Linux-Hardware Digest #523, Volume #10 Fri, 18 Jun 99 15:13:36 EDT
Contents:
hdparm corrupts OS on SP-97V even with patches. (Kyle Petersen)
Re: Power save on a Sony Multiscan 15sf monitor (Randy Olinger)
Re: Adaptec 2940UW and IBM 9ES headache (Johan Groth)
Re: 2 GB JAZ (Paul Gray)
Somebody Help! ("Michael Vachon")
Re: Diamond Viper TNT AGP + XWindows (mj)
Serial mouse in ps/2 port? (Pankaj Pant)
installation problem with Red Hat 6.0!!!!! ("matteo badino")
Frozen Serial MS mouse (Bob Gamble)
Re: Can I probe my video card to get Dot Clocks? (Howard Mann)
Re: BIOS Upgrade Suggestion (Walter Tautz)
Re: The ultimate backup program for any OS? (killbill)
Re: modem problems (Randy Olinger)
Re: 3c905B (Phil DeBecker)
Re: BIOS Upgrade Suggestion (Villy Kruse)
Need AAmazing cm8426 specs. (Ron Alexander)
Re: Linux Backup Solution. (killbill)
----------------------------------------------------------------------------
From: Kyle Petersen <[EMAIL PROTECTED]>
Subject: hdparm corrupts OS on SP-97V even with patches.
Date: 18 Jun 1999 08:47:37 -0700
I can't get DMA or UDMA or other supposedly supported features of my
disk drives to work. If you've worked problems like this, please
post any suggestions. Have you had success asking the UDMA or kernel
people about such problems?
Using a stock 2.2.9 kernel or one with the
2.2.9.uniform-ide-6.19.kilo.patch patch from
http://www.dyer.vanderbilt.edu/server/udma/
and a stock or latest patched "hdparm" from the same place,
my OS gets corrupted ("time", "sleep", "shutdown" all fail)
after I use hdparm's -c# or -m# or -d1 or -X#. The -c and -m
do speed up the disks and the disk isn't corrupted, but it can't be
used.
I had just upgraded my Red Hat 5.2 OS to kernel 2.2 following a couple of
2.2 upgrade documents and the kernel "Changes" page. Everything seems
fine. I had thought this would detect my DMA hard drive and my UDMA
hard drive and speed up the quite slow speed "hdparm -tT" had been
showing. No luck; hdparm and /proc shows no DMA for either drive.
I've got an ASUS SP-97V MB with "Award BIOS V. 4.51PG (01/22/98)".
A document said BIOSs dated 1998 should work OK. I'd don't have
M$ Windows. Can I upgrade my BIOS from Linux (or DOSEMU or WINE)
while using the MB? Anybody got a URL for documentation on that?
The boot messages indicate the drives (or BIOS?) want DMA.
======================
Hardware Details:
/dev/hda: [[ UDMA ]]
Model=Maxtor 90576D4, FwRev=WAS8283C
Config={ Fixed }
RawCHS=11166/16/63, TrkSize=0, SectSize=0, ECCbytes=29
BuffType=3(DualPortCache), BuffSize=256kB, MaxMultSect=16, MultSect=off
DblWordIO=no, maxPIO=2(fast), DMA=yes, maxDMA=2(fast)
CurCHS=700/255/63, CurSects=11255328, LBA=yes, LBAsects=11255328
tDMA={min:120,rec:120}, DMA modes: mword0 mword1 mword2
IORDY=on/off, tPIO={min:120,w/IORDY:120}, PIO modes: mode3 mode4
/dev/hdd: [[ EIDE ]]
Model=ST32140A, FwRev=08.08.01
Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
RawCHS=4095/16/63, TrkSize=65535, SectSize=951, ECCbytes=22
BuffType=3(DualPortCache), BuffSize=128kB, MaxMultSect=32, MultSect=off
DblWordIO=no, maxPIO=2(fast), DMA=yes, maxDMA=0(slow)
CurCHS=1023/64/63, CurSects=4127760, LBA=yes, LBAsects=4127760
tDMA={min:120,rec:120}, DMA modes: sword0 sword1 sword2 *mword0 mword1 mword2
IORDY=yes, tPIO={min:180,w/IORDY:120}, PIO modes: mode3 mode4
Thanks.
