Linux-Setup Digest #512, Volume #19 Tue, 29 Aug 00 21:13:11 EDT
Contents:
Problem with RPM and loading of source (Ken Kachnowich)
Re: Problem with TomCat 3.1 ("Turk")
installing linux ("Michael Rosel")
Re: Annoying X problem - PLEASE HHHHEEEELLLLLPPPPP!!!!!!!! (Gregory)
Re: Linux RH6.2 RAID1 - Anyone succeed at this? (Professor Bruno)
Re: Linux, XML, and assalting Windows ([EMAIL PROTECTED])
Re: scsi bus resets with aic7xxx (vlado)
Re: Restaurant Booking System (vlado)
Re: Linux, XML, and assalting Windows ([EMAIL PROTECTED])
Re: wron major or minor version number for partitions. What am I doing wrong?
(Jonathan)
Ghostscript Make errors; jpeglib.h (Roger Jagoda)
Re: sendmail is acting like a snail (Colin Watson)
Re: Problem with RPM and loading of source (Colin Watson)
Re: mount fails on hda2 (Colin Watson)
----------------------------------------------------------------------------
From: Ken Kachnowich <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Problem with RPM and loading of source
Date: Tue, 29 Aug 2000 19:25:07 -0400
I must be doing something wrong.
I want to load the source for PostgreSQL from the Redhat 6.2 CD.
I put in disk 2 and run the Gnome RPM tool
I look in the SRPM directory and find the postgreSQL.src rpm file
I click install
And nothing. I tried rpm from the command line and still nothing.
What am I doing wrong here.
Thanks,
Ken
------------------------------
From: "Turk" <[EMAIL PROTECTED]>
Subject: Re: Problem with TomCat 3.1
Date: Wed, 30 Aug 2000 00:52:50 +0200
Krzysztof Binias <[EMAIL PROTECTED]> wrote in message
news:bFNq5.4461$[EMAIL PROTECTED]...
> I have RedHat 6.2 with TomCat 3.1.
> When I run TomCat "tomcat.sh run" everything is OK.
> When I try to connect to TomCat server (with telnet for example) I get
> message
> message and server is stopped:
> "java.net.SocketException: Interrupted system call"
> ...
> Endpoint ServerSocket[addr=0.0.0.0/0.0.0.0,port=0,localport=8007] shutdown
> due to exception:
> Endpoint ServerSocket[addr=0.0.0.0/0.0.0.0,port=0,localport=8080] shutdown
> due to exception:
>
> What 's the matter ?!
>
> Krzysiek.
>
>
I've the same problem !!!!!!
Could someone help us ??????
Davide
------------------------------
From: "Michael Rosel" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.hardware,comp.os.linux.networking
Subject: installing linux
Date: Wed, 30 Aug 2000 07:17:51 +0800
Hello everybody!!!
I am very much new to Linux and I want to have informations about it.
--What do I need
--limitations
--how to
Can anyone refer me to a very informative site.
I am planning to install linux on a Compaq deskpro 5120 (pentium 120Mhz)
series and on a Deskpro 4000 (pentium 166) series.
------------------------------
From: [EMAIL PROTECTED] (Gregory)
Subject: Re: Annoying X problem - PLEASE HHHHEEEELLLLLPPPPP!!!!!!!!
Date: Tue, 29 Aug 2000 23:38:17 GMT
In message <[EMAIL PROTECTED]> - Bozzy
<cbostock@=antispamremove=btinternet.com> writes:
:>
:> I have recently installed Mandrake 7.0 on my new PC:
:>
:> Athlon 600
:> ASUS K7M Motherboard
:> G400 Graphics card (32Mb)
:> Mitsubishi Pro 710 monitor
:>
:>The problem is that that periodically, possibly anywhere between 30
:>seconds to several minutes, my screen blinks. That is it goes blank
:>for about a second after which the desktop display returns. I have
:>tried upgrading to 7.1 of Mandrake, but that did not help. I am
:>running KDE, but the same problem also manifests itself when I switch
:>to Gnome. I did notice that when I try to configure my graphics card,
:>specifying 32M, that when I go back in to configure it again it seems
:>to reset it to a 16M G400, though I don't know whether this is an
:>issue.
