Linux-Development-Sys Digest #605, Volume #8 Tue, 3 Apr 01 04:13:17 EDT
Contents:
Re: Kernel 2.4.3 reverses assignment order of AHA2940UW and NCR810 disks (Bernd
Strieder)
Re: blob support by linux-db (Juergen Sauer)
Re: 8200 Tape drive problems (Jerry Peters)
Re: mysql or postgresql? which is better?
Re: mysql or postgresql? which is better?
did anyone have problems with Daylight Savings changeover? (jtnews)
Re: smp question (Rick Ace)
DEV_TINT question. ("Interested Reader")
Re: blob in linux-database (Bob Hauck)
Re: Kernel 2.4.3 reverses assignment order of AHA2940UW and NCR810 disks (Harmon
Seaver)
Re: What is the size of Linux 2.4.1 Kernel (Toby A Inkster Esq)
Re: mysql or postgresql? which is better?
RFC 868 inconsistency? (Luben Tuikov)
The roadblocks to printing are infinitely numbered ("Gene Heskett")
signed vs unsigned ints discussion please ("Gene Heskett")
Re: two question about linux kernel source? (Villy Kruse)
which raidtools for kernel 2.4.3 ??? (Frank Rust)
LILO & bootsect.S
Re: Quiet Boot for Linux (Daniel Franklin)
Re: chinese browser! (Rolf Magnus)
----------------------------------------------------------------------------
From: Bernd Strieder <[EMAIL PROTECTED]>
Subject: Re: Kernel 2.4.3 reverses assignment order of AHA2940UW and NCR810 disks
Date: Mon, 02 Apr 2001 23:48:06 +0200
"Hal R. Brand" wrote:
>
> Executive Summary: Kernel 2.4.3 assigns disks on the NCR810 controller
> before the Adaptec 2940UW controller. This is the opposite
> behavior of previous kernels and can lead to all kinds of "fun".
>
> Just tried kernel 2.4.3 and ran into the following "difference" in
> 2.4.3 vs its predecessors. I have both an Adaptec 2940UW and
> NCR810 SCSI controllers. Up till 2.4.3, the 2940UW was initialized
> (and the disks assigned) before the NCR810. With 2.4.3, the NCR810
> is initialized first. In my case, with one disk on each controller,
> this reversed disks sda and sdb.
[...]
> CONFIG_SCSI_AIC7XXX=y
> CONFIG_SCSI_NCR53C8XX=y
Why not leaving only the driver needed to get access to the root
partition in the kernel. The other one could be a module.
If you were using initrd, it'd be possible to have them both as modules,
specifying somewhere appropriate in the initrd startup the correct order
of loading the drivers.
There is a kernel parameter scsihosts, which AFAIK has something to do
with the order of initialization of the drivers builtin into the kernel,
but I haven't tried.
Bernd Strieder
------------------------------
From: Juergen Sauer <[EMAIL PROTECTED]>
Subject: Re: blob support by linux-db
Date: 2 Apr 2001 19:45:47 GMT
Hans-Peter Maurer <[EMAIL PROTECTED]> schrieb
am Mon, 2 Apr 2001 02:32:24 +0200 in comp.os.linux.development.system:
> hello world!
> i need a linux-database with blob-support...
> do you know one...!
Adabas D.
mfG
Jojo
--
J�rgen Sauer - AutomatiX GmbH, +49-4209-4699, [EMAIL PROTECTED]
http://www.automatix.de to Mail me: remove: -not-for-spawm-
------------------------------
From: Jerry Peters <[EMAIL PROTECTED]>
Subject: Re: 8200 Tape drive problems
Date: Mon, 02 Apr 2001 22:06:16 GMT
Manoj Patil <[EMAIL PROTECTED]> wrote:
> Jerry Peters wrote:
>> Manoj Patil <[EMAIL PROTECTED]> wrote:
>> > Hi !
