Linux-Misc Digest #672, Volume #27 Sat, 21 Apr 01 18:13:01 EDT
Contents:
Re: WM that uses scripting? (Robert Heller)
Re: Could Linux be used in this factory environment ? (The Ghost In The Machine)
Re: RAID 5 or 0 for performance? (Dan Smith)
Re: RAID 5 or 0 for performance? ("Ron Reaugh")
Problem with: PPPD dying with HangUp (SIGHUP) (Joachim Aurbacher)
Re: KWord crashes ("Markus Reitz")
Re: KWord crashes ("Markus Reitz")
Re: Could Linux be used in this factory environment ? ("Monte Milanuk")
Re: RedHat 7.1 too early ?,... ("Monte Milanuk")
Re: My address is very flawed! ("Jim C. Brown")
Upgrade RPM3 to RPM4 (George Trapkov)
help on the boot option (Hui Ji)
Re: top level dir permissions (Juergen Heinzl)
Re: top level dir permissions (Michael Heiming)
Re: MySQL Problem. (Jack Pan)
Re: error when mounting FAT32 partition: kernel: fatfs: bogus cluster size (Dances
With Crows)
reading multisession cd's (Shea A Martin)
Voodoo (usual) problems (Bruno Barberi Gnecco)
Re: Reliability of "time" command? (Jean-David Beyer)
----------------------------------------------------------------------------
From: Robert Heller <[EMAIL PROTECTED]>
Subject: Re: WM that uses scripting?
Date: 21 Apr 2001 19:29:43 GMT
[EMAIL PROTECTED] (Andrew Purugganan),
In a message on 21 Apr 2001 13:43:39 GMT, wrote :
AP> I don't know if you're familiar with Litestep which runs on top of that
AP> OS from Redmond, which allows me to make it look like I'm running Linux
AP> instead. It's a bit like getting your themes but with the flexibility of
AP> scripting which allows me to maximize deskspace atthe same time. Bothered
AP> by wharf? A little button tucks it out of sight. Toolbar too big? A
AP> similar button. Or, one button tucks them all away. Or skip the wharf if
AP> you like. Which WM allows me to do that?
Basically *all* of them allow some degree of what you call scripting.
There are some WM/Desktop packages that are *written* in a scripting
language (TkDesk -- written in Tcl/Tk).
Note further: the *source code* for all of the WM/Desktop packages for
Linux are available (and linux comes with the compilers, etc. needed to
(re-)build them), so you have the *flexibility* to customize a WM or
Desktop at that level as well.
AP>
AP> I would like to port some of those themes to my Linux box. If you want
AP> some theme ideas feel free to browse litestep.net for inspiration ;-)
AP> --
AP> jazz
AP> Registered linux user no. 164098 +--+--+--+ Litestep user no. 386
AP> Doesn't it bother you, that we have to search for intelligent life
AP> --- OUT THERE??
AP>
--
\/
Robert Heller ||InterNet: [EMAIL PROTECTED]
http://vis-www.cs.umass.edu/~heller || [EMAIL PROTECTED]
http://www.deepsoft.com /\FidoNet: 1:321/153
------------------------------
From: [EMAIL PROTECTED] (The Ghost In The Machine)
Crossposted-To: comp.os.linux.advocacy,comp.os.linux.hardware
Subject: Re: Could Linux be used in this factory environment ?
Date: Sat, 21 Apr 2001 19:16:36 GMT
In comp.os.linux.advocacy, Charles Lyttle
<[EMAIL PROTECTED]>
wrote
on Sat, 21 Apr 2001 14:30:09 GMT
<[EMAIL PROTECTED]>:
>Erik Funkenbusch wrote:
>>
[snip for brevity]
>>
>> Further, the same author that wrote the gcn article (which quotes from the
>> usni article) also clarifies his statements in a followup article:
>> http://www.gcn.com/archives/gcn/1998/november9/6.htm
>That last one is even worse than my story. A divide by zero in the
>controller for a fuel valve caused the entire LAN to go down crashing 27
>remotes? Industry (mostly) fixed that problem 30 years ago. For what its
>worth, I had an NT machine I was working with bring down an entire LAN
>of over 1000 machines. It was called the "ping of death". Some
>applications could cause the NT software to start issuing network pings
>at high speed. These faults often also caused a BSOD, but not always.
Yeah, but is the BSOD on the machine pinging, or one of the
many machines being pinged? :-)
[.sigsnip]
--
[EMAIL PROTECTED] -- insert random misquote here
EAC code #191 5d:12h:04m actually running Linux.