:>
:>Since upgrading to 7.1 I have selected my monitor from the list
:>provided by the X-config utility and the Vertical and Horizontal sync
:>rates match between my monitor manual and the XF86Config file.
:>
:>This is a really anoying bug that's driving me up the wall so any help
:>would be appreciated.
:>
:>Thanks in advance,
:>
:>Clive.
I can't help with your problem but I would like to know how you got it
installed. I have a similar system to yours, only diff is the speed of the
cpu, mine is 750. I can't get X to configure....I get an error saying that it
can't find the file for the server SVGA.......any help?
------------------------------
From: [EMAIL PROTECTED] (Professor Bruno)
Subject: Re: Linux RH6.2 RAID1 - Anyone succeed at this?
Date: 29 Aug 2000 23:51:02 GMT
error@hell wrote:
: On Tue, 29 Aug 2000, Professor Bruno wrote:
: >[ Article crossposted from panix.user.unix,panix.questions ]
: >[ Author was Professor Bruno ([EMAIL PROTECTED]) ]
: >[ Posted on 25 Aug 2000 17:17:04 GMT ]
: >
: >Per subject line, I've been trying to setup RAID1 on a RH6.2
: >which has two Seagate 20.5 GB drives, but can't
: >get my root and boot partitions on the RAID1.
: >
: >I've combed through the HOWTOs and the linux-raid
: >archives trying to figure this out. Anyone
: >on this panix newsgroup been successful?
: >
: >Thanks!
: >
: >
: >Max Pyziur BRAMA - Gateway Ukraine
: >[EMAIL PROTECTED] http://www.brama.com/
: There is a how-to for exactly this at www.linuxdocs.org
: --
: moonie ;)
Like I said above, I've followed those HOWTOs to the letter. I've gotten
all of my partitions going on RAID1, except the two critical ones - /boot
and / (root).
Max
: Registered Linux User #175104
: http://counter.li.org
: KDE2
: Kernel 2.4.0-test5
: XFree86 4.0 Nvidia .94 drivers
: RAID 0 Striped
: Test-Pilots-R-Us ;)
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: alt.os.linux,comp.text.xml,comp.os.linux.misc
Subject: Re: Linux, XML, and assalting Windows
Date: Wed, 30 Aug 2000 00:21:12 GMT
In article <[EMAIL PROTECTED]>,
bobh{at}haucks{dot}org wrote:
> On Tue, 29 Aug 2000 04:09:06 GMT, paul snow <[EMAIL PROTECTED]>
wrote:
>
> >Bob Hauck <[EMAIL PROTECTED]> wrote in message
> >news:[EMAIL PROTECTED]...
> >> On Mon, 28 Aug 2000 02:28:10 GMT, paul snow <[EMAIL PROTECTED]>
wrote:
> >> >Bob Hauck <[EMAIL PROTECTED]> wrote in message
> >> >news:[EMAIL PROTECTED]...
> >>
> >> >> You've got it exactly backwards. Raw storage is just numbered
blocks
> >> >> on the disk. Filesystems are an abstraction created by the OS.
> >>
> >> >No, you have it backwards. Where is the OS when your computer is
off?
> >>
> >> In a pile of bits on the hard disk.
> >
> >So, your OS is in storage. And Obviously that storage can be
changed, so
> >long as the reasonable set of changes possible are documented.
>
> Sure, certainly. And being able to do this correctly is tantamount to
> re-implementing the file system layer of your OS.
Of all the complexity of a modern OS, this is only a very small part of
it. In the case of Linux, I already have an implementation (Linux
itself) and I can use it just as long as it is a different one than the
one I am supporting.
Or I can simply be *logically* separate, running within the OS and
giving up on supporting those constructs that I can't change while the
OS is running. I still get most of my advantages with this approach.
