Linux-Hardware Digest #405, Volume #12 Sun, 5 Mar 00 11:13:07 EST
Contents:
PCMCIA card configuration..... ("fail006")
Re: Advice on PartitionMagic on all-Linux system? (Eric Wick)
Some progress (was Re: unknown disk drive activity) ("Peter Cameron")
KDE SOUND PROBLEM2 ("Jarek \"Krusher\" Onuszko")
Re: adding a scsi controller (Ellen Koinz)
Re: SCSI tape drive, device not ready (phil)
Re: 96Mb = 64Mb RAM??? (Thord Nilson)
Re: 4 Celeron motherboard? (Thord Nilson)
Re: Xircom CBEM56G (S Bond)
Re: DPT 2044UW SCSI controller (Markus Kossmann)
Re: PCI nics ("Wm. Lobb")
Re: dual xeon or single athlon? (Thord Nilson)
3COM SuperStack II 3300 ("Sara")
S3 86365/viewsonic 17GA/rolling screen (Harry Park)
Hardware Switch To Completely Disconnect From Ethernet (mike)
LILO und Win200 ? (Benjamin Henne)
Re: AVA1505A with RedHat 6.1? (Rebeccah H. Prastein)
Re: aha152x not detected (Rebeccah H. Prastein)
Re: Legacy 150x Tape Drive
Re: not sloooow, but sluggish linux modem (Ellen Koinz)
setting up multi processors. ("Mr Paul")
OnStream DI30 Tape Drive (Sami Shaaban)
redhat FS (iso9660?) vs. os/2's HPFS
Re: heating linux (Thord Nilson)
----------------------------------------------------------------------------
From: "fail006" <[EMAIL PROTECTED]>
Subject: PCMCIA card configuration.....
Date: Sun, 5 Mar 2000 23:01:01 +1300
Hi
I know how to configure a ethernet network card..by editing one of the
files...
I would like to know how can i do the same for a PCMCIA card....i know that
the driver or the module that i want is supported by the kernel but i don't
know which file to edit....
i am using redhat 6.1
thanks allot
------------------------------
From: Eric Wick <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.misc
Subject: Re: Advice on PartitionMagic on all-Linux system?
Date: Sun, 5 Mar 2000 09:32:46 +0100
Reply-To: [EMAIL PROTECTED]
Steve Snyder wrote:
> also says that the software is shipped on a "Win32 CD". How does one
> use PM on an all-Linux, no Win32 system? Do/can you run it from a
> DOS-formatted boot diskette?
Since Version 4 there is a script for linux that creates 2 Dos-based
Bootdisks. The Programm is rather useful.
--
Linux-Computing, SpeedDragon http://www.hanse-net.de/eric.wick
ByeBye
Eric
------------------------------
From: "Peter Cameron" <[EMAIL PROTECTED]>
Subject: Some progress (was Re: unknown disk drive activity)
Date: Sun, 5 Mar 2000 10:38:47 -0000
The Central Scrutiniser said:
> My disk drive (actually scsi activity) light flashes every two seconds
> constantly and the disk seems to make actuating noises about every ten.
I've just re-installed RedHat 6.1 after putting a CD-writer into my system
(and having windows crash, overwrite the MBR, and no linux boot floppy :).
Currently my system now consists of two UDMA66 drives (windows on first,
linux on second) connected to Promise Ultra66 card, and on the motherboard:
CD-writer as primary master, DVD as primary slave, and LS-120 floppy as
secondary master.
Since re-installing linux yesterday, I now have disk drive access every
second.
Using the KDE desktop, I investigated all processes with ps, after someone
had mentioned kflushd. I immediately saw a candidate process: an autorun
with an interval of 1000ms. I killed this process and the drive access
immediately stopped. (The autorun is a detector for CD insertion.)
However, under Gnome, there is no autorun process running, yet the constant
disk activity is there. Perhaps there's a Gnome equivalent of autorun.
I'll report back anything else I find.