We are all naked underneath our clothes.
------------------------------
From: Dan Smith <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.hardware,comp.periphs.scsi
Subject: Re: RAID 5 or 0 for performance?
Date: 21 Apr 2001 15:06:42 -0400
I see it, but I thought that it was a reference signal. I didn't
think I was supposed to put a jumper on it...
(These are SCA drives, so the spindle sync is on the adapter...)
Thanks for helping me with this..
--Dan
------------------------------
From: "Ron Reaugh" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.hardware,comp.periphs.scsi
Subject: Re: RAID 5 or 0 for performance?
Date: Sat, 21 Apr 2001 19:40:50 GMT
Dan Smith wrote in message ...
>I see it, but I thought that it was a reference signal. I didn't
>think I was supposed to put a jumper on it...
Right.
>(These are SCA drives, so the spindle sync is on the adapter...)
------------------------------
Date: Sat, 21 Apr 2001 21:54:40 +0200
From: Joachim Aurbacher <[EMAIL PROTECTED]>
Subject: Problem with: PPPD dying with HangUp (SIGHUP)
Hello,
i 've got a similar problem as Bill as well.
My system ist RH 6.2 with kernel 2.2.5-15 and pppd 2.3.7 with KDE (kppp
1.6.10).
When dialing up with kppp pppd dies unexpectedly without any obvious
reason few
seconds after connect.
(/var/log/messages: Hangup (SIGHUP)) This ist the same with all
providers.
I have tryed everything the kppp-help says (kppp and pppd setuid root,
no lock,
/dev/ttyS0 instead of /dev/modem, wait 1 second, send ppp ) but it still
does
not work.
I can although dial up to just one provider (ras.uni-hohenheim.de) with
netcfg
but all the other providers don't work in this manner neither. Within
MS-Windows everything works well.
What can I still do? I would appreciate any help
Joachim Aurbacher
Bill Unruh wrote:
> In <3ad83846@grissom> "Greg Williams" <[EMAIL PROTECTED]> writes:
>
> >When dialling up to the Internet, it connects ok, sometimes after a redial
> >or retrain (don't worry about that, i'll try some modem strings or
> >something). Sometimes it drops out. The log (/var/log/messages) doesn't
> >give an termination signal, it says Hangup (SIGHUP).
>
>
------------------------------
From: "Markus Reitz" <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux.mandrake,comp.os.linux.help,linux.dev.newbie,linux.help
Subject: Re: KWord crashes
Date: Sat, 21 Apr 2001 12:25:59 +0200
I think this isn't a problem of Mandrake but KOffice. Can you reproduce the
error? If so, have a look at the KOffice-Bug-List and if the error isn't
already mentioned, post a bug report.
Markus
------------------------------
From: "Markus Reitz" <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux.mandrake,comp.os.linux.help,linux.dev.newbie,linux.help
Subject: Re: KWord crashes
Date: Sat, 21 Apr 2001 12:34:08 +0200
> An application mostly receives the SIGEGV signal due to a bug in the
> application. The application was asked to save it's documents.
>
> Is there anyway to fix this? Actually I've had a couple other programs
crash
> on me.
Sure, there is always a way to fix it. Get the sources, find the bug,
correct it and the problem disappears ;-)
Which other programs crash and what are the error messages?
Markus
------------------------------
From: "Monte Milanuk" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.advocacy,comp.os.linux.hardware
Subject: Re: Could Linux be used in this factory environment ?
Date: Sat, 21 Apr 2001 20:26:12 GMT
Charles Lyttle <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> A good project for Linux open source would be a ladder logic
> development/run-time enviornment. It should have provisions for
> constructing and printing ladder diagrams. It should have a runtime
> enviornment that intreprets the diagram routing signals to/from the
> external hardware. For emergency situations, it should have provisions
> for over-riding the state of objects (password required).
>
I know... it's one of the things that is driving me to try to learn to
program... i.e. I'm very used to ladder logic, after dealing w/ large
programs on Reliance Automax DCS and AB PLC5 racks, and since in my new job
I will probably be not allowed near the techie type stuff (I'm an operator,
now ;p ) I will have to settle for doing some home automation stuff. And to
me, it seems like a decent ladder logic implementation would be just the
ticket for stuff like irrigation systems, lighting schemes, burglar alarms,
and what not. Hopefully someone more skilled than I will start it, because
if it has to wait for me to learn enough to start it, it'll be a long time
coming ;)
Monte
------------------------------
From: "Monte Milanuk" <[EMAIL PROTECTED]>
Subject: Re: RedHat 7.1 too early ?,...