> >All that about how you would have to stop the OS to manage it. Give
> >that up too. Surely you can figure out at least one way around
that.
> >I can think of several, depending on the OS.
>
> Given the agressive caching done by modern operating systems, it is
not
> a trivial problem to change the filesystem out from under them without
> causing inconsistencies between what the operating system's idea of
the
> filesystem is and what it really is. And if that happens, your system
> is toast. This sort of thing can probably be done, but not trivially
> and yes the method would depend on the OS. I thought we were trying
to
> get away from that.
That is certainly one way to approach the problem. Another, easier way
might be to simply use an agent that runs within the OS to modify
structures that can be modified while the OS runs. Most of the
structures likely fall in this category.
Keep in mind too that the modifications we are talking about are those
involved in configuration management. This does not have anything to
do with the typical operation of the execution environment itself. If
we are simply verifying the system, then all we need is read access.
An agent can perform this for us in almost all cases.
Changes don't generally have to be made unless I am already modifying
the system's configuration.
> >Suppose the hard disk crashes. I can buy another, and assuming I can
> >lay my hands on all my CDs, I can rebuild my machine yet again
(losing
> >only my unqiue work, if I failed to transfer it too to some external
> >storage). And I supply all the answers to all the decision points
yet
> >one more time.
>
> I've had hard disk crashes and not ever had to reinstall using the
> standard installers that make you answer questions. Boot up from
> floppy, partition and format the disk, restore from the most recent
> backup. Reboot. Done. But then, I don't use Windows.
Is there *any* circumstance that you might imagine constructing a
computer system from the initial installation medium? And besides, I
don't think this situation has any sensitivity to Linux vs Windows.
Backups are backups, good to have, great when they just work. I don't
think I have ever claimed otherwise, and in fact have repeatedly
pointed out that the fact that outside construction of the disk (disk
images, disk copies, and other outside means of constructing and
configuring storage like backups) are indicative of principles that we
can get more leverage out of than we are currently.
> >Are you really saying no standard form, with a single separate
install
> >facility for a given computer system can be reasonably define that is
> >equivalent to running a bunch of installs off a set of CDs?
>
> I don't think I said it was impossible. I think I said that it was
> much more difficult that you seem to think it is. You are describing
> abstractions and just ignoring the nitty-gritty implementation details
> that make it complicated to actually do. If you think it is worth
what
> might be years of effort, well, then you are free to start work. If
> you come back in six months with a prototype that does a subset of
what
> you want and it shows promise, then some folks might actually want to
> help you. That's how it works.
You are simply being purposely dense.
You haven't pointed out a single nitty-gritty detail that makes the
approach I have described any more difficult than what we are doing
today (from a developer's point of view or install facility point of
view). You think I am ignoring the implementation details, then give
me one. Just one example of something an application needs to have
constructed in storage during an install that is in any way more
difficult to describe in XML and interpreted by a separate installation
facility as compared to describing it for a tool like NeXT packages,
RPM or even Installshield.
In fact, as you might guess, this model is likely to be far easier on
Windows than what that world is doing now.
Keep in mind that the separate installation facility gets to cheat.
While it might be nice to be a pure, literally separate facility, most
of the advantages are gained even if it is only a logically separate
facility. At the very least it can duplicate the advantages of RPM
(since the XML is easily versioned and we can specify prerequisites).
But a single, separate facility can also record the install parameters
(and/or take them as XML) so that they can be used to duplicate fully
configured, multiple installs onto other existing machines. (This is
useful if you need the same basic configuration of a number of software
components, but the targets are not homogenous.)
The XML can also be fairly easily extended to describe distributed
installs. (Most of the time a distributed install amounts to a set of
parameters for each machine based on its role; Pretty easy to describe
in XML)
The magic of XML? No, like any decent algorithm, it can be implemented
in any language. It would probably be just as easy or easier to simply
extend RPM as a format, and beef up the implementation. Support of
other platforms would amount to porting the code and adding features to
handle differences required of the other file systems and Operating
Systems.