>> > I have IBM PC with Red Hat 6.2 (kernel 2.2.14)
>> > I have connected two SCSI tape drives to this machine
>> > I am not able to change the density and block size for one of the tape
>> > drive
>>
>> > $ cat /proc/scsi/scsi shows me following
>> > Attached devices
>> > Host: scsi0 Channel: 00 ID: 02 Lun:00
>> > Vendor: EXABYTE Model: IBM-8505 Rev: 7J0A
>> > Type: Sequential-Access ANSI SCSI
>> > revision: 02
>> > Host: scsi0 Channel: 00 ID: 03 LUN:00
>> > Vendor: EXABYTE Model: EXB-8200 Rev:2680
>> > Type: Sequential-Access ANSI SCSI
>> > revision: 01 CCS
>>
>> > $mt -f /dev/nst1 status
>> > SCSI 1 tape drive
>> > File number =-1, block number=-1
>> > Tape block size 0 bytes. Density code 0x0 (default)
>> > Soft error count since last status=0
>> > General status bits on (1010000):
>> > ONLINE IM_REP_EN
>>
>> > On this tape if i try to change the blksize or the density, I get
>> > following error
>> > ( mt -f /dev/nst1 setblk 1024 and setdensity 0x14)
>>
>> > st1: Error with sense data: [valid=0] Infofld=0x0 , EOM current
>> > st09:01:sense key Illiegal request
>> > /dev/nst1: Input/Output Error
>>
>> > However, I am able to do a tar cvf and tar tvf on this tape using a
>> > 112M cartridge.
>>
>> > Can some one please help me in how to go about changing the blk size .
>>
>> > (I am able to change block size and density on the other tape i.e
>> > /dev/nst0)
>>
>> It looks like you don't have a tape in the drive. My 4mm dat rejects
>> certain commands, including setblksize when there's no tape present.
>> This appears to be dependent on the drive itself, I've used other 4mm
>> drives that will accept setblksize even without a tape.
>>
>> Jerry
> Jerry,
> Thanks for your note. I have confirmed that the drive has a tape in it. But
> it does not allow me to change the block size and the density. But I am
> able to do it for the other tape. I think this must be something to do with
> the scsi device driver subsytem
The density may not be changeable, there may be only 1 valid value.
Have you tried to change the default blocksize? Even with no tape in
it mt defblk 0 is accepted by my drive.
When you boot the system or insert the st module, what messages do you
get in syslog? Here's what I get when I add the device:
(I've trimmed the lines somewhat to fit)
kernel: scsi singledevice 1 0 0 0
kernel: Vendor: ARCHIVE Model: Python 25501-XXX Rev: 5.72
kernel: Type: Sequential-Access ANSI SCSI revision: 02
kernel: Detected scsi tape st0 at scsi1, channel 0, id 0, lun 0
kernel: st: bufsize 32768, wrt 30720, max init. buffers 4, s/g segs 16.
kernel: sym53c810a-1-<0,*>: FAST-10 SCSI 6.7 MB/s (150.0 ns, offset 8)
kernel: st0: Block limits 1 - 16777215 bytes.
kernel: st0: Normal timeout set to 300 seconds.
kernel: st0: Default block size set to 0 bytes.
kernel: st0: Mode 0 options: buffer writes: 1, async writes: 1, read ahead: 1
kernel: st0: can bsr: 0, two FMs: 0, fast mteom: 0, auto lock: 0,
kernel: st0: defs for wr: 0, no block limits: 0, partitions: 1, s2 log: 0
kernel: st0: sysv: 0
kernel: st0: Mode 0 options: buffer writes: 1, async writes: 1, read ahead: 1
kernel: st0: can bsr: 1, two FMs: 0, fast mteom: 0, auto lock: 0,
kernel: st0: defs for wr: 0, no block limits: 0, partitions: 1, s2 log: 0
kernel: st0: sysv: 0
darkstar Diskdump: Dump started to tape: 14 Level: 0
IIRC the Block limits 1 - 16777215 bytes has something to do with the
valid blocksizes.
Jerry
------------------------------
From: [EMAIL PROTECTED] ()
Subject: Re: mysql or postgresql? which is better?