Date: Sat, 21 Apr 2001 20:26:13 GMT
Christian Rose <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Just run up2date once in a while. Or take real use of RHN and create
> system profiles (basically "what packages are installed") for your
> machines. Then you can go to RHN once in a while and see exactly which
> updates apply to which machine.
>
> Or go even one step further - select the "Software Manager" level for
> one of your machines - then you will get errata reports for that system
> mailed to you, specifically targeted to the software you have on that
> machine. You can even get them automatically installed, if you like that
> kind of thing.
>
Sounds good... just not sure about having a 'profile' of my machines
residing on someone else's box. I don't think RedHat is going to try
hacking my machines anytime soon, but it just doesn't give me a warm fuzzy,
nonetheless.
Just an idea percolating around in my head... I'll have to deal w/ it more
when I get a house bought and move into and connectivity restored to my
Linux boxen.
Thanks
Monte
------------------------------
From: [EMAIL PROTECTED] ("Jim C. Brown")
Subject: Re: My address is very flawed!
Date: Sat, 21 Apr 2001 20:30:17 +0000 (UTC)
Are you sure? I tried it, but it is still incorrect.
If it helps, I can tell this: my pop3 server is pop.hotpop.com,
and their smtp server is smtp.hotpop.com, but I have to use
smtp.email.msn.com because my ISP blocks other smtp servers to prevent spam.
Also, I recently fixed up Kmail to use the hotpop pop3 server and my ISP's
SMTP server, and it works fine. Kmail connects directly to both get and send
email.
TIA,
jbrown
<reply text removed due to size>
--
Posted from cpimssmtpu12.email.msn.com [207.46.181.87]
via Mailgate.ORG Server - http://www.Mailgate.ORG
------------------------------
From: George Trapkov <[EMAIL PROTECTED]>
Subject: Upgrade RPM3 to RPM4
Date: Sat, 21 Apr 2001 16:40:42 -0400
I have Mandrake 7.0 and I want to install RPm4. However the files for
RPm 4 were build on 4. Is there a work around that?
George
------------------------------
From: Hui Ji <[EMAIL PROTECTED]>
Subject: help on the boot option
Date: 21 Apr 2001 20:42:51 GMT
I just installed Redhat 7.1, so far it's good, but there is one thing
I don't know how to setup. I used a laptop, sometimes without connected
to LAN, sometime I do. But the system just hang on if I didn't connect
to LAN, Network daemon keep trying.
Is there any way to setup an option when boot, so I could choose whether
start network daemon or not?
Thanks for help in advance
------------------------------
From: [EMAIL PROTECTED] (Juergen Heinzl)
Subject: Re: top level dir permissions
Date: Sat, 21 Apr 2001 21:01:23 GMT
In article <[EMAIL PROTECTED]>, Jeffrey J. Bacon wrote:
>what folders on the top level is it safe to remove Read permission for
>all users (except root, or group root)?
[-]
I'd say short of none, as various applications may fail in various
ways. Try and see what stops working.
>ie. I removed Read permission on /home so that users can not list home
>directories, but left eXecute on so they could get to their own home
>directory
>
>what about /dev? /proc? /var? others?
[-]
See before, opendir() and readdir() are bound to fail then and how
can you know ?
BTW this is going to be a good way to make your system short of
unusual as not even find is going to work anymore. I'd rather not
be a user on your system 8-/
Ta',
Juergen
--
\ Real name : Juergen Heinzl \ no flames /
\ EMail Private : [EMAIL PROTECTED] \ send money instead /
------------------------------
Date: Sat, 21 Apr 2001 23:12:32 +0200
From: Michael Heiming <[EMAIL PROTECTED]>
Subject: Re: top level dir permissions
"Jeffrey J. Bacon" wrote:
>
> what folders on the top level is it safe to remove Read permission for
> all users (except root, or group root)?
>
> ie. I removed Read permission on /home so that users can not list home
> directories, but left eXecute on so they could get to their own home
> directory
>
> what about /dev? /proc? /var? others?
Why not set the users shell to /sbin/false or even better, get rid of
all those users...:-)
awk '{FS=":"} ! (/root/ || ! /home/) {system ("userdel -r " $1)}'
/etc/passwd
(would speed this up)
Honestly, just keep the default perms, you chances are big, if you use
chmod/chown/chgrp with the -R option and some strange regexp, you may
render your whole system unusable with only one keystroke.