All you need is a structured, extendable format that can describe
storage. XML fits the bill, but that fact seems to cause you grief,
and the use of the term seems to trip you up. So ignore it already.
Besides, in six months I am willing to bet you are going to see the
first products delivered using this approach.
Paul Snow
[EMAIL PROTECTED]
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: vlado <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.hardware,comp.os.linux.misc
Subject: Re: scsi bus resets with aic7xxx
Date: Tue, 29 Aug 2000 09:54:31 +0100
Reply-To: [EMAIL PROTECTED]
Thanks Dave,
>> I've run up on a "cute" problem with receiving scsi bus resets under
> > heavy load.
>
> If you've already got current drivers, then check cable lengths.
cable lengths are ok - less than 1.5m
>
>
> > Hardware details
> > ------------
> > Motherboard - ASUS P2B-DS SCPI Rev1012
> > Award bios v4.51p6
> > with onboard Adaptec 7890 bios v.2.11
> >
> > "HD" - ZERO-D 400 IDE-RAID system, firmware v.2.2W
>
> You don't say what kind of SCSI bus you have (Ultra, Ultra-Wide,
> Ultra-2, etc.)
Ultra Wide SCSI Single Ended $) MB/sec data rate
> If you mix different device types on a single bus, your overall limit is
> the tightest limit of all the devices.
There is just one single device, which is properly terminated.
Well, to sum it up a little:
- very low chance of a cabling/termination problems - a single device on
the bus, properly terminated,
anyway I tested the setup with several different scsi cables,
terminators with the same results.
- the ZERO-D box behaves properly, in other enviroments - SGI Octane,
but I haven't tried VERY heavi loads there
this is what I am going to do the next couple of hours
- the SCSI tiomeouts which cause the resets appear under a heavy load -
with vmstat I get a constant 1 - processes in
uninterruptable sleep - presumably blocked io
The chances are more towards a bug/undocumented feature of the AHA2940UW
controller and/or a related aic7xxx driver behaviour.
Anyway, things and solutions of the sort of no_resets does not cure the
problem of the occuring bus timeouts.
Cheers,
Vlado
------------------------------
From: vlado <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux,alt.os.linux.suse,comp.os.linux.misc,uk.comp.os.linux
Subject: Re: Restaurant Booking System
Date: Tue, 29 Aug 2000 10:46:18 +0100
Reply-To: [EMAIL PROTECTED]
Hi,
> What I would like to know is are there any linux based applications
> available out there that would allow online bookings, and hopefully would
> also allow this to be done for multiple locations.
Well, you could have a look at the various "shopping cart" apps available for
php, perl, zope, java and some other more exotic.
have a look at:
php.org
zend.org
www.enhydra.org
I dont't remember the URLs of the others but you can alwways do some smart
search :)
Cheers,
Vlado
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: alt.os.linux,comp.text.xml,comp.os.linux.misc
Subject: Re: Linux, XML, and assalting Windows
Date: Wed, 30 Aug 2000 00:30:57 GMT
In article <8ogup7$ea4$[EMAIL PROTECTED]>,
<[EMAIL PROTECTED]> wrote:
>
> paul snow <[EMAIL PROTECTED]> wrote in message
> news:RQNq5.22657$[EMAIL PROTECTED]...
> > <[EMAIL PROTECTED]> wrote in message
> > news:8off6l$f3c$[EMAIL PROTECTED]...
>
> > > Have you not ever encountered or heard of any of these system?
>
> > Of course. But how is the persistent state of a swich interestingly
> > different than storage? Or in other words, how is this
computational
> model
> > significantly different than a Turing machine? (Hint: it isn't)
>
> I was not addressing my reply in any way to your citation of Turing
machines
> since your citation was an invalid attempt to try to falsely
intellectualize
> your position. I for one am not foolish enough to fall into your
trap that
> you have not even tried to hide.