Cheers,
Peter
------------------------------
From: "Jarek \"Krusher\" Onuszko" <[EMAIL PROTECTED]>
Subject: KDE SOUND PROBLEM2
Date: Sun, 05 Mar 2000 11:07:48 GMT
HELP ME !!
some time ago someone told me to link /etc/sysconfig/soundcard to
/etc/sysconfig/sound but the problem is that I don't have
/etc/sysconfig/soundcard either.... help
------------------------------
From: [EMAIL PROTECTED] (Ellen Koinz)
Subject: Re: adding a scsi controller
Date: Sat, 4 Mar 2000 12:39:29 +0100
In article <[EMAIL PROTECTED]>,
"Ron S" <[EMAIL PROTECTED]> writes:
> I have a system running RH6.1. It is booting from an IDE drive. I added a PCI
> based 53C810 SCSI controller card to the system. When I rebooted, kudzu
> detected the controller and asked if I want to configure the device. I
> responded yes. However when the system came up, it did not recognize the
> controller. I checked and found that an entry had been added to
> /etc/conf.modules for the controller:
>
> alias scsi_hostadapter ncr53c8xx
>
> There was also an entry in /etc/sysconfig/hwconf for the device:
>
> -
> class: SCSI
> bus: PCI
> detached: 0
> driver: ncr53c8xx
> desc: "Symbios|53c810"
> vendorId: 1000
> deviceId: 0001
> -
>
> However it was not seen at boot. The only hint of an error I saw was in
> /var/log/messages:
>
> kernel: scsi : 0 hosts.
> kernel: scsi : detected total.
>
> I thought maybe it isn't loaded since there isn't a target drive. I installed a
> drive, but the results were the same.
>
> If I manually do an insmod ncr53c8xx however, the controller is recognized and
> I can use the attached drive.
> So the question is, why isn't the module loaded automatically at boot time?
Why should it? The module loads as soon as it is needed i.e. when you mount the
attached drive(s). Put the attached drive(s) into the fstab and the NCR will be
recognized at boot time. But why bother? Just mount your drives when you need
them.
Cheers
Ellen
> What do I need to change to make
> it do so?
>
> TIA,
> Ron S.
>
>
------------------------------
From: phil <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup
Subject: Re: SCSI tape drive, device not ready
Date: Sun, 05 Mar 2000 11:44:34 +0000
[EMAIL PROTECTED] wrote:
>
> Hi. Please help...
> I have a Seagate TapeStor DAT 24 internal SCSI tape drive, and a
> Adaptec 2930 Ultra SCSI card. I'm on RH Linux 6.1, kernel 2.2.12.
>
> If root does "mt -f /dev/nst0 erase" (or rewind), the reply to stderr
> is "/dev/nst0: Input/output error". Also, each time root tries this,
> something like these two lines is appended to /var/log/messages :
>
> Mar 4 13:33:12 madhanilab kernel: Device not ready. Make sure there
> is a disc in the drive.
> Mar 4 13:33:44 madhanilab last message repeated 31 times
>
> snipped
Try changing the Dat address to 2 or higher.
Phil Jones
------------------------------
From: Thord Nilson <[EMAIL PROTECTED]>
Subject: Re: 96Mb = 64Mb RAM???
Date: Sun, 5 Mar 2000 12:30:16 +0100
On Wed, 1 Mar 2000, Victor A. Grinberg wrote:
> Marcelo Muzilli wrote:
> >
> > Try this,
> >
> > alter your /etc/lilo.conf for:
> > append="mem=96M"
> >
> > Regards,
> >
> > Muzilli
> >
> > trevorjf <[EMAIL PROTECTED]> wrote in message
> > news:[EMAIL PROTECTED]...
> > > I have 96 Mb ram, but Linux only recognises 64Mb. Win 95 uses all 96Mb,
> > > so its installed OK, how can I get Linux to play with the rest?
> > >
>
> And why is it that I have to do this?? Kernel docs say it's because the
> BIOS doesn't report the size >64M. Well, windoze found it, so how come
> linux can't?