Check the various howto's on how to harden your distro, but I don't
really understand what you want to achieve, if those users shouldn't do
anything on the shell, that may, or may not be harmful to the system,
you could setup some chroot environment for those users.
Michael Heiming
------------------------------
From: Jack Pan <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: MySQL Problem.
Date: Sat, 21 Apr 2001 21:18:10 GMT
Joe wrote:
> In article <[EMAIL PROTECTED]>, [EMAIL PROTECTED]
> says...
> > Hi
> > This is my problem with mysql. when I run mysql as root. everything
> > is so fine. but when I try to run mysql as an user. say joe. I already
> > add user account joe in the mysql database. it gives me a error message.
> >
> > Warning: MySQL Connection Failed: Can't connect to local MySQL
> > server through socket '/var/lib/mysql/mysql.sock' (111).
> >
> > Anyone has any idea of how to solve this problem? Thank U
> >
> > Jack
> >
> >
>
> See if this solves your problem:
> http://www.mysql.com/doc/C/h/Changing_MySQL_user.html
I am sorry, I didn't put my problem correctlly. when I say "run mysql" I mean
"running mysql client".
The mysql server has already started at the boot time. when I log in as root.
I can execute the mysql client program. But, When I log in as an user. I
can't run mysql client program. and I got the messages shown in the previous
messages. I also have an user account defined in the mysql database.
Sorry to bother you again. I really need your help. Thanks
Jack
------------------------------
From: [EMAIL PROTECTED] (Dances With Crows)
Subject: Re: error when mounting FAT32 partition: kernel: fatfs: bogus cluster size
Reply-To: [EMAIL PROTECTED]
Date: 21 Apr 2001 21:40:04 GMT
On Sat, 21 Apr 2001 20:11:31 +0200, Carsten Cimander staggered into the
Black Sun and said:
>I installed SuSE Linux 7.1 (kernel 2.4.0) on my box.
>Using the preconfigured Icon "Windows_C" on my desktop I manage to mount
>/dev/hda2 as vfat successfully:
>/dev/hda2 on /windows/C type vfat (rw,noexec,nosuid,nodev,user=me)
>
>When trying the same with /dev/hda7 supposed to be /windows/D i failed
>getting the error:
>kernel: fatfs: bogus cluster size
>kernel: VFS: Can't find a valid MSDOS filesystem on dev 03:07
Hmmm. Make sure you are mounting hda7 as type vfat and not type msdos,
as type msdos may not handle FAT32 correctly. I presume Doze doesn't
report any problems with the partition. Run ScanDisk on the filesystem
anyway (skip the "surface scan") or use dosfsck if you're brave.
Also, see if you can boot kernel 2.2.x and try to mount hda7. 2.4.0 has
had a few problems (VIA chipsets, loopback problems) which is why 2.4.3
is out now....
--
Matt G|There is no Darkness in Eternity/But only Light too dim for us to see
Brainbench MVP for Linux Admin / Workin' in a code mine, hittin' Ctrl-Alt
http://www.brainbench.com / Workin' in a code mine, whoops!
=============================/ I hit a seg fault....
------------------------------
From: Shea A Martin <[EMAIL PROTECTED]>
Crossposted-To: linux.debian.user
Subject: reading multisession cd's
Date: Sat, 21 Apr 2001 21:41:11 GMT
I am trying to copy a multi-session cd. This was an image I downloaded,
and then used cdrecord to make. Now I would like to make a copy of the
disc.
I have tried the command 'readcd -V dev=0,2,0 f=cannon.iso'
I get this error:
<quote>
Executing 'read_g1' command on Bus 0 Target 2, Lun 0 timeout 20s
CDB: 28 00 00 00 00 00 00 00 01 00
readcd: Input/output error. read_g1: scsi sendcmd: retryable error
CDB: 28 00 00 00 00 00 00 00 01 00
status: 0x2 (CHECK CONDITION)
Sense Bytes: 70 00 05 00 00 00 00 0A 00 00 00 00 64 00 00 00
Sense Key: 0x5 Illegal Request, Segment 0
Sense Code: 0x64 Qual 0x00 (illegal mode for this track) Fru 0x0
Sense flags: Blk 0 (not valid)
cmd finished after 0.003s timeout 20s
readcd: Input/output error. Error not corrected.