Date: Mon, 02 Apr 2001 22:27:58 -0000
In article <9aakvc$jbh$04$[EMAIL PROTECTED]>,
Hans-Peter Maurer <[EMAIL PROTECTED]> wrote:
>which is better mysql or postgresql? Which is more reliable and more
>powerful...?
It depends on what you want to do. If all you want is good read
performance, mysql is better. But mysql doesn't do transactions
which is a major problem if you ever do concurrent updates.
--
http://www.spinics.net/linux/
------------------------------
From: <[EMAIL PROTECTED]>
Subject: Re: mysql or postgresql? which is better?
Date: Mon, 02 Apr 2001 22:58:28 GMT
[EMAIL PROTECTED] wrote:
> It depends on what you want to do. If all you want is good read
> performance, mysql is better. But mysql doesn't do transactions
> which is a major problem if you ever do concurrent updates.
Even that's up in the air. postgresql outperforms mysql by a handy
margin on many workloads. However, the design goals of both are quite
different, so direct comparisons are probably unfair.
As a general rule, postgreql is faster and more robust, while using
more resources. mysql is alot slower under pathological workloads,
while using less resources normally. It's definately worth
benchmarking both for critical applications.
--
Matt Gauthier <[EMAIL PROTECTED]>
------------------------------
From: jtnews <[EMAIL PROTECTED]>
Subject: did anyone have problems with Daylight Savings changeover?
Date: Mon, 02 Apr 2001 23:48:55 GMT
I was at my system while the changeover from
standard time to daylight savings time
occurred and everything appeared ok.
But when I rebooted my machine after the
changeover time, my system clock was set
wrong on boot.
When I set my system clock again, everything
was ok, even after another reboot.
What's going on?
------------------------------
From: Rick Ace <[EMAIL PROTECTED]>
Subject: Re: smp question
Date: Mon, 02 Apr 2001 19:55:09 -0400
[EMAIL PROTECTED] wrote:
>
> Tim Roberts <[EMAIL PROTECTED]> wrote:
> > ([EMAIL PROTECTED]) wrote:
>
> >>For smp's the value smp_num_cpus is used alot.
> >>I wonder what happens to that value when one of the cpus crashes -
> >>assuming linux doesn't crash because of this event.
>
> > Exactly how, in your experience, would a CPU "crash"?
>
> > Processes crash. Hard disks crash. CPUs do not crash.
>
> For example the fan on the CPU ceases to spin, and the CPU gets hot and stops
> working. This is very common, much more frequent in my experience than hard
> disk crash.
I'd guess that the kernel is not prepared to handle failures of
this nature. If that's correct, a hung system is a likely result,
making the question about the value in smp_num_cpus moot.
------------------------------
From: "Interested Reader" <[EMAIL PROTECTED]>
Subject: DEV_TINT question.
Date: Mon, 2 Apr 2001 20:37:52 -0500
I was reading a device driver book and found that the LAN example will not
compile on my 2.2 kernel box because of dev_tint().
I recall reading somewhere that dev_tint() died some where around 2.1.18 or
something like that. What I have not found is what replaced it and how to
adjust for it.
Can some one give me a URL to help me out?
Please reply to the news group and thanks in advance.
------------------------------
From: [EMAIL PROTECTED] (Bob Hauck)
Subject: Re: blob in linux-database
Reply-To: bobh = haucks dot org
Date: Tue, 03 Apr 2001 01:43:41 GMT
On Mon, 2 Apr 2001 02:38:29 +0200, Hans-Peter Maurer
<[EMAIL PROTECTED]> wrote:
>what kind of database supports blobs in linux!!
Probably most of them. I know PostgreSQL does.
--
-| Bob Hauck
-| To Whom You Are Speaking
-| http://www.haucks.org/
------------------------------
From: Harmon Seaver <[EMAIL PROTECTED]>
Subject: Re: Kernel 2.4.3 reverses assignment order of AHA2940UW and NCR810 disks
Date: Tue, 03 Apr 2001 01:49:00 GMT
I found sort of the same thing in 2.4.1 and 2.4.2 in that they switched my
two nics, so that my 3com was now eth0 and my pc32net was eth1 -- really
confused me at first until I figured out what happened to my net/nat etc. I
think what is going on is that it reads the pci bus from the other end.
