Linux-Misc Digest #970, Volume #26 Tue, 30 Jan 01 11:13:02 EST
Contents:
Re: Tar to Tape Above Capacity (-ljl-)
Re: I wish to RTFM, but where is TFM I need? ("Tom Edelbrok")
Adding Win98SE to W2K and Linux system. (-ljl-)
Re: Tar to Tape Above Capacity (Bill Buchan)
Re: at queue problems ("Tom Edelbrok")
rpm failed dependencies? (Bob Terrell)
Re: help please ("Tom Edelbrok")
Re: I wish to RTFM, but where is TFM I need? (Jean-David Beyer)
Re: I wish to RTFM, but where is TFM I need? ("Peter T. Breuer")
Re: rpm failed dependencies? (Enrico Lunghi)
Re: at queue problems (Mike E.)
Re: Slow Disk Check for Large Drives (Robert Heller)
Re: Tar to Tape Above Capacity (Robert Heller)
Re: Adding Win98SE to W2K and Linux system. (Eric)
Re: copying /dev/* files (Eric)
smp on pc server 330? ("M Osborne")
----------------------------------------------------------------------------
From: -ljl- <[EMAIL PROTECTED]>
Subject: Re: Tar to Tape Above Capacity
Date: Tue, 30 Jan 2001 14:11:16 GMT
In article <9564b3$mho$[EMAIL PROTECTED]>,
Bill Buchan <[EMAIL PROTECTED]> wrote:
> Hi,
>
> I've inherited a machine running Suse6.4 Linux which has a
> DDS4(20/40Gb) tape drive attached.
>
> There is a cron job already on this machine which writes a tar archive
> to tape of every file (tar -cvf /u01 /u02 /u03 /u04).
>
> It then verifies the tar archive (tar -tvf /dev/rmt0).
I'm running TurboLinux 6.
ls -l /dev/rmt*
crw-rw---- 1 root disk 12, 8 Apr 17 1999 /dev/rmt16
crw-rw---- 1 root disk 12, 6 Apr 17 1999 /dev/rmt8
Looking in "/usr/src/linux/Documentation/devices.txt":
12 char QIC-02 tape
2 = /dev/ntpqic11 QIC-11, no rewind-on-close
3 = /dev/tpqic11 QIC-11, rewind-on-close
4 = /dev/ntpqic24 QIC-24, no rewind-on-close
5 = /dev/tpqic24 QIC-24, rewind-on-close
6 = /dev/ntpqic120 QIC-120, no rewind-on-close
7 = /dev/tpqic120 QIC-120, rewind-on-close
8 = /dev/ntpqic150 QIC-150, no rewind-on-close
9 = /dev/tpqic150 QIC-150, rewind-on-close
DDS-4 ? I've an old DDS-DC and it uses "/dev/nst0" and "/dev/st0".
> However the total amount it writes is about 58Gb which is way above
> the maximum capacity of the tape. The log files for writing and
> verifying the tape contain the same list of files, and if you sum
> up the number of bytes of all the files in the verification log you
> get 58Gb.
>
> How can this be?
Compression? Most advertisements assume 2x (e.g. 20/40), it can be
higher (or lower). Some backup programs assume something like 1.3.
Create a file of zeros and gz it. Another explanation, what is the
numeric value of a GB.
Have you tried to recover one of the files near the end of tape. Use
tar's -t switch and see what is actually on the tape.
--
Louis-ljl-{ Louis J. LaBash, Jr. }
Sent via Deja.com
http://www.deja.com/
------------------------------
From: "Tom Edelbrok" <[EMAIL PROTECTED]>
Subject: Re: I wish to RTFM, but where is TFM I need?
Date: Tue, 30 Jan 2001 06:26:00 -0800
I think you've got way too much junk in there. I am using ppp right now to
send you this message, from Linux Redhat 6.0. It has worked reliably for
about 8 months now, and I only have 3 lines in
/etc/sysconfig/network-scripts/ifcfg-ppp0 as follows:
ONBOOT=yes
PERSIST=yes
HOLDOFF=70
One problem I do have is that the line drops consistently after about 15
minutes of idle connection time, but as long as I'm using the net the line
stays up forever.