</quote>
the command 'cdrecord dev=0,2,0 -msinfo' gives me
"207799,327022". I can mount & browse the cd, and I can use readcd on
non-multi's.
Does anyone know how to make a copy of a multi-session cd???
ps - I have also tried cdrdao, but it only reads the 1st Session.
--
Shea Martin
_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-
Elbows out...
...stick on the ice! \_.
------------------------------
From: Bruno Barberi Gnecco <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.x
Subject: Voodoo (usual) problems
Date: Sat, 21 Apr 2001 17:45:55 -0300
Reply-To: [EMAIL PROTECTED]
I have a Voodoo 3 PCI, which used to work with acceleration
in X4.0.1/kernel 2.2.17. I decided to upgrade to X4.0.3, expecting an
improved driver. I got the tdfx drivers in dri.sourceforge.net, and
they asked for kernel 2.4; so I updated to 2.4.3 (which doesn't find my
parallel port scanner, but that's another story;).
In short, I can run X (using the drivers that came with 4.0.3),
without acceleration.
If I try the dri.sourceforge drivers, however, I get the following
error:
(II) LoadModule: "tdfx"
(II) Loading /usr/X11R6/lib/modules/drivers/tdfx_drv.o
(II) Module tdfx: vendor="The XFree86 Project"
compiled for 2.49.119Ez, module version = 1.0.0
Module class: XFree86 Video Driver
ABI class: XFree86 Video Driver, version 0.4
(EE) module ABI minor version (4) is newer than the server's version (3)
(II) UnloadModule: "tdfx"
(II) Unloading /usr/X11R6/lib/modules/drivers/tdfx_drv.o
(EE) Failed to load module "tdfx" (module requirement mismatch, 0)
(...)
(EE) No drivers available.
Fatal server error:
no screens found
And that's it. Does anybody know how to solve this?
Best regards,
--
Bruno Barberi Gnecco <[EMAIL PROTECTED]>
http://www.geocities.com/RodeoDrive/1980/
Quoth the Raven, "Nevermore". - Poe
------------------------------
From: Jean-David Beyer <[EMAIL PROTECTED]>
Subject: Re: Reliability of "time" command?
Date: Sat, 21 Apr 2001 17:53:27 -0400
"Peter T. Breuer" wrote:
>
> Francis Litterio <[EMAIL PROTECTED]> wrote:
> > MH <[EMAIL PROTECTED]> writes:
>
> >> As a comparison, I ran "time" on the "du" command for each system, thus:
> >>
> >> time du -s /
> >>
> >> For the larger data source (~7 GB) "du" reported 8.95s
> >> For the smaller data source (~2.5 GB) "du" reported 4.63s
>
> > I'm not sure these numbers bear any relation to the disk cloning anomaly.
>
> Indeed .. it's just reading the filesystem metadata. However, I can
> probably give him some comparisons.
>
> P450, intel 810 mobo, 128MB ram, symbios 10MB/s ultra scsi controller
> holding 4 1996 vintage quantum fireball 1.28GB disks (which do about
> 7.5MB/s), ext2fs filesystem, 2.2.15 kernel:
>
> first time (1.37GB in 2 mins 53s wall time, 3.87s cpu time)
>
> oboe:/usr/oboe/ptb% time du -sx .
> 1372582 .
> 0.390u 3.480s 2:53.33 2.2% 0+0k 0+0io 95pf+0w
>
> second time (1.37GB in 2.94s wall time, 2.82s cpu time)
> oboe:/usr/oboe/ptb% time du -sx .
> 1372599 .
> 0.270u 2.550s 0:02.94 95.9% 0+0k 0+0io 94pf+0w
>
> so you can see that caches play a large part in the experiment. I don't
> feel that this is a valid measurement technique, therefore. A
> hundred-fold difference in results shows that the experimental protocol
> is not at all well-defined.
>
> I'll run a more sensible test. Writing 200MB of zeros after mounting
> the filesystem sync, using 1MB blocks (this will be quite slow, as
> the fs is not quiet, and since it's mounted sync now, all kinds of
> things will interfere with the write, which isn't atomic):
>
> oboe:/tmp% time dd if=/dev/zero of=writetest bs=1024k count=200
> 200+0 records in
> 200+0 records out
> 0.010u 9.310s 8:03.13 1.9% 0+0k 0+0io 96pf+0w
>
> yes, well, that just shows you why one doesn't normally mount local
> fs's (especially tmp!) sync. 400KB/s !!