> -vg
>
May be a bug in the BIOS, when i updated the BIOS in my Abit BP6 Linux
2.2.10 would not recognize more than 64 Mb without the append=128M.
The bad thing with using the append is that if you have to remove some
memory, Linux won't boot anymore, so make sure you have a boot-disk
without the append as safety boot.
(I reverted to the older BIOS after a while)
-- Thord Nilson --
------------------------------
From: Thord Nilson <[EMAIL PROTECTED]>
Subject: Re: 4 Celeron motherboard?
Date: Sun, 5 Mar 2000 12:23:57 +0100
On Thu, 2 Mar 2000, Atle wrote:
> "Lee Sau Dan ~{@nJX6X~}" wrote:
> >
> > >>>>> "Atle" == Atle <[EMAIL PROTECTED]> writes:
> >
> > Atle> Personally, I would prefer to have 16x150Mhz Pentiums
> > Atle> instead of one 700Mhz Athlon, because a multiprocessor
> > Atle> system should be a lot more stable if done right. One
> > Atle> processor hangs and corrupts its kernel, 15 are still alive,
> > Atle> and one of them restarts the dead one ...
> >
> > This is too ideal. The actual senario could be as bad as: One
> > processor hangs and corrupts its kernel. This makes it become crazy,
> > so crazy that it starts intervening the other processors, corrupting
> > the in-memory code being executed by other processors. The other
> > processors become crazy, too, and drives the remaining processor
> > crazy.
> This is too bad, assuming that the OS is healthy - that the offending
> program only runs on the downed processor. It would be very difficult to
> confuse a processor running a sound kernel (this was the case on the
> systems I refer to).
> >
Yes, but remember all the cpu:s are sharing the same memory. So if one
cpu goes bad and corrupts its kernel, it has also corrupted the other
cpu's kernel.
Then there is also only one bus to the memory, so if the crashed
processor/bus arbiter manages to lock the bus, the system is also dead.
My experience from dual cpu PC:s with Linux is that reliability is worse
than for a single cpu system. If one processor locks up, the whole system
locks up. I have newer seen any console message like...
"CPU 1 out of order, continiuing with only CPU 0" in linux.
Fot the SMP DEC-10:s (from the middle/late 1970:s) you could see this,
but then it was the hardware that was mostly unreliable.
[snip]
Thord Nilson
------------------------------
From: S Bond <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup,alt.os.linux,comp.os.linux.portable
Subject: Re: Xircom CBEM56G
Date: Sun, 05 Mar 2000 12:03:12 +0000
Reply-To: [EMAIL PROTECTED]
CC wrote:
>
> Hi.
>
> I've installed winlinux 2000 in my Inspiron 3700 a few days ago.
> Similarly, my card is correctly recognised but the modem was installed
> on ttyS0 for some reasons. But i've check that the /dev/modem is
> actually point to ttyS4?! I'm wondering do i have to assign my modem to
> COM5 (which is set in my machine under win98)? If yes, how?
>
ttyS4 is COM5
As long as modem is pointing to the correct ttyS* there shouldn't be any
problem.
Have a look at dmesg to see which ttyS* is correct for your modem, and
which is just the serial port.
> Actualy, i'm a newbie in Linux and this is the first time i've
> installed Linux in my machine. I've read some howtos in order to find
> some help but still don't have a clue of what's going on.
>
> It would be greatly appreciated if any of the experts out there could
> drop me some hints. Thanks.
>
> Regards,
>
> Cedric
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
------------------------------
From: Markus Kossmann <[EMAIL PROTECTED]>
Subject: Re: DPT 2044UW SCSI controller
Date: Sun, 05 Mar 2000 11:15:23 +0100
Scott Stafford wrote:
>
> I'm trying to set up Mandrake, but when it goes to autoprobe for my
> SCSI controller it doesn't find it.
>
> The controller is a DPT 2044UW.
>
> I select the ETA_DMA driver.
Well , the eata_dma driver is marked as obsolete in the current 2.2.14
sources.
AFAIK it is replaced ba the eata.c . So try the eata.o module instead.