Bernd Strieder <[EMAIL PROTECTED]> wrote:
> "Hal R. Brand" wrote:
>>
>> Executive Summary: Kernel 2.4.3 assigns disks on the NCR810 controller
>> before the Adaptec 2940UW controller. This is the opposite
>> behavior of previous kernels and can lead to all kinds of "fun".
>>
>> Just tried kernel 2.4.3 and ran into the following "difference" in
>> 2.4.3 vs its predecessors. I have both an Adaptec 2940UW and
>> NCR810 SCSI controllers. Up till 2.4.3, the 2940UW was initialized
>> (and the disks assigned) before the NCR810. With 2.4.3, the NCR810
>> is initialized first. In my case, with one disk on each controller,
>> this reversed disks sda and sdb.
> [...]
>> CONFIG_SCSI_AIC7XXX=y
>> CONFIG_SCSI_NCR53C8XX=y
> Why not leaving only the driver needed to get access to the root
> partition in the kernel. The other one could be a module.
> If you were using initrd, it'd be possible to have them both as modules,
> specifying somewhere appropriate in the initrd startup the correct order
> of loading the drivers.
> There is a kernel parameter scsihosts, which AFAIK has something to do
> with the order of initialization of the drivers builtin into the kernel,
> but I haven't tried.
> Bernd Strieder
--
Harmon Seaver, MLIS
CyberShamanix
Work 920-203-9633 [EMAIL PROTECTED]
Home 920-233-5820 [EMAIL PROTECTED]
------------------------------
From: Toby A Inkster Esq <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux.advocacy,comp.os.linux.development.apps,comp.os.linux.help,comp.os.linux.misc
Subject: Re: What is the size of Linux 2.4.1 Kernel
Date: Tue, 03 Apr 2001 03:59:26 +0100
In our last episode, Alumne FIB - MARC COLL CARRILLO wrote:
:The compressed file is about 20 Mb. Once uncompressed, it can grow up to
:100 Mb.
ROFL, Free Agent hadn't downloaded the body of this message when I
first read it, do all I saw was this:
|Re: What is the size of Linux 2.4.1 Kernel
|
| [13 lines]
|
| [Press M to mark this message for later retrieval.]
A 13 line Kernel... he he.
------------------------------
From: [EMAIL PROTECTED] ()
Subject: Re: mysql or postgresql? which is better?
Date: Tue, 03 Apr 2001 03:50:59 -0000
In article <ou7y6.27987$[EMAIL PROTECTED]>,
<[EMAIL PROTECTED]> wrote:
>As a general rule, postgreql is faster and more robust, while using
>more resources. mysql is alot slower under pathological workloads,
>while using less resources normally. It's definately worth
>benchmarking both for critical applications.
Ok, I can believe that.
--
http://www.spinics.net/linux/
------------------------------
From: Luben Tuikov <[EMAIL PROTECTED]>
Subject: RFC 868 inconsistency?
Date: Tue, 03 Apr 2001 03:54:25 GMT
Hello,
RFC 868 describes the time protocol TCP/UDP port 37 service,
which it says returns the number of seconds since 0:0 Jan 1, 1900 GMT
as a signed 32 bit binary number.
My questions are: is the format big-endian or little-endian?
Also the example in the RFC says that 2,208,988,800 corresponds
to 0:0 Jan 1, 1970 GMT, but how is this possible as this number is
greater than a 32 bit LONG_MAX (which is 2,147,483,647)?
Either the RFC means 1970 rather than 1900 or the number returned is NOT
a 32 bit signed number?
What is the story?
--
Luben
------------------------------
Date: 24 Mar 2001 9:37:28 -0500
From: "Gene Heskett" <[EMAIL PROTECTED]>
Subject: The roadblocks to printing are infinitely numbered
Ok, heres one for your contemplation...