Tom
Jean-David Beyer <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> I have gotten ppp to work by guess and by golly, but I would prefer to
> not be shooting in the dark as I am now. The "manual" I need pertains to
> file /etc/sysconfig/network-scripts/ifcfg-ppp0. It contains the
> following lines at the moment, but I would like to know the definitions
> of all the items listed (some are obvious to me, but some are a
> mystery), and also those others that are not there, but might be, in
> sufficient detail to make intelligent decisions as to what should be in
> there.
>
> PERSIST=no
> DEFROUTE=yes
> ONBOOT=no
> INITSTRING=ATZ
> MODEMPORT=/dev/modem
> LINESPEED=115200
> ESCAPECHARS=no
> DEFABORT=yes
> HARDFLOWCTL=yes
> DEVICE=ppp0
> PPPOPTIONS=
> DEBUG=yes
> PAPNAME=
> PEERDNS=no
> REMIP=
> IPADDR=
> WVDIALSECT=
> BOOTPROTO=none
> MTU=
> MRU=
> DISCONNECTTIMEOUT=5
> RETRYTIMEOUT=10
> USERCTL=yes
>
> --
> .~. Jean-David Beyer Registered Linux User 85642.
> /V\ Registered Machine 73926.
> /( )\ Shrewsbury, New Jersey
> ^^-^^ 9:05am up 1 day, 17:35, 3 users, load average: 1.15, 1.13, 1.16
------------------------------
From: -ljl- <[EMAIL PROTECTED]>
Subject: Adding Win98SE to W2K and Linux system.
Date: Tue, 30 Jan 2001 14:17:00 GMT
Hi,
Currently, I have two drives in this machine and wish to add Win98SE.
Drive 0 has DOS (FAT-16), and NT4 (NTFS); drive 1 has an FAT partition,
W2K (NTFS), and Linux (ext2 & swap). Both drives have => 1GB of space
that can be used for the Win98SE installation.
I'm using W2K's boot-manager (BM) to boot any of these OSes; DOS and
Linux have boot-sector files loaded by BM.
Any and all help will be appreciated.
--
Louis-ljl-{ Louis J. LaBash, Jr. }
Sent via Deja.com
http://www.deja.com/
------------------------------
From: Bill Buchan <[EMAIL PROTECTED]>
Subject: Re: Tar to Tape Above Capacity
Date: Tue, 30 Jan 2001 14:22:07 GMT
In article <956i1u$120$[EMAIL PROTECTED]>,
-ljl- <[EMAIL PROTECTED]> wrote:
> In article <9564b3$mho$[EMAIL PROTECTED]>,
> Bill Buchan <[EMAIL PROTECTED]> wrote:
> > Hi,
> >
> > I've inherited a machine running Suse6.4 Linux which has a
> > DDS4(20/40Gb) tape drive attached.
> >
> > There is a cron job already on this machine which writes a tar
archive
> > to tape of every file (tar -cvf /u01 /u02 /u03 /u04).
> >
> > It then verifies the tar archive (tar -tvf /dev/rmt0).
>
> I'm running TurboLinux 6.
> ls -l /dev/rmt*
> crw-rw---- 1 root disk 12, 8 Apr 17 1999 /dev/rmt16
> crw-rw---- 1 root disk 12, 6 Apr 17 1999 /dev/rmt8
>
> Looking in "/usr/src/linux/Documentation/devices.txt":
>
> 12 char QIC-02 tape
> 2 = /dev/ntpqic11 QIC-11, no rewind-on-close
> 3 = /dev/tpqic11 QIC-11, rewind-on-close
> 4 = /dev/ntpqic24 QIC-24, no rewind-on-close
> 5 = /dev/tpqic24 QIC-24, rewind-on-close
> 6 = /dev/ntpqic120 QIC-120, no rewind-on-close
> 7 = /dev/tpqic120 QIC-120, rewind-on-close
> 8 = /dev/ntpqic150 QIC-150, no rewind-on-close
> 9 = /dev/tpqic150 QIC-150, rewind-on-close
>
> DDS-4 ? I've an old DDS-DC and it uses "/dev/nst0" and "/dev/st0".
Actually /dev/rmt0 is just a symbolic link to /dev/st0 on my system.
I don't know if it comes that way or if somebody set it up. What
are /dev/st1, /dev/st0m etc.?
>
> > However the total amount it writes is about 58Gb which is way above
> > the maximum capacity of the tape. The log files for writing and
> > verifying the tape contain the same list of files, and if you sum
> > up the number of bytes of all the files in the verification log you
> > get 58Gb.