--
Markus Kossmann
[EMAIL PROTECTED]
------------------------------
From: "Wm. Lobb" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.networking,comp.os.linux.setup
Subject: Re: PCI nics
Date: Sun, 05 Mar 2000 12:53:33 GMT
You are right. I messed up the numbers. Thank you for your help.
Bill
"Vladimir Florinski" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> "Wm. Lobb" wrote:
> >
> > Hello:
> >
> > I'm sure this question has been asked before, I am sorry.
> >
> > I have 2 3Com PCI 3C509B NIC's installed in this machine. I have read
that
> > you need to specify IO and INT for each NIC when you are using 2 i.e.,
> > ether= XX, XX, eth0 ether=XX, XX eth1
> >
> > My question is with PCI nics you cannot specify the IO and INT. they are
> > assigned by the BIOS. Do you just use a line like:
> >
> > ether= eth0 ether=eth1 ? to tell the kernel that there are 2 nic's? I
have
> > not had much luck getting this to work. I am missing something.
> >
> > Thank You.
> >
> > Bill
>
> Are you sure they are PCI? 3c509b are ISA cards:
> http://www.3com.com/products/dsheets/400230a.html#specifications
>
> Unless it's 905 and not 509. In that case you probably don't need to do
> anything, except maybe a couple of alias lines in /etc/conf.modules
(something
> like 'alias eth0 3c59x').
>
> --
>
>
> Vladimir
------------------------------
From: Thord Nilson <[EMAIL PROTECTED]>
Subject: Re: dual xeon or single athlon?
Date: Sun, 5 Mar 2000 13:43:07 +0100
On Tue, 29 Feb 2000, Hal Burgiss wrote:
> On Mon, 28 Feb 2000 20:18:10 -0700, Vladimir Florinski
> <[EMAIL PROTECTED]> wrote:
>
> [...]
>
> >If a lot of people post here saying they had SMP stability problems on
> >BX boards (in a workstation, not in a server environment), I might
> >reconsider my position. --
>
> Here's one vote. Hard lockups on BP6 (BX), despite trying all the voodoo
> magic, etc. With kernels up thru 2.2.15pre7, XFree 3.3.6. 7-10 day
> average between locks, 24 day max uptime. No ping, no telnet, no SysRq,
> no nothing. Mostly during unattended, idle time. A very intractable
> situation so far (since early Oct).
>
I have experience with the following dual-BX machines:
SOYO Dual P-II 300, hard lockup when starting big jobs if CPU:s was
preheated by running some other applications for atleast a few hours.
Had to underclock to 2x266 MHz, no hard lockups since then -- about 1.5
years. (this is a server, does not run X)
Machine mostly used for heavy FEM calculations.
MSI 6120: Dual Celeron 300A at 450 MHz, a little more than 1 year old.
No hard lockups with linux 2.0.xx, 2.2.5. Had one lockup with 2.2.14 a
few days after upgrade, after wich i backed to 2.2.5. (runs X)
Abit BP6: Dual Celeron 400 at 450 MHz. (from october-99)
Have had much hard-lockups with this one. (runs X)
I have tested serveral "voodoo" things on this one. The things that seems
to make it work stable is:
1) Use Linux kernel 2.0.38 (or older)
2) append="noapic", and try swapping the CPU:s if you still get lockups.
If this does not help, i would recommend testing with different
celerons. (since celerons are not specified for SMP they may be buggy
there) (I have not tested other cpus yet, since I have only 5 days
2xsetiatome on swapped setup yet, the unswapped gave lockup in about
36 hours)
/Thord Nilson.
------------------------------
From: "Sara" <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux.networking,comp.os.ms-windows.networking.tcp-ip,comp.os.os2.networking.misc,comp.os.os2.networking.server
Subject: 3COM SuperStack II 3300
Date: Sun, 5 Mar 2000 05:31:42 -0800
I have a 3Com Superstack II 3300 used for my colocation at my ISP. When I
hook all the cables the status lights signal a-ok except the line provided
by my ISP (which I think is called the uplink). I've heard that this line
is a crossover cable, but not sure. How do I go about making this
connection work?