Trying to build a gs I can use with the newest printing setup from
rawhide.
1. The gcc-2.96-78 compiler dies on zlib includes, kgcc works past that.
2. Finally getting the XLIBDIRS path right, then the make of gs6.61 dies
with this, using kgcc:
/bin/sh <./obj/ldt.tr
./obj/zutil.o: In function `zcalloc':
./obj/zutil.o(.text+0x2a): undefined reference to `ssh_xcalloc'
./obj/zutil.o: In function `zcfree':
./obj/zutil.o(.text+0x37): undefined reference to `ssh_xfree'
/usr/X11R6/lib/libXt.so: undefined reference to `[EMAIL PROTECTED]'
/usr/X11R6/lib/libX11.so: undefined reference to `[EMAIL PROTECTED]'
/usr/X11R6/lib/libXt.so: undefined reference to `[EMAIL PROTECTED]'
collect2: ld returned 1 exit status
make: *** [bin/gs] Error 1
Next I can check the GLIBC versions:
[root@gene gs6.61]# rpm -qa|grep glibc
glibc-profile-2.2.2-7
glibc-common-2.2.2-7
glibc-devel-2.2.2-7
compat-glibc-6.2-2.1.3.2
glibc-2.2.2-7
In other words, the latest (as of yesterday anyway) glibc install.
Ideas anyone? What dependencies are broken? I'm about burnt, been
fighting with first one thing then another in the path to a working
printer setup for about 2 weeks now.
Cheers, Gene
--
Gene Heskett, CET, UHK |Amiga A2k Zeus040, Linux @ 500mhz
email gene underscore heskett at iolinc dot net
#Amiga based X10 home automation program EZHome, see at:#
<http://www.thirdwave.net/~jimlucia/amigahomeauto>
This messages reply content, but not any previously quoted material,
is � 2001 by Gene Heskett, all rights reserved.
--
------------------------------
Date: 26 Mar 2001 0:40:51 -0500
From: "Gene Heskett" <[EMAIL PROTECTED]>
Subject: signed vs unsigned ints discussion please
Hi all;
I'm trying to clean up some c++ code but don't know a lot about c++. One
of the problems in this code is that there are many places in it where
the format is:
if var >= 0
where the var is an int and isn't spec'd as an unsigned, and in fact
can't be or the compiler bails out totally since such a comparison is
always true. I actually tried that :(
So the real problem is that the casting of the numeral '0' must be an
unsigned value according to the warnings given.
The question then is: How can I specify, in c++, a globally available
signed int containing the value of zero, or 0x80000000? Then all I have
to do is change all the '0's to 'zero'.
Yeah, I know, its a stupid question but this has bit me in C many times
as its so darned easy to forget that little detail...
Cheers, Gene
--
Gene Heskett, CET, UHK |Amiga A2k Zeus040, Linux @ 500mhz
email gene underscore heskett at iolinc dot net
#Amiga based X10 home automation program EZHome, see at:#
<http://www.thirdwave.net/~jimlucia/amigahomeauto>
This messages reply content, but not any previously quoted material,
is � 2001 by Gene Heskett, all rights reserved.
--
------------------------------
From: [EMAIL PROTECTED] (Villy Kruse)
Subject: Re: two question about linux kernel source?
Date: 3 Apr 2001 06:33:50 GMT
On Mon, 2 Apr 2001 23:06:08 +0800, hushui <[EMAIL PROTECTED]> wrote:
>1.int bootsect.S , the variable root_dev is standing for the root device
>???
>Is ROOT_DECV defined when compile a kernel ???
It is also set using the "rdev" command, and is used if you copy the
kernel to a raw floppy disk and boot from it. When booting from lilo
or loadlin the rootdev is patched into the loaded kernel from the root
parameter given to lilo or loadlin.
Villy
------------------------------
From: [EMAIL PROTECTED] (Frank Rust)
Subject: which raidtools for kernel 2.4.3 ???