------------------------------
From: Randy Olinger <[EMAIL PROTECTED]>
Subject: Re: Power save on a Sony Multiscan 15sf monitor
Date: Fri, 18 Jun 1999 11:47:42 -0500
Reply-To: [EMAIL PROTECTED]
Did you set the BIOS to use DPMS for monitor low power?
APM is a mixture between op-sys and BIOS settings.
Randy
PhilD wrote:
> Hey, I have a Sony Multiscan 15sf monitor and am running RH6 with
> e/gnome. I was wondering how to get the monitor to power down. The
> screen saver works correctly, and I have the power save set to
> powerdown after the screensaver has been on for 20min. But it
> doesn't. Does anyone know how to fix this? Thanks for any help.
>
> PhilD
------------------------------
From: Johan Groth <[EMAIL PROTECTED]>
Crossposted-To: comp.periphs.scsi
Subject: Re: Adaptec 2940UW and IBM 9ES headache
Date: Fri, 18 Jun 1999 18:35:10 +0200
Nils Remmers wrote:
>
> > So I did and now comes the promblems. I can't do anything with the drive
> > except talk to its firmware. I've tried Adaptec SCSI benchmark program
> > in Win95 but it reports sense error 0xBh and in Linux I receive data-in
> > sync problems and when it tries to reset the SCSI bus the system
> > freezes. The strange thing is that I can low format the drive and verify
> > the surface in the adapter bios but I can't create a partition on the
> > drive in either DOS, WIN95 or Linux. Has anyone any clue what might be
> > wrong? Can it be the cable? It's a SE cable with 5 connectors.
> >
> The cable must not be longer than 1.5 meters. Might it be that yours is
> longer ???
> Min space between 2 devices is 10cm. Try connecting the se-devices
> directly to the internal connector with a flat cable (just for testing)
> does it work ?
Everything else works without the LVD-drive. I've tried to remove one SE
and kept the LVD and the terminated SE but no success. The space between
devices is about 20 cm. The only thing I've done is to add a LVD drive
to a working system and now it refuses to work with the LVD.
///Johan
--
! Ericsson Infotech AB ! ECN: 863 3554 !
! Johan Groth KS/EIN/T/R ! Phone: +46 54 19 3554 !
! Box 1038 ! Fax: +46 54 19 3466 !
! Lagergrens gata 3 ! Email: Johan.Groth(at)ein.ericsson.se !
! S-651 15 Karlstad, Sweden ! Email: Johan.Groth(at)ks.ericsson.se !
------------------------------
From: [EMAIL PROTECTED] (Paul Gray)
Subject: Re: 2 GB JAZ
Date: 18 Jun 1999 15:55:22 GMT
Jens,
Before you format (using mke2fs, I assume?), do you re-partition the
disk? When I run fdisk on a new Jaz media, I get many partition
boundary mis-alignment errors. So, before I mke2fs on these drives, I
like to remove the (somewhat inane) partitioning which places all data
on the primary partition 4 to a single primary partition 1.
I can't tell from your post if you're using the disk "as-is." That
is, in its pre-formatted state from the factory. If so, I assume that
you're mounting it as vfat?
The only problem that I've had with the media is that the metal
closure comes off-track and scrapes against the drive when ejecting.
Good luck and regards,
Jens (ein HSS-Fan) ([EMAIL PROTECTED]) wrote:
: Hi,
: We are using Linux + a internal 2 GB JAZ drive.
: The truth of the matter is we use each new disk for only one time: we can
: read/write on it when it is new; due to the filesystem of linux, we have to
: re-format the disks. After having formatted, all disks schow up with "read
: errors".
: We tried several tools to check out where the error stems from: all tools
: resulted as follows: sense key -- medium error
:
: Any-one who can help..?
:
:
:
:
--
Paul Gray
gray@[mathcs.emory.edu | cns.uni.edu]
Math/Computer Science Department
Emory University, Atlanta, GA
150 North Decatur Building
1784 N. Decatur Rd.
Atlanta, GA 30322
------------------------------
From: "Michael Vachon" <[EMAIL PROTECTED]>
Subject: Somebody Help!
Date: Fri, 18 Jun 1999 17:06:51 GMT
I am brand new to linux and I got through the installation and when I boot
Linux I get to the Bask prompt....but how do I RUN Linux from the prompt??