------------------------------
From: Harry Park <"hpark01_nospam"@sprynet .com>
Subject: S3 86365/viewsonic 17GA/rolling screen
Date: Sun, 05 Mar 2000 13:51:38 +0000
Reply-To: [EMAIL PROTECTED]
I am attempting to get a viewsonic 17ga monitor working with a
S3 86365 (Trio 3d) card. I am using rh6.0. I have upgraded to the newest
svga server and the newest XFree (3.6 I think, working from memory
here). I have succeeded in getting the Xserver to work except that the
screen rolls. If I just had a vertical control nob I could fix the
problem. The viewsonic does not have one though. My reading has lead me
to believe that a bad vertical sync pulse will make the screen roll. Is
there a way that I can repair this problem with xvidtune? Is there
another way of fixing this? I'm at the end of my rope on this one.
Thanks for the help
H
------------------------------
From: mike <[EMAIL PROTECTED]>
Subject: Hardware Switch To Completely Disconnect From Ethernet
Date: Sun, 05 Mar 2000 09:43:59 -0500
Hi,
I would like to know if it is possible to put simple hardware
switch in an ethernet cable, for example 10 Base T twisted pair,
so as to completely disconnect a computer from the local network.
If this could work reliably, it would be a secure way to prevent
unauthorized hacking of a system.
Mike
P.S. I am concerned that the switch could cause some type
of imbalance in the signal maybe creating internal reflections
or noise or general impedance mismatch in the system that
would interfere with the connection.
------------------------------
From: Benjamin Henne <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.misc
Subject: LILO und Win200 ?
Date: Sun, 05 Mar 2000 15:42:07 +0100
Reply-To: [EMAIL PROTECTED]
Hallo
habe soeben versucht auf Meinem PC (Win2000 und Linux) neu zu
konfiguriere.
Doch wenn ich ich Lilo installieren lasse auf Diskette und er zum
Boot-Eintrag von Win2000 kommt, serviert Lilo als Ausgabe:
Partitionseintrag nciht gefunden.
Kann LiLo Win2000 nicht booten ?!?
------------------------------
From: Rebeccah H. Prastein <[EMAIL PROTECTED]>
Subject: Re: AVA1505A with RedHat 6.1?
Date: Sun, 5 Mar 2000 08:55:24 -0600
Sorry, my mistake. 0 is disconnect/reconnect disabled.
The other parameters you can use are parity, synchronous, delay, and
extended translation.
Rebeccah
In article <89jfj3$qcc$[EMAIL PROTECTED]>, [EMAIL PROTECTED] says...
>
>
> > - Be sure you don't have an interrupt conflict: run the adaptec
> > software and configure the card for an unused interrupt
> >
> > - Then, as root:
> >
> > modprobe aha152x aha152x=0x340,9,7,0
> >
> > Where 0x340 is the IO port
> > 9 is the interrupt
> > 7 is the controller's SCSI ID
> > 0 is well, I forget
> >
> > Peter
> >
>
>
> 0 is parity checking enabled.
>
> Rebeccah
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>
------------------------------
From: Rebeccah H. Prastein <[EMAIL PROTECTED]>
Subject: Re: aha152x not detected
Date: Sun, 5 Mar 2000 08:57:27 -0600
One major hurdle surmounted. I can find the card with modprobe. Now I
just need some help getting it detected at bootup. Any ideas?
Rebeccah
------------------------------
From: <[EMAIL PROTECTED]>
Subject: Re: Legacy 150x Tape Drive
Date: Sun, 05 Mar 2000 15:16:54 GMT
You are absolutly correct, I found the website for Legacy (www.legacy.ca)
and well, it is an 8bit non-scsi card that comes as a microchannel or ISA
card. The only drivers that I see available on there website is for
dos/windows 3.1, so this is a really old device.
I would still love to get this working under linux though. If something is
not out yet that would support it, what would be the first steps one should
take to get it supported?