> >
> > How can this be?
>
> Compression? Most advertisements assume 2x (e.g. 20/40), it can be
> higher (or lower). Some backup programs assume something like 1.3.
> Create a file of zeros and gz it. Another explanation, what is the
> numeric value of a GB.
>
> Have you tried to recover one of the files near the end of tape. Use
> tar's -t switch and see what is actually on the tape.
>
Thanks for your reply.
I tried to recover a file at the end of the tape as you suggest
and got the error "Not found in archive". So it must be overrunning the
end of tape.
But this file appears in the list generated by tar -tvf /dev/rmt0. Is
this 't' verification just checking an index at the top of the tar file
rather than checking if the file is actually there?
Thanks
- Bill.
> --
> Louis-ljl-{ Louis J. LaBash, Jr. }
>
> Sent via Deja.com
> http://www.deja.com/
>
Sent via Deja.com
http://www.deja.com/
------------------------------
From: "Tom Edelbrok" <[EMAIL PROTECTED]>
Subject: Re: at queue problems
Date: Tue, 30 Jan 2001 06:32:16 -0800
This may be of only little help to you in isolating the problem but I have a
similar problem with Redhat 6.0 when I get the "at" command to run certain
scripts from within /etc/rc.d/rc.local at boot-up time. They will sit in the
queue forever, even though a display of the queue shows that the time they
are scheduled to run has approached, come, and then gone. I ended up solving
the problem by using crontab instead, and within the crontab-executed script
I place some bash lines that ensured the script only performs its intended
function once. A dumb band-aid solution, but no one has yet been able to
tell me why some batch jobs go into never-never land.
Tom
Gregory A Greenman <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> I have two users users set up on my Mandrake Linux 7.0
> computer that are both trying to use the at queue to run
> programs. They both have a shell script that calls the at
> command. The line where at is called is:
>
> at -q b -f diprun $2 $3 $4
>
> This works fine for one of the users, but for the other, when
> the appropriate time comes, diprun does not execute. It just
> stays in the queue indefinitely without ever executing.
>
> If anyone knows why this is happening and how to fix it, I'd
> greatly appreciate your help.
>
>
>
>
> Greg
> ------------------
> greg -at- spencersoft -dot- com
------------------------------
From: Bob Terrell <[EMAIL PROTECTED]>
Subject: rpm failed dependencies?
Date: Tue, 30 Jan 2001 09:34:19 -0500
Hi. How do I find out which package provides the
missing libwraster.so.1?
# rpm -i AfterStep*
error: failed dependencies:
libwraster.so.1 is needed by AfterStep-APPS-20000124-6
(I can't see how the --whatprovides stuff works.)
thanks,
Bob
------------------------------
From: "Tom Edelbrok" <[EMAIL PROTECTED]>
Subject: Re: help please
Date: Tue, 30 Jan 2001 06:39:07 -0800
I had a similar problem with Redhat Linux 6.0 using a standard PC with an
NEC 3D screen. Every time I went into XWindows I got only a portion of the
viewing area, and as a result couldn't get to certain buttons that had
disappeared off the screen.
I tried all of the graphic options within the install procedure and could
not get the thing to work properly, (I have a Creative Labs Blaster Banshee
16meg 3DFX card).
Finally I went through the detailed setup where you specify the horizontal
and vertical frequencies, and all the rest of the absolute nitty gritty for
the screen and card. Now it works fine. But it didn't work as long as I
stayed in the standard RedHat installer with all of its useless options.
Tom
gvary <[EMAIL PROTECTED]> wrote in message
news:odkd6.7833$[EMAIL PROTECTED]...
> I just got Corel linux but my problem is when i try to install it the
screen
> is too large for my monitor, so im unable to see the install button. My
> guess is that its something related to my video card. My card is Neomagic
> 256Avp.
> My laptop display is 1024x768.
> Can anyone help?
> [EMAIL PROTECTED]
>
>
>
>
------------------------------
From: Jean-David Beyer <[EMAIL PROTECTED]>
Subject: Re: I wish to RTFM, but where is TFM I need?