------------------------------
From: [EMAIL PROTECTED] (mj)
Crossposted-To: comp.os.linux,linux.redhat.misc,comp.os.linux.x
Subject: Re: Diamond Viper TNT AGP + XWindows
Date: 18 Jun 1999 15:22:42 GMT
Hi,
Go to xfree86.org and download the latest SVGA server (3.3.3.1 or somewhat
like that). I works great with my ELSA ERAZOR II (Riva TNT too).
You won't find these server on DEBIAN 2.2 or REDHAT latest releases. Download
directly from www.xfree86.org (there you'll find a FTP site for download).
You'll at least need the following:
SVGA
Xbin
Xcfg
Xlib
XVGA16 (XF86 setup is included here)
Xdoc
Xset
(i hope i'm not missing some package)
untar this tar's over your existing /usr/XF86R6 directory, and run XF86Setup.
This may not be the best way to do it, but it works.
Have Fun!
bye.
>Does anyone know what chip set / if an X server exists (or where I can
>get one) for a Diamond Viper TNT AGP card (16 MB)?
>
>If anyone just knows a place where I can search for free (or even
>commercial, god forbid) X-servers, I would really appreciate it.
>
>TIA,
>John
>
------------------------------
From: Pankaj Pant <[EMAIL PROTECTED]>
Subject: Serial mouse in ps/2 port?
Date: 18 Jun 1999 12:44:16 -0400
Hi all,
With some help from this group we have our dual-boot setup functioning
pretty well.
Yesterday we got a 3-button serial mouse and it seems to work OK in both W98
and RH6.0. However, we want to free up the serial port for connecting a
terminal (actually a 486). So we tried a serial-to-ps/2 adapter, but the
system dosen't boot ie. LILO dosen't come up. If we pull the mouse out, it
wakes up again and the LILO prompt appears (we can't use the mouse, off
course).
Has anyone else tried this or heard of this problem. I suppose the system is
trying to recognize the mouse but failing. Could the adapter be bad?
Thanks once again for any tips.
- Pankaj.
------------------------------
From: "matteo badino" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.portable,comp.os.linux.setup
Subject: installation problem with Red Hat 6.0!!!!!
Date: Thu, 17 Jun 1999 20:05:29 +0200
I have a problem to install the distribution of Red Hat 6.0 on my notebook
TOSHIBA SATELLITE 4010CDT. The configuration of system software/hardware is
remain that original without some modification:
Hardware:
Pentium II at 266 Mhz
RAM Memory: 64 MB
BIOS Name: Toshiba 07/25/98
BIOS Version: Satellite4000CDT v7.40 TOSHIBA
Hard Drive Memory: 4.1 GB
Screen Size: 12.1 inches
Resolution: 800x600 DPI
CD-ROM speed: 24x TEAC CD-224E
BUS TYPE: ISA
Video Card: Chips and Tech. 65555 PCI
Internal modem: Megahertz 3CCM156 B 56K Modem
Floppy drive: 1.44MB Generic NEC Floppy Disk
Monitor: Monitor Plug and Play
SOFTWARE:
Operating System pre-installed: Windows 98, version 4.10, Build 1998
I am an italian user and I am not expert in UNIX/LINUX system.
I execute the installation procedure of Red Hat 6.0 correctly and everything
is right. But when I start the boot again fron the floppy boot created
during installation (but the same thing verifies even if I start it from
the first sector boot partition or from MBR), after the prompt "boot: linux"
and dil loading of different daemon, have:
Red Hat Linux release 6.0 (Hedwig)
kernel 2.2.5-15 on an i686
localhost login:_
At this point I can not do nothing else as anything I type is not inserted.
Therefore I cannot type the password to enter the system Linux.
Is it an installation problem or is it due to the hardware of my notebook ?
I hava already read hundred of documents like, Hardware Compability, Linux
on Laptops, HOWTO etc....., without resolve my problem!
How can I do ?
------------------------------
From: Bob Gamble <[EMAIL PROTECTED]>
Subject: Frozen Serial MS mouse
Date: Wed, 16 Jun 1999 17:06:07 -0600
I had an almost flawless installation, except the Microsoft serial mouse
with my computer is completely dead. X loads fine, and the install
probe identified the mouse correctly. The port was assigned cua0, which
is where it should be. Do I need to edit another file?