>
> What persistance? The switches used in this style are higly volatile
more
> so than core storage and are perhaps the equals of the processor's
registers
> in terms of the survivalibility of the data. Those toggle switches
used to
> provide the machine code instruction the the processor a byte or word
at a
> time are not used for storage. They are a means of communications to
with
> the processor or to the core storage. If you consider those toggles
as
> storage then you would have to consider the keyboards as storage as
well
> because they are used for a similar purpose.
You are simply being emotional, and as fun as it is to argue with you,
I really don't have the time.
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: Jonathan <[EMAIL PROTECTED]>
Subject: Re: wron major or minor version number for partitions. What am I doing wrong?
Date: Wed, 30 Aug 2000 00:30:50 GMT
In article <[EMAIL PROTECTED]>,
[EMAIL PROTECTED] (Marcus Webb) wrote:
> On Sat, 26 Aug 2000 18:06:19 GMT, [EMAIL PROTECTED]
> (Marcus Webb) wrote:
>
> I've now tried installing from boot floppies and the same problem
> still exists. I jsut thought I'd add that here to stop anyone
> answering with the same thought.
>
> Is there a wy for correcting the /dev/hda? file to reflect the
> necessary major or minor version numbers and if not are there any
> other suggestions out there in LinuxLand.
>
> Knowing all you lovely linux types are hardware hackers of the highest
> order and will come up with a solution soon.
>
> Regards
>
> Marcus
>
> >Hi there
> >
> >Hoping that someone here can help. I have a home built system that
has
> >two SCSI CD Drives and an IDE Hard drive. The partitioning work has
> >all been done and Linux installs from a bootable CD in either drive
-
> >no probs there. I mount the partitions I want when setting up under
> >the install, but when I reboot if I try to mount any partition other
> >than the Linux ones (i.e. my Windoze 98 stuff) I get the error
"wrong
> >major or minor version number". This is very frustrating.
> >
> >I can reinstall if necessary to fix this, so am not too fussed by the
> >situation, but it would be nice to get it fixed. Any suggestions?
> >
> >I can use RedHat, SuSE, Mandrake or TurboLinux to install. and they
> >all produce the same probs.
> >
> >Regards
> >
> >Marcus
>
>
I had this error for vfat partitions when upgrading from kernel 2.2.15-4
(Mandrake) to 2.4test7. The fix was to upgrade to modutils 2.3.15 and
rename /etc/conf.modules to modules.conf
Hope this helps ...
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: [EMAIL PROTECTED] (Roger Jagoda)
Subject: Ghostscript Make errors; jpeglib.h
Date: 30 Aug 2000 01:00:02 GMT
The hardware:
PentIII
128MB RAM
10G Hard drive
The software:
Mandrake 7.1 (new clean install)
GhostScript 6.01 (fresh download from sourceforge.net)
The problem:
We continually get error messages about the additional libraries
not being defined properly in the makefile. Here's the lines from the last Make
capture:
./obj/echogs -e .dev -a- ./obj/psf0read -oper zcfont zfcmap zfont0
./obj/echogs -e .dev -w- -l-obj ./obj/compfont -include ./obj/psf0lib
./obj/psf0
read
make: *** No rule to make target `jpeg/jpeglib.h', needed by `obj/jpeglib0.h'.
Stop.
Now, in the src tree, there are these files:
79 mandlinux % ls -la !$
ls -la *g.mak*
-rw-r--r-- 1 root system 14992 Mar 09 03:40 jpeg.mak
-rw-r--r-- 1 root system 6897 Mar 09 03:40 libpng.mak
80 mandlinux %
We're using the normal make file link as described in the docs:
lrwxrwxrwx 1 root system 16 Aug 11 14:36 Makefile ->
src/unix-gcc.m
ak
Sooo, what do we have to do to get this to work? Does anyone have
any changes that they have used in either the main .mak files or the jpeg,
libpng, etc. Makefiles?
There must be a very simple solution here, but after two weeks of
trying we're just not seeing it.