Date: Tue, 30 Jan 2001 09:44:00 -0500
Tom Edelbrok wrote:
>
> I think you've got way too much junk in there. I am using ppp right now to
> send you this message, from Linux Redhat 6.0. It has worked reliably for
> about 8 months now, and I only have 3 lines in
> /etc/sysconfig/network-scripts/ifcfg-ppp0 as follows:
>
> ONBOOT=yes
> PERSIST=yes
> HOLDOFF=70
>
Could be, but It came almost like that with my distro. I had to add the
line:
PEERDNS=no
to keep it from overwriting my /etc/resolv.conf file,
and I changed
PERSIST=yes to PERSIST=no
in an attempt to keep it from redialing as soon as I took down the
connection to my ISP. I am not sure what that is for, but it is one of
the things I did to keep the /sbin/ppp-watch program from re-dialing as
soon as I hung up.
I do not dare take your advise to use only what you have there. I want
to RTFM so I will know what I am doing, and to do that, I NEED TFM.
--
.~. Jean-David Beyer Registered Linux User 85642.
/V\ Registered Machine 73926.
/( )\ Shrewsbury, New Jersey
^^-^^ 9:40am up 1 day, 18:10, 3 users, load average: 1.17, 1.12, 1.09
------------------------------
From: "Peter T. Breuer" <[EMAIL PROTECTED]>
Subject: Re: I wish to RTFM, but where is TFM I need?
Date: Tue, 30 Jan 2001 14:55:44 GMT
Jean-David Beyer <[EMAIL PROTECTED]> wrote:
> file /etc/sysconfig/network-scripts/ifcfg-ppp0. It contains the
> following lines at the moment, but I would like to know the definitions
> of all the items listed (some are obvious to me, but some are a
:-) A vain wish .. you'll have to talk to the author of the script,
probably!
> mystery), and also those others that are not there, but might be, in
> sufficient detail to make intelligent decisions as to what should be in
> there.
> PERSIST=no
Dunno .. possibly a "don't die when the connnection drops". Or
possibly a "try again when the connection dies".
> DEFROUTE=yes
let pppd (try to) set up a default route
> ONBOOT=no
This script must be executable at boottime! They're saying not to start
pppd at that time.
> INITSTRING=ATZ
The string used to initialize the modem
> MODEMPORT=/dev/modem
The modem special device (this should not be /dev/modem, but should
instead be a /dev/ttyS? ... without that information, there is some
danger of locking failing through different naming conventions across
different applications - mgetty being the principal one I have in
mind).
> LINESPEED=115200
The initial line training info for talking to the modem.
> ESCAPECHARS=no
uh, I think that's a ppp thing. You can escape certain characters or
otherwise leave them out of the commuunications. If it's a serial port
thing I'm at a loss to guess what it is.
> DEFABORT=yes
dunno.
> HARDFLOWCTL=yes
Modem flow control protoocl.
> DEVICE=ppp0
The ppp device name
> PPPOPTIONS=
any last wishes?
> DEBUG=yes
Good news for manufacturers of hard disks.
> PAPNAME=
uh, in case you want to autheticate yourself in an unusual way to the
other side. This overrides the default (machine name).
> PEERDNS=no
I think this says to NOT get dns from the peer! That doesn't seem
consistent with what youy allowed above.
> REMIP=
You can set the IP address of the remote here. Presumably only if you
are serving?
> IPADDR=
Your preferred IP address, if you want to override what's given you.
The peer might not accept your choice!
> WVDIALSECT=
Eh? Something for wvdial? What?
> BOOTPROTO=none
Dunno? Are they talking about bootp? Are they asking if we want to boot
over ppp? They're crazy!
> MTU=
Obviously, if you want to override the default MTU for the connection.
It's often a good idea to set it low on noisy lines.
> MRU=
ditto MRU.
> DISCONNECTTIMEOUT=5
Some timeout thing. Dunno for what exactly.
> RETRYTIMEOUT=10
Obviously, if retrying (persist?), the interval between redial
attempts.
> USERCTL=yes
Eh?
Peter
------------------------------
From: Enrico Lunghi <[EMAIL PROTECTED]>
Subject: Re: rpm failed dependencies?
Date: Tue, 30 Jan 2001 15:52:45 +0000
Bob Terrell wrote:
>
> Hi. How do I find out which package provides the
> missing libwraster.so.1?
>
> # rpm -i AfterStep*
> error: failed dependencies:
> libwraster.so.1 is needed by AfterStep-APPS-20000124-6
>
> (I can't see how the --whatprovides stuff works.)
>
> thanks,
> Bob
Hi Bob.
This not the most elegant and hortodox way to solve the problem but it
works.