Date: 3 Apr 2001 07:17:31 GMT
Reply-To: [EMAIL PROTECTED]
Hi all,
Im looking for the correct version of raidtools for kernel 2.4.3
All versions I tried till now don't compile:
raidtools-0.41
raidtools-pre3-0.42
raidtools-0.50beta3
In all of these versions I get
*** common.h: linux/md.h: No such file
if I change this include to linux/raid/md.h
I get lots of error and warning messages ending in
/usr/include/linux/raid/md_k.h: #error MD doesnt handle bigger kdev yet
?!?
Can anybody help?
Regards Frank.
--
Frank Rust Technische Universitaet Braunschweig
Tel.: +49 531 391 9525 Institut fuer Theoretische Informatik
Fax.: +49 531 391 9529 Fallersleber-Tor-Wall 22
email: [EMAIL PROTECTED] 38100 Braunschweig
------------------------------
From: <[EMAIL PROTECTED]>
Subject: LILO & bootsect.S
Date: Tue, 03 Apr 2001 07:30:06 -0000
Hi,
I'm a little bit confused with LILO and bootsect.S and their
relationship. My understanding is that bootsect.S is used to boot linux
from a floppy. This is probably the reason why there is a code in the
bootsect.S that initialize the FDC and then read the 1st 2 sectors from it.
It is also my understanding that the bootsect.S is part of the 1st 512
bytes of the kernel image.
My qustions now are:
1)If LILO is to boot the kernel image, will bootsect.S be loaded and
executed ? If so then what happen to the part where it tries to read the
floppy drive?
2)If bootsect.S is not loaded then how can LILO find setup.S ?
Tks.
AL
(a linux newbie)
--
Posted via CNET Help.com
http://www.help.com/
------------------------------
From: [EMAIL PROTECTED] (Daniel Franklin)
Subject: Re: Quiet Boot for Linux
Reply-To: [EMAIL PROTECTED]
Date: 3 Apr 2001 11:47:02 +1000
Kasper Dupont wrote:
>Benjamin Scherrey wrote:
>>
>> I've got a techno-phobic client who gets concerned when he sees all the
>> things going across his screen when linux boots up. This is enough of an
>> issue that I'm prepared to go in and build a customer kernel or whatever
>> is necessary to make a linux boot appear even simpler than windows. How
>> do I get control over what gets to the screen when linux is booting? Is
>> there some parameter I can pass while building the kernel to suppress its
>> screen output? Perhaps route it to /dev/null or something?
>>
>> thanx & later,
>>
>> Ben Scherrey
>
>All you have to do is give "console=null" as argument to
>the kernel. The kernel argument can be specified in your
>loader configuration. If you use lilo the configuration
>is in /etc/lilo.conf. Find the default option and make a
>copy, then edit the append option. When the new config
>is saved run /sbin/lilo.
>
There's also a Linux Progress Patch which displays pretty pictures and
flashing lights as the machine boots, but really, this guy sounds like a
total retard, even if he is paying you money you should only tolerate his
cluelessness so far. You should just hit him in the head with your
combination Clue Stick / LART (LARTs are very useful in my experiencing for
the Readjustment of Luser Attitude) and let him figure it out himself. Or
advise him to turn his monitor on a minute after switching on his computer.
Maybe tell him that if there is a problem and you can't see the error
messages you'll have to charge him double for the extra time required to
work out the problem.
- Daniel
--
******************************************************************************
* Daniel Franklin - Postgraduate student in Electrical Engineering
* University of Wollongong, NSW, Australia * [EMAIL PROTECTED]
******************************************************************************
------------------------------
From: Rolf Magnus <[EMAIL PROTECTED]>
Subject: Re: chinese browser!
Date: Tue, 3 Apr 2001 10:03:51 +0200
Leo Naboro wrote:
> hi,i wanna look for a small browser under embedded environment,and it
> should support the chinese.Does anywhere i can find? thanks!
Konqueror supports chinese afaik, and there is an embedded version of it.
Don't know if this one also suppors chinese though.
------------------------------
** 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 the
comp.os.linux.development.system newsgroup.
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-Development-System Digest
******************************
