 3.01
MB/sec
Dissapointing, 2.17 to 3.01 is o n l y a 38.7% increase
# hdparm -c1 -d1 -X11 /dev/hdb
/dev/hdb:
setting 32-bit I/O support flag to 1
setting using_dma to 1 (on)
setting xfermode to 11 (PIO flow control mode3)
I/O support =3D 1 (32-bit)
using_dma =3D 1 (on)
# hdparm -t /dev/hdb
/dev/hdb:
Timing buffered disk reads: 64 MB in 20.05 seconds =3D 3.19
MB/sec
A little better, but still only a 47% increase; nowhere near
the 350% !!
# hdparm -c1 -d1 /dev/hda
/dev/hda:
setting 32-bit I/O support flag to 1
setting using_dma to 1 (on)
I/O support =3D 1 (32-bit)
using_dma =3D 1 (on)
[root@localhost ide]# hdparm -t /dev/hda
/dev/hda:
Timing buffered disk reads: 64 MB in 18.34 seconds =3D 3.49
MB/sec
Dissapointing, 2.17 to 3.01 is o n l y a 9% increase
Now, I'd like some advice, rather than taking shots in the
dark, and risking corrupting my drives...
What would be good parameters, and what do the symbols
really mean ?
If you have good advice, please reply to me personally as
well as posting to the group.
Thanks in advance,
KR.
--
=3D=3D
I don't like spammers. So take the warning
out of my address before you reply.
++
------------------------------
** 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.setup) 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-Setup Digest
******************************