Just go to this site
http://rpmfind.net/linux/RPM/index.html
put there the name of the package you are installing and go to its html
page. There you can see what it provides and what it requires.
Follow the link of the required library that you need and you'll get a list
of rpm's which contain what you need.
By the way, libwraster.so.1 is provided, for instance, by
wmaker-0.61.1-82-i386.rpm .
But you should check which architecture you actually need.
Bye,
Enrico
------------------------------
From: Mike E. <[EMAIL PROTECTED]>
Subject: Re: at queue problems
Date: Tue, 30 Jan 2001 15:03:22 GMT
In article <49Ad6.61407$[EMAIL PROTECTED]>,
"Tom Edelbrok" <[EMAIL PROTECTED]> wrote:
> This may be of only little help to you in isolating the problem but I
have a
> similar problem with Redhat 6.0 when I get the "at" command to run
certain
> scripts from within /etc/rc.d/rc.local at boot-up time. They will sit
in the
> queue forever, even though a display of the queue shows that the time
they
> are scheduled to run has approached, come, and then gone. I ended up
solving
> the problem by using crontab instead, and within the crontab-executed
script
> I place some bash lines that ensured the script only performs its
intended
> function once. A dumb band-aid solution, but no one has yet been able
to
> tell me why some batch jobs go into never-never land.
>
> Tom
>
> Gregory A Greenman <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]...
> > I have two users users set up on my Mandrake Linux 7.0
> > computer that are both trying to use the at queue to run
> > programs. They both have a shell script that calls the at
> > command. The line where at is called is:
> >
> > at -q b -f diprun $2 $3 $4
> >
> > This works fine for one of the users, but for the other, when
> > the appropriate time comes, diprun does not execute. It just
> > stays in the queue indefinitely without ever executing.
> >
> > If anyone knows why this is happening and how to fix it, I'd
> > greatly appreciate your help.
> >
> >
> >
> >
> > Greg
> > ------------------
> > greg -at- spencersoft -dot- com
>
>
Not really much help again, but I was taught that the AT command was
designed more for scheduling a one time task and the CRON function was
more for running repeated tasks. If one user can use AT and not the
other though, it is probably a permission problem or else they do not
belong to the right group (which is really a permission problem also).
Mike
--
Support provided by Linuxgruven, Inc.
http://www.linuxgruven.com
314-727-0918
Sent via Deja.com
http://www.deja.com/
------------------------------
From: Robert Heller <[EMAIL PROTECTED]>
Subject: Re: Slow Disk Check for Large Drives
Date: 30 Jan 2001 09:17:33 -0600
Jean-David Beyer <[EMAIL PROTECTED]>,
In a message on Tue, 30 Jan 2001 09:06:01 -0500, wrote :
JB> [EMAIL PROTECTED] wrote:
JB> >
JB> > I'm leading up the development of a Linux-based product that requires
JB> > alot of storage space. The main problem that we are experiencing comes
JB> > when the system loses power unexpectedly and reboots. The following
JB> > disk check takes forever if the system has 3 or 4 60 Gb disks. We
JB> > really need to speed up this reboot process or we may be forced into
JB> > using Windows 2000 (which I loathe). Even with a UPS connected, we
JB> > have to take into account the fact that some people will just power the
JB> > thing on and off.
JB> >
JB> If you cannot afford the time it takes to do the fsck's (some of these
JB> can be done in parallel to speed it up if you have different drives),
JB> perhaps your system(s) deserve to be run from a UPS that can coast over
JB> the short-term power interruptions, and cause a controlled shut-down for
JB> the longer ones. That way, the fsck's can be bypassed.
And you need to seriously train your people NOT to randomly power down
the box! Stick a sign (PostIt note?) about the proper shutdown
procedure. Note: if it boots to console mode, Ctrl-Alt-Del will do a
clean re-boot -- you can power off the box as soon as the BIOS comes up
with its memory test. If you have it set up as a GUI workstation, most
of the modern system (Gnome, KDE, etc.) have a hack (ala WinNT-ish) to
allow anyone to do a clean shutdown instead of just logging out. At
worst, they can select a different virtual terminal (Ctrl-Alt-F2) and do
a Ctrl-Alt-Del to start a re-boot.
JB>
JB> --
JB> .~. Jean-David Beyer Registered Linux User 85642.
JB> /V\ Registered Machine 73926.