RayS
"M. Buchenrieder" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> "Andrew E. Schulman" <[EMAIL PROTECTED]> writes:
>
> >> I have an old Legacy 150x tape drive system, with a controler that I
believe
> >> to be SCSI, but that aside I was attempting to find out if at all there
> >> would be a way to get this to operate under the linux OS
>
> >If your SCSI card is working correctly, your tape drive should be
> >autodetected at boot. Then all you need is the right device special
> >files.
>
> Read again. IF it is a SCSI card at all, he'll most probably have
> to specify I/O address and IRQ settings, not to mention loading
> the correct driver module for it. The SCSI cards that shipped with
> tape drives usually were BIOSless cards.
>
> Note: This 150MB drive is possibly not a SCSI drive at all.
> If it's older than just 4 - 5 years, then it's perhaps a QIC-02
> tape drive with a proprietary adapter card that looks just like
> an 8-bit SCSI controller (using a 50-pin cable), but needs specific
> driver support. Without further information about the kind of controller
> card used, it's probably not possible to give detailed instructions
> at this time.
>
> Michael
> --
> Michael Buchenrieder * [EMAIL PROTECTED] * http://www.muc.de/~mibu
> Lumber Cartel Unit #456 (TINLC) & Official Netscum
> Note: If you want me to send you email, don't munge your address.
------------------------------
From: [EMAIL PROTECTED] (Ellen Koinz)
Subject: Re: not sloooow, but sluggish linux modem
Date: Sun, 5 Mar 2000 13:27:09 +0100
In article <[EMAIL PROTECTED]>,
J.R. Lockwood <[EMAIL PROTECTED]> writes:
> Hi--
>
> I previously posted this question but I believe it was to the wrong
> audience. I recently installed a 3Com USR 56K external faxmodem for use
> with a dual boot Win98/Linux system. The modem performs wonderfully under
> Windows, maintaining download speeds of better than 5 KB/s for all file
> sizes. However, under Linux, I can only maintain about half of that
> speed. I have read about people having problems with modems under Linux
> because of an IRQ conflict. However, I don't think this is my problem, as
> I believe my download speeds would then be abysmal instead of just
> disappointing. The following are common to the use of the modem under both
> OSs: connect speed of 49,333 bps, IRQ 4, and the modem initialization
> string.
>
> Here is more information that may help to diagnose the problem: using the
> modem monitor, I noticed that my modem throughput is cyclical. I will have
> a few seconds of download speeds around 6 KB/s, but then the modem stalls
> and has to build back up for a couple seconds. This behavior is repeated,
> with constant frequency, for all downloads. If I download a small file, I
> get good speed because the whole thing can transfer during the "fast
> phase", but for larger files, I can barely maintain 2.8 KB/s because of the
> cyclical behavior of the modem speed.
>
Sounds to me as if your serial port isn't set to operate at hi speed.
put something like
${SETSERIAL} -b /dev/ttyS1 irq 3 port 0x2F8 skip_test autoconfig spd_vhi ${STD_F
LAGS}
into your setserial command (adjust the port specs to your needs).
Cheers
Ellen
> My questions are this:
>
> 1) What could cause the modem throughput to cycle?
> 2) If 1) is related to the priority placed on the serial port by Linux, is
> there a way to tell Linux to pay more attention to it?
> 3) If 1) is unrelated to port priority, what other suggestions do folks
> have?
>
> Thanks,
> J.R.
>
> --
> Posted via CNET Help.com
> http://www.help.com/
------------------------------
From: "Mr Paul" <[EMAIL PROTECTED]>
Subject: setting up multi processors.
Date: Sun, 5 Mar 2000 09:43:35 -0600
i am a new new newbie to linux, so new i haven't even installed it yet. i
just got an old 486 that i'm going to load with linux so i can dink around
with it. actually i have two old 486s and know that linux supports
multiprocessors. i'd like to combine the two chips into one machine... maybe
this isn't the right place to ask this question.
is this even possible to have two 486dx2 chips (one 50 the other 66) slapped
into one machine and have linux know what to do?
thanks in advance!