------------------------------
From: Howard Mann <[EMAIL PROTECTED]>
Subject: Re: Can I probe my video card to get Dot Clocks?
Date: 18 Jun 1999 16:30:50 GMT
John Miller wrote:
> I have a trident 3DImage 985 video card which linux recognizes and
> supports. When I run xf86config, I am told NOT to probe the clocks.
>
> Here is the problem that I'm having:
>
> If I use the modelines that are written into the XF86Config file, my
> monitor shuts off. Only by reducing the horizontal scan rate to 35kHz
> (where the actual upper bound is 55KHz) will I even be able to run the
> 680x400 mode. By commenting out modelines with high Dot clocks (over 65
> which is the bandwidth of my monitor) I'm able to get 1024x768 at 60hz
> using a 65 MHz dot clock. Unfortunately, the screen experiences
interference when
> the hard drive is accessed or when the mouse is moved. The problem is
> even worse, for some reason, when I use the 50 Mhz dot clock for the
> 800x600 resolution.
>
> Here is what I think is the solution:
>
> Since the monitor will run 800x600 under win95 without interference,
there
> must be a dot clock which is stable between 50 and 65. Getting this info
> will let me also run 800x600 under linux. But I need to know what dot
> clocks the 9850 chipset uses.
>
> Can I probe the clocks without damage to find out which ones it uses?
> Or, does anyone have a better idea what is going on here?
>
I do not know a reason why you could not check the maximum
dot clock provided by your videocard with the "X -probeonly "command.
( See: http://www.xmission.com/~howardm/Xwoes.html)
1. You should be able to "use" the programmable pixel generator on the
videocard. See: http://www.xfree86.org/3.3.3/trident1.html#1
You may have to specify an option, as indicated, in the XF86Config file.
2. Peruse : http://www.xmission.com/~howardm/modelines.html
for a method of writing a modeline using the VESA monitor
timing standards.
Choose a screen resolution and refresh rate supported by your monitor.
( Let me know if this works for you)
Cheers,
Howard Mann.
================== Posted via SearchLinux ==================
http://www.searchlinux.com
------------------------------
From: [EMAIL PROTECTED] (Walter Tautz)
Subject: Re: BIOS Upgrade Suggestion
Date: 18 Jun 1999 15:54:07 GMT
On Thu, 17 Jun 1999 15:58:47 GMT, M. Buchenrieder <[EMAIL PROTECTED]> wrote:
>Don McKeown <[EMAIL PROTECTED]> writes:
>
>>I have a 486-25 into which I'd like to install a 2.5 gig drive; the
>>BIOS, however, will only allow up to like .5 gig drives.
>
>Install it. Linux doesn't care about the BIOS settings. All
>you need is a small boot partition (say, 150 MB) for your
>kernel etc. to be living in. You'd only need an add-on
>EIDE card with BIOS if you wanted to use that drive for more
>than just Linux.
>
Isn't it true at least if one reads the mini-howto on large disk that
all is moot if one uses the latest kernels (2.0.37 or 2.2). Simply
install it...perhaps lilo still needs it.... Some one clarify the
state of the art including what can and cannot be turned off in the
BIOS settings in a linux only installation. Does having other
operating systems booting make a difference?
-walter
>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.
--
===============================================
Walter Tautz
Office: MC5136A, x6895
Department of Combinatorics & Optimization
Faculty of Mathematics
University of Waterloo
Waterloo, Ontario, Canada N2L 3G1
------------------------------
From: killbill <[EMAIL PROTECTED]>
Subject: Re: The ultimate backup program for any OS?
Date: Fri, 18 Jun 1999 17:10:40 GMT
In article <hDsa3.189$[EMAIL PROTECTED]>,
[EMAIL PROTECTED] (Peter F. Curran) wrote:
> In article <7k6cjb$afl$[EMAIL PROTECTED]>,
> "Brad Ball" <[EMAIL PROTECTED]> writes:
> >Do you know what would be cool? A program where you can boot
> >from a floppy and select a "Backup" option that will make an
> >image file of your ENTIRE hard drive. It doesn't matter what
> >OS's you are running or how many.