JB> /( )\ Shrewsbury, New Jersey
JB> ^^-^^ 9:00am up 1 day, 17:30, 3 users, load average: 1.23, 1.13, 1.17
JB>
--
\/
Robert Heller ||InterNet: [EMAIL PROTECTED]
http://vis-www.cs.umass.edu/~heller || [EMAIL PROTECTED]
http://www.deepsoft.com /\FidoNet: 1:321/153
Posted Via Nuthinbutnews Premium Usenet Newsgroup Services
==========================================================
** SPEED ** RETENTION ** COMPLETION **
==========================================================
http://www.nuthinbutnews.com
------------------------------
From: Robert Heller <[EMAIL PROTECTED]>
Subject: Re: Tar to Tape Above Capacity
Date: 30 Jan 2001 09:17:35 -0600
Bill Buchan <[EMAIL PROTECTED]>,
In a message on Tue, 30 Jan 2001 14:22:07 GMT, wrote :
BB> Actually /dev/rmt0 is just a symbolic link to /dev/st0 on my system.
BB> I don't know if it comes that way or if somebody set it up. What
BB> are /dev/st1, /dev/st0m etc.?
/dev/st1 is the second tape drive (if any). /dev/nst0 is the
non-rewinding version of /dev/st0.
BB> I tried to recover a file at the end of the tape as you suggest
BB> and got the error "Not found in archive". So it must be overrunning the
BB> end of tape.
No, you need to specify the file *exactly* as tar stored it. For
example, I also use tar. Here is a fragment of my tar listing:
-rw-r--r-- root/root 4 2001-01-28 15:05 ./var/spool/lpd/color/lock
-rw-rw-r-- root/root 30 2001-01-28 14:12 ./var/spool/lpd/color/status
-rw-r----x root/root 4 2001-01-28 14:47
./var/spool/lpd/fastcolor/.seq
-rw-r--r-- root/root 4 2001-01-28 15:05
./var/spool/lpd/fastcolor/lock
-rw-rw-r-- root/root 30 2001-01-28 14:47
./var/spool/lpd/fastcolor/status
-rw-r----x root/root 4 2001-01-25 17:27 ./var/spool/lpd/lp0/.seq
-rw-r----- root/root 4 2001-01-28 15:05 ./var/spool/lpd/lp0/lock
-rw-rw-r-- root/root 30 2001-01-25 17:27 ./var/spool/lpd/lp0/status
-rw-rw---- heller/mail 0 2001-01-28 15:17 ./var/spool/mail/heller
-rw------- news/news 8192 2001-01-16 16:01 ./var/spool/news/core
-rw-rw-r-- news/news 0 2001-01-25 16:28
./var/spool/news/innfeed/innfeed-dropped.A000440
-rw-rw-r-- news/news 0 2001-01-28 15:05
./var/spool/news/innfeed/innfeed-dropped.A000443
-rw-rw-r-- news/news 0 2001-01-14 03:45
./var/spool/news/innfeed/innfeed-dropped.A000452
-rw-rw-r-- news/news 64361 2000-12-19 01:30
./var/spool/news/lhbbs/r-fidonews/100
-rw-rw-r-- news/news 7585 2000-12-19 01:30
./var/spool/news/lhbbs/r-fidonews/101
-rw-rw-r-- news/news 64274 2000-12-27 01:30
./var/spool/news/lhbbs/r-fidonews/102
-rw-rw-r-- news/news 11995 2000-12-27 01:30
./var/spool/news/lhbbs/r-fidonews/103
If I wanted to recover the file /var/spool/news/lhbbs/r-fidonews/103, I
would have to specify:
tar xv ./var/spool/news/lhbbs/r-fidonews/103
NOT tar xv /var/spool/news/lhbbs/r-fidonews/103
'/var/spool/news/lhbbs/r-fidonews/103' is not in the archive,
'./var/spool/news/lhbbs/r-fidonews/103' is. Same file, different
'name'. Also GNU tar will strip off a leading / (convert from abs to
relative name). (Tar is keeping the leading '/' on my backups because
I am doing a cd /, then tar cv ., rather than a tar cv / -- same files
just a *slightly* different way of naming them.) The advantage of
always using relative names in a tar archive is that it allows you to
restore the files someplace other than where they were backed up from --
good for comparing or for recovery from say a rescue system, where
'/' is not the real system (eg a rescue RAMDISK/Zip/Jaz/CD-ROM or a
completely different box -- hdd transplant games).