Paul
------------------------------
From: Sami Shaaban <[EMAIL PROTECTED]>
Subject: OnStream DI30 Tape Drive
Date: Sun, 05 Mar 2000 15:50:06 GMT
I recently purchased an OnStream DI30 tape drive and tried using it with
Amanda on a Linux system (with the 2.2.14 kernel compiled with the patch
needed to make this drive work). The drive works fine if tar is writing
to it, but to does not seem to work with dd. After getting restore
errors with Amanda's amrestore and amrecover commands, I tried reading
with dd, which didn't work, then I tried writing to a new tape with the
following simple dd command:
dd -of=/dev/nht0 -if=./somefile -bs=32k
with the following output:
dd: /dev/nht0: invalid argument
0+1 records in
0+0 records out
My guess is that the DI30 driver just isn't ready for prime time and
that I should just return this drive for one that is supported. If
anyone has any experience with this, I would love to know before mailing
this drive back. Otherwise, hopefully my message will serve as a
warning for those considering purchase of this drive for use with Linux.
------------------------------
From: [EMAIL PROTECTED] ()
Crossposted-To: comp.os.linux.redhat
Subject: redhat FS (iso9660?) vs. os/2's HPFS
Reply-To: [EMAIL PROTECTED]
Date: Sun, 05 Mar 2000 15:52:56 GMT
Since switching from os/2 to linux (redhat 6.1), I've had little to complain
about with once exception: the way the filesystem reacts to an unaticipated
halt.
On os/2, even with very unreliable hardware, I never ever lost a byte of data
in 6 years of use. It was a welcome change after windoze' piss poor FAT file
system where anytime there was a crash, there were crosslinked files.
My computer has a bad AGP video board that sometimes doesn't initialize
properly. Killing the power during the boot process has twice caused file
system errors that the automatic fsck has been unable to correct. Perhaps it
was just being panicy. The lost+found directory hasn't had anything of
importance -- last time it was just some netscape crud.
------------------------------
From: Thord Nilson <[EMAIL PROTECTED]>
Subject: Re: heating linux
Date: Sun, 5 Mar 2000 17:01:27 +0100
On Tue, 29 Feb 2000, Paul Ingram wrote:
>
>
> "Robert W. Cunningham" wrote:
>
> >
> >
> > My original reply was not designed to impress anyone with my knowledge, but
> > merely to provide a minimal explanation to enable the original poster to
> > measure the power consumed by his system, and make sense of the numbers.
> >
> > What was your reply intended to do? Prove that you're taking your first EE
> > class?
> >
> > -BobC
>
> Hi Guys,
>
> Thanks for your input, what I am actually trying to do is ascertain the heating
> effect of several machines in a computer room. The place I am installing them in
> want to know the heat output so they can check if their aircon unit is up to the
> job. So, what they asked me was if I could tell them the heat output figure, which
> is where I came in.
>
> So, if I take the current (actual) x line voltage , this will give me heat output
> in Watts right?
>
> Pi
As a rule of thumb you can take:
power = line_voltage_rms * line_current_rms * PF
Where PF is the powerfactor of your supply. PF will depend on your supply
as well as the impedance in your mains.
For a direct rectifaction supply (= normal switched PC power sypply) PF
can be in the range of 0.5 to 0.7 for a total load of 0.5 to 2 KW.
PF increases with increasing load, mostly due to that the voltage waveform
is distorted more and the voltage ripple in the capacitor in the
power supply increases.
For best results, you should measure the total power to all machines
you intend to run. If you are in an office building whith many PC:s on
the same mains, you might experience a lower PF when all the other
machines are turned on compared to when they are off. But it all depends
on how "stiff" your manins are.
Summary: the best you can do, is to borrow/rent a power-meter that can
show you power-factor etc. I know FLUKE has one with grapical display
that can show both voltage and current waveforms, but I don't remember
the model # now.
If measure with this, you will be amazed how far from sinousodial the
current waveform is.
/Thord Nilson.
------------------------------
** 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.hardware) 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-Hardware Digest
******************************