> [snip]
>
> Heh, such a beastie already exists and comes with every
> linux distribution. The 'dd' command can copy an entire
> hard drive (say /dev/hda) for you into a single file raw
> sector by sector. You just need a bootable disk with this
> proggie and support for your target devices. (Or you could
> just use it without rebooting to a floppy, but you might
> want to remount partitions read only for safety.) I suggest
> that you pipe to gzip to make the image smaller, (since the
> image file for an 8Gig drive would be 8Gigs regardless of how
> much space was used on the drive!)
>
The bad news is "dd" needs a little help for multivolume archives to
CDRW, for a lot of subtle technical reasons.
The good news is that I have released a GPL backup utility that gives dd
(or tar, for that matter) exactly the little bit of help it needs. See
"backburner" on www.freshmeat.net.
Note that the standard Linux programs 'find', 'tar', 'gzip' and 'dd', if
combined with two custom programs (backburner, and cdrecord), will do
everything necessary both image and native full and incremental file
system backup and recovery of any amount of data for just about any file
system.
--
Bil Kilgallon ([EMAIL PROTECTED])
--"I believe, what I believe, has made me what I am. I did not make
it, It is making me, it is the very truth of God, not the invention
of any man". Rich Mullins, quoting G.K. Chesterton.
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
------------------------------
From: Randy Olinger <[EMAIL PROTECTED]>
Subject: Re: modem problems
Date: Fri, 18 Jun 1999 12:18:52 -0500
Reply-To: [EMAIL PROTECTED]
Incorrect information all over previous response.
I have the DIGICOM modem in my linux box and it works just fine.
First of all, COM[3,4] are not used by the motherboard unless you
specifically tell the MB to use them. I use COM2, but COM4 should
work fine. Be aware that you cannot use COM2 and COM4 at the
same time unless you assign them different interrupts. Best bet is
to disable any motherboard com ports that you are not using.
I had to tell the setserial command the exact uart to use. I cannot remember
the
incantation, but it was in the man page. If you want, send an e-mail to my
home address ( [EMAIL PROTECTED]) asking for my command syntax
and I'll send it to you. I had the exact same problem. (remove the _NOSPAM)
Randy
"M. Buchenrieder" wrote:
> "Jase" <[EMAIL PROTECTED]> writes:
>
> >Hi I have a digicom systems (makers of the modem blaster series) 3660 56k
> >internal isa 56k non winmodem
> ^^^^^^^^^^^^
>
> Are you sure it isn't a winmodem ? Does it work in DOS ?
>
> >it runs in on com 4 when i use it in windows
> >however its not detected at sturtup on linux rh6.0 when i use isapnptools
> >ive configured it to ttys3 irq4 using the same resources as windows isapnp
>
> Don't use IRQ 4/3 for the internal modems. These are already in use
> by the onboard serial ports.
>
> >works untill i try these commands
> >setserial /dev/ttyS3 irq 3 autoconfig
> >setserial /dev/ttyS3 spd_vhi
> >setserial /dev/ttyS3
> >the last line gives an output with uart unknown does this mean that even
>
> [...]
>
> UART unknown usually means that it is a winmodem - if you edited
> isapnp.conf for ttyS3, that is.
>
> 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.
------------------------------
Date: Fri, 18 Jun 1999 13:18:00 -0400
From: Phil DeBecker <[EMAIL PROTECTED]>
Subject: Re: 3c905B
ELVIS wrote:
> System (2.0.35) stability seems to have gone to hell after adding this
> adapter. I will probably recompile to 2.0.36 and ditch the module approach
> (never worked well with the old NE200).
>
> Here's the curious part: after running 'shutdown -h now' I can ping the box
> and get a reply.
>
> So... what part of 'HALT' don't I understand?
Are you sure you don't have another machine on the network with the same
IP address? Try turning the linux box off and pinging again, if you still get
a reply then that's the culprit.
I run a 3c905b as a module with 2.0.35 on one of my linux machines and have
seen no problems; I also run a NE2000 clone on 2.0.36, again a modular driver,
and again zero problems. I'm currently writing this on a 2.2.10 system with a
3c905b, also modular driver, also zero hiccups.
So unless you give more info on your configuration, all I can guess is that
you've got some kind of IP address conflict.
Phil D.