>
> I'll do it again, this time with the fs mounted async, as normal. If
> I were clever, I would also set it noatime.
>
> oboe:/tmp% time dd if=/dev/zero of=writetest bs=1024k count=200
> 200+0 records in
> 200+0 records out
> 0.000u 2.540s 0:39.07 6.5% 0+0k 0+0io 121pf+0w
>
> A rather more impressive 5MB/s wallclock. One should note that neither
> test occupied the cpu in the least, which was why I was able to
> continue watching tv and writing this message without noticing it,
> while also receiving log files from every machine in 100M to the same
> partition. That's the advantage of scsi - not linear speed. But if
> necessary I'll perform the same test on my ultra 160 controllers and
> lvd disks, for more interesting data.
Here it is on my machine with an ultra-2 SCSI controller on 10,000rpm
hard drives.
valinux:jdbeyer[~]$ time dd if=/dev/zero of=writetest bs=1024k
count=200
200+0 records in
200+0 records out
real 0m4.841s
user 0m0.000s
sys 0m1.680s
valinux:jdbeyer[~]$
It seemed to be doing 5.3 megabytes/second according to xosview.
This is what vmstat thinks of it:
valinux:jdbeyer[~]$ vmstat 1
procs memory swap io
system cpu
r b w swpd free buff cache si so bi bo in cs
us sy id
3 0 0 12692 3508 249648 102664 0 0 16 27 142 201
89 6 4
2 0 0 12692 4536 249648 102664 0 0 0 0 170 334
94 6 0
2 0 0 12692 3508 249648 102664 0 0 0 16 198 2091
88 12 0
2 0 0 12692 3640 249648 102664 0 0 0 0 109 300
92 8 0
2 0 0 12692 3640 249648 102664 0 0 0 0 112 261
91 9 0
2 0 0 12692 4152 249648 102664 0 0 0 0 116 265
93 7 0
2 0 0 12692 4668 249648 102664 0 0 0 0 105 245
93 7 0
2 0 0 12692 4668 249648 102664 0 0 0 8 119 254
91 9 0
2 0 0 12692 3640 249648 102664 0 0 0 0 112 300
92 8 0
2 0 0 12692 3640 249648 102664 0 0 0 0 115 275
94 6 0
6 0 0 12692 3560 249648 102664 0 0 0 1500 163 229
34 67 16909319
2 1 1 12692 4316 249648 102664 0 0 0 6000 468 904
71 29 0
3 1 1 12692 3552 249648 102664 0 0 0 6198 501 810
69 30 0
2 1 1 12692 3800 249648 102664 0 0 0 5648 469 804
67 33 0
2 1 1 12692 3420 249648 102664 0 0 0 5627 462 798
73 27 0
2 0 0 12692 4664 249648 102664 0 0 0 527 234 631
91 8 1
2 0 1 12692 3504 249648 102664 0 0 0 4084 277 576
90 10 0
2 0 1 12692 3504 249648 102664 0 0 0 5552 476 920
86 14 0
2 0 1 12692 4532 249648 102664 0 0 0 5280 453 923
84 15 1
2 0 1 12692 3504 249648 102664 0 0 0 5872 503 1061
77 23 0
2 0 0 12692 4016 249648 102664 0 0 0 4988 495 911
87 13 0
2 0 0 12692 3504 249648 102664 0 0 0 0 194 335
92 8 0
4 0 0 12692 4664 249648 102664 0 0 0 0 132 313
92 8 0
2 0 0 12692 3376 249648 102664 0 0 0 0 139 319
92 8 0
2 0 0 12692 3376 249648 102664 0 0 0 0 139 333
93 7 0
valinux:jdbeyer[~]$
>
> Does (substitute bozo A) now understand why his "data" is not data?
>
> > The du command doesn't read all the data from the disk (as happens when
> > you clone a disk). Plus, these numbers are biased by the filesystem
> > buffer cache. In fact, the buffer cache will bias any I/O speed
> > measurement (except maybe for a full read of the entire disk using its
> > block device, because there should be no cache hits in that case).
>
> Peter
--
.~. Jean-David Beyer Registered Linux User 85642.
/V\ Registered Machine 73926.
/( )\ Shrewsbury, New Jersey http://counter.li.org
^^-^^ 5:45pm up 1 day, 11 min, 3 users, load average: 2.12, 2.16, 2.19
------------------------------
** 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
******************************