Should we make the jpeg and other libraries separately, then install them into
the /usr/local/lib,include trees? Wouldn't we then also have to change to
main Makefile to look there instead of the existing gs src tree?
Or is there something that needs to be done in the other .mak files to get
some more defines for these libraries to get made, linked, and compiled.
Any help would be very much appreciated!
Thanks in advance, really!
-R
Roger Jagoda
Cornell University
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (Colin Watson)
Subject: Re: sendmail is acting like a snail
Date: 30 Aug 2000 00:45:51 GMT
Rune Elvemo <[EMAIL PROTECTED]> wrote:
>Yeah, that's right. Right now sendmail is running REALLY slow here at
>my system.
>Fetching mail takes forever, even on my connection (6.9 mbit), sending
>mail isn't fast either, pine/mutt sits around waiting for a confirm
>from sendmail for several minutes....
>APART from that everything is like it should...
>
>BUT this is kinda annoying, so I was wondering if ANYONE out there
>would happen to know what might be causing this ?? (starting up
>sendmail on bootup takes several minutes aswell, btw)
Notoriously, this is due to DNS being broken. Check that you can
correctly look up hostnames from IP addresses, and in particular that
any local non-Internet hosts that might communicate with your sendmail
daemon (including the system running sendmail itself) are listed in
/etc/hosts.
--
Colin Watson [[EMAIL PROTECTED]]
"Alas, alas poor Willie, we'll hear his voice no more.
What he took for H2O was H2SO4" - Ruthless Rhymes for Heartless Homes
------------------------------
From: [EMAIL PROTECTED] (Colin Watson)
Subject: Re: Problem with RPM and loading of source
Date: 30 Aug 2000 00:46:48 GMT
Ken Kachnowich <[EMAIL PROTECTED]> wrote:
>I want to load the source for PostgreSQL from the Redhat 6.2 CD.
>I put in disk 2 and run the Gnome RPM tool
>I look in the SRPM directory and find the postgreSQL.src rpm file
>I click install
>
>And nothing. I tried rpm from the command line and still nothing.
>What am I doing wrong here.
Are you sure it hasn't installed into /usr/src/redhat?
--
Colin Watson [[EMAIL PROTECTED]]
"Saint Jean le Desincarne, priez pour nous! But please - let us not
have to follow your example for at least another million years."
------------------------------
From: [EMAIL PROTECTED] (Colin Watson)
Subject: Re: mount fails on hda2
Date: 30 Aug 2000 00:50:46 GMT
Rodney Loisel <[EMAIL PROTECTED]> wrote:
>physical C:drive is msdos (vfat) - Linux - msdos (vfat)
>shows up on boot as hda1 hda2 <hda5..6...7.....8>
> = (C:) (data:) (Linux /boot /root
>/swap etc)
[...]
>[root@localhost zip]# mount -t msdos /dev/hda2 /mnt/dat
>[MS-DOS FS Rel. 12,FAT 0,check=n,conv=b,uid=0,gid=0,umask=022,bmap]
>[me=0x0,cs=0,#f=0,fs=0,fl=0,ds=0,de=0,data=0,se=0,
> ts=0,ls=0,rc=0,fc=4294967295]
>Transaction block size = 512
>VFS: Can't find a valid MSDOS filesystem on dev 03:02.
>mount: wrong fs type, bad option, bad superblock on /dev/hda2,
> or too many mounted file systems
> (aren't you trying to mount an extended partition,
> instead of some logical partition inside?)
Like it says - hda2 is an extended partition, according to the output
above. You need to mount hda5, ..., hda8 separately. If you think you
have another DOS-based partition, maybe it's somewhere else?
--
Colin Watson [[EMAIL PROTECTED]]
"... and by God I *KNOW* what this network is for, and you can't have
it." - Russ Allbery (http://www.eyrie.org/~eagle/writing/rant.html)
------------------------------
** 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.setup) 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-Setup Digest
******************************