------------------------------
From: [EMAIL PROTECTED] (Villy Kruse)
Subject: Re: BIOS Upgrade Suggestion
Date: 18 Jun 1999 19:46:54 +0200
In article <[EMAIL PROTECTED]>,
Walter Tautz <[EMAIL PROTECTED]> wrote:
>Isn't it true at least if one reads the mini-howto on large disk that
>all is moot if one uses the latest kernels (2.0.37 or 2.2). Simply
>install it...perhaps lilo still needs it.... Some one clarify the
>state of the art including what can and cannot be turned off in the
>BIOS settings in a linux only installation. Does having other
>operating systems booting make a difference?
Linux itself never depended on the bios for disk io or any other device
access for that matter. The limit here is in the IDE interface and that
interface has a much higher limit than the int13 bios call interface.
The lilo loader has always and still does depend on the bios to provide
disk acces as well as console and keyboard io.
You could make a loader that is not dependent on bios, not even for
disk io. The FreeBSD people have done that, but they do have problem
with differences between different scsi adapters and limited space to
support all of them. Thus using the bios for disk io makes the loader
device independent.
There are roumors that some bioses can access disks beyound the 8Gb limit,
but so far I havn't seen any detail. It would require an alternate bios
call instead of the standard int13 interface, and when such an alternate
bios call becomes widely available lilo could start using this and thus
no more 1024 limit problem.
Villy
------------------------------
Subject: Need AAmazing cm8426 specs.
From: [EMAIL PROTECTED] (Ron Alexander)
Date: Fri, 18 Jun 1999 17:49:57 GMT
Help,
I can't find my manual since I moved. I am installing, and need to specify
the monitor freqs etc. for an AAmazing CM8426 monitor (it might really be a
Sony of some description).
I tried to find aamazing on the web, but no luck so far.
TIA,
Ron Alexander
------------------------------
From: killbill <[EMAIL PROTECTED]>
Subject: Re: Linux Backup Solution.
Date: Fri, 18 Jun 1999 17:24:32 GMT
In article <7k94nb$s90$[EMAIL PROTECTED]>,
[EMAIL PROTECTED] (Matthias Kilian) wrote:
> > I don't understand why the multi-volume option and the compress
option
> > are mutually exclusive...
>
> I did some tests a few months ago. I don't remember the results
exactly, but
> here are some hints:
>
> The Linux tape driver and the streamer often have different ideas
about
> the current tape position. You can see this using mt(1) with command
> "status" (ioctl MTIOCGET, field mt_blkno of struct mtget), which
prints
> Linux's idea of the tape, i.e. the driver status), and "tell" (ioctl
> MTIOCPOS, field mt_blkno of strict mtpos), which queries the streamer
for
> the current position. The latter is only supported by some SCSI tapes
> (most DAT streamers). When writing until a write error occurs (short
> write), the current block number reported by the tape driver will be
> greater than the block number reported by the streamer.
>
> Yet worse, even if the software asks the streamer for the current
block
> number, it appears that this number is too high.
>
> [speculations on]
> This may have to do with the cache built in the streamer. When using
> hardware compression, several megabytes may fit on the last inches of
the
> tape, e.g. if the data written consists of zeroes only. If the data is
> "uncompressable", the streamer may fail to put the last blocks on the
tape.
> [speculations off]
>
> May be one could patch tar to use ask the streamer for the tape
position at
> end of tape, but my experiments showed that reliable synchronization
only
> works with hardware compression and write buffering disabled. Thus, a
write(2)
> to the tape waits until the data has been physically written to the
tape.
> Unfortunately, this slows down backup by a factor of 5 to 10.
>
> If all the above is total nonsense, please tell me, I'm willing to
learn more
> about tape operations :-)
Well, that is probably "a" problem, but you cannot (at least on my red
hat 5.2 distribution) give tar both the compress and the multivolume
archive flag simultaneously, even if you are just sending the tar file
to the hard drive. I would think this would have nothing to do with
tape drive subtleties (which are indeed a mess). Besides, tar can
stream output, so I would think it cant go in and mess with all that
stuff as a normal operation.
I worked around the problem with backburner, so it is no big deal. It
just never made sense to me why it was impossible...
You can, (if I recall correctly), do this on HPUX. On the sparc station
here in front of me, there is no compress option for tar (thats one way
to avoid the problem).
--
Bil Kilgallon ([EMAIL PROTECTED])
--"I believe, what I believe, has made me what I am. I did not make
it, It is making me, it is the very truth of God, not the invention
of any man". Rich Mullins, quoting G.K. Chesterton.
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
------------------------------
** 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
******************************