BB>
BB> But this file appears in the list generated by tar -tvf /dev/rmt0. Is
BB> this 't' verification just checking an index at the top of the tar file
BB> rather than checking if the file is actually there?
Tar stores the 'index' as a 512 byte header in front of each file on the
table:
Typical tar file (on tape or otherwise):
[512-byte tar header][file data][512-byte tar header][file data]...[1024 bytes of
zeros]
BB>
BB>
BB> Thanks
BB> - Bill.
BB>
BB>
BB> > --
BB> > Louis-ljl-{ Louis J. LaBash, Jr. }
BB> >
BB> > Sent via Deja.com
BB> > http://www.deja.com/
BB> >
BB>
BB>
BB> Sent via Deja.com
BB> http://www.deja.com/
BB>
--
\/
Robert Heller ||InterNet: [EMAIL PROTECTED]
http://vis-www.cs.umass.edu/~heller || [EMAIL PROTECTED]
http://www.deepsoft.com /\FidoNet: 1:321/153
Posted Via Nuthinbutnews Premium Usenet Newsgroup Services
==========================================================
** SPEED ** RETENTION ** COMPLETION **
==========================================================
http://www.nuthinbutnews.com
------------------------------
From: Eric <[EMAIL PROTECTED]>
Subject: Re: Adding Win98SE to W2K and Linux system.
Date: Tue, 30 Jan 2001 16:22:10 +0100
Reply-To: [EMAIL PROTECTED]
-ljl- wrote:
>
> Hi,
> Currently, I have two drives in this machine and wish to add Win98SE.
>
> Drive 0 has DOS (FAT-16), and NT4 (NTFS); drive 1 has an FAT partition,
> W2K (NTFS), and Linux (ext2 & swap). Both drives have => 1GB of space
> that can be used for the Win98SE installation.
>
> I'm using W2K's boot-manager (BM) to boot any of these OSes; DOS and
> Linux have boot-sector files loaded by BM.
>
> Any and all help will be appreciated.
Best help I can give you:
look at where you posted
comp.os.linux.misc
What does this have to do with linux?
repost at comp.semi-os.win98.misc
Eric
------------------------------
From: Eric <[EMAIL PROTECTED]>
Subject: Re: copying /dev/* files
Date: Tue, 30 Jan 2001 16:27:49 +0100
Reply-To: [EMAIL PROTECTED]
Juergen Pfann wrote:
>
> Eric wrote:
> >
> > [EMAIL PROTECTED] wrote:
> > >
> > > Patrick Machado <[EMAIL PROTECTED]> wrote:
> > > : The problem arises when I atempt to copy the files located in /dev.
> > > : Does anybody know how to copy this special files to another dir?
> > >
> > > You can not copy "/dev/files". They are not files.
> >
> > yes you can and yes they are.
> > They are special files though.
> >
> > /home/moorse>sudo cp -a /dev/sda1 .
> > /home/moorse>ls -al ./sda1
> > brw-rw---- 1 root disk 8, 1 May 5 1998 ./sda1
> >
>
> Correct - but you should stress the importance of the "-a"
> option, which is a shortcut for "-dpR" (man cp). Without this
> option, the poor user would destroy his home directory and
> hence the fs it is on with the contents of the first SCSI
> disk !!!
> So - be careful when dealing with devices, please...
>
People that don't think before they act
shouldn't get permission to use sudo
(or any root priviliges for that matter)
But.... I agree, copying to . is bad practice too
<shame shame shame>
so it should have read:
`sudo cp -a /dev/sda1 ./sda1`
Eric
------------------------------
From: "M Osborne" <[EMAIL PROTECTED]>
Subject: smp on pc server 330?
Date: Tue, 30 Jan 2001 09:21:04 -0600
Reply-To: "M Osborne" <[EMAIL PROTECTED]>
I'm trying to get smp support to work with linux on an IBM PC Server 330.
This server has 2 PPro 200's.
When I install the RPM for Redhat 7 and change the lilo.conf to point to the
new smp kernel, I get a kernel panic on reboot (at the VFS part where it's
trying to mount root).
I tried Mandrake too, but it does not automatically load the smp support for
my machine. Is there something special I'm supposed to do at install time?
Thank you.
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: [EMAIL PROTECTED]
You can send mail to the entire list by posting to comp.os.linux.misc.
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
******************************
