Linux-Setup Digest #353, Volume #21 Sat, 2 Jun 01 09:13:07 EDT
Contents:
Troubleshooting Linux I/O Network or Process Problems ("Herb Wiser")
Re: Problem with Apache and SUSE 7.0 ("Syco")
Re: sshd - slow initializing connection (Michael Heiming)
Change NIC (Smoothwall/Redhat) (Darren Wyn Rees)
Re: Problem with Apache and SUSE 7.0 (Michael Heiming)
UDMA drives and linux DMA support ("Jan =?iso-8859-1?q?Oberl=E4nder=22?=)
Can't get out of READONLY ("mkb")
new monitor X Problem (julian)
Re: 2GB File size limitation ([EMAIL PROTECTED])
Re: Can't get out of READONLY ("Peter T. Breuer")
Re: 2GB File size limitation ([EMAIL PROTECTED])
Re: configuring the modem (Uwe Malzahn)
Permission denied ("Stephen Kirby")
Re: Printer stopped working with eth1 installed (Steve Martin)
Re: Where did the program go? (Bit Twister)
Re: new monitor X Problem ("Peter T. Breuer")
Re: sshd - slow initializing connection (Bit Twister)
power up in wee hours, cron jobs run twice (Dan Jacobson)
----------------------------------------------------------------------------
From: "Herb Wiser" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.questions,comp.os.linux.hardware
Subject: Troubleshooting Linux I/O Network or Process Problems
Date: Sat, 2 Jun 2001 00:04:52 -0700
I have been running an Intel based Linux (Redhat) web server for quite
awhile now, but lately have bumped into some problems that really have me
puzzled. Whenever the machine gets into trouble, I like to go ahead and
reboot it to see if that clears it up. That is when I noticed that it takes
a VERY long time to get through the 'killall' phase of shutdown. We're
talking about way over 5 minutes, maybe closer to 10 minutes.
The next thing I noticed is that when some of my PHP code gets executed that
is doing an 'fopen' on a URL (a socket read essentially), that too takes a
long, long time. About 4 minutes or so.
Finally, I can no longer ftp into the web server using WS-Ftp. I get a
'blocking call cancelled' error. Trying to ftp into it from a DOS prompt
just hangs until it finally comes back with 'Connection closed by remote
host'.
Anyone have ANY ideas where to begin on troubleshooting this problem?
------------------------------
From: "Syco" <[EMAIL PROTECTED]>
Subject: Re: Problem with Apache and SUSE 7.0
Date: Sat, 2 Jun 2001 09:51:16 +0200
So, I can try squid.
Is it available somewhere ( I don't see it in my distro ), and is it free ?
--
SYCO de SycoWeb
[EMAIL PROTECTED]
http://ww.chez.com/sycoweb/puces/index.htm
http://ww.multimania.com/sycoweb/puces/index.htm
Chad Whitten <[EMAIL PROTECTED]> a �crit dans le message : 3b17f0bb$[EMAIL PROTECTED]
> I think what you need here would be squid and diald along with some
> ipchains/iptables rules. I know apache can be used as a proxy but squid
is
> easier and diald is what your server uses to connect when a pc on the lan
> needs and outside ip.
>
> "syco" <[EMAIL PROTECTED]> wrote in message
> news:9f7tk3$1qg$[EMAIL PROTECTED]...
> > Hi all,
> > I'm using apache 1.3 and SUSE 7.0
> > I will use apache as a PROXY, so I uncommented the lines about proxy in
> > httpd.conf
> > - apache is OK : a station on my LAN can connect my server. Apache send
> his
> > banner to the station.
> > - my dialer is OK : when I run wvdial on my server, the modem connects
my
> > provider and I can browse the web ( using lynx ).
> >
> > So,
> > how have I to do for automatic dialing when apache receives a request
from
> > LAN stations, to the proxy, for the web ?
> >
> > Thanks for answer.
> >
> >
> >
> > --
> > SYCO
> > de Sycoweb
> > ( BROCANTES DU NORD-EST http://www.chez.com/sycoweb/puces/index.htm )
> > - Let me Alone but not Forever -
> >
> >
>
>
------------------------------
Date: Sat, 02 Jun 2001 10:50:27 +0200
From: Michael Heiming <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.misc
Subject: Re: sshd - slow initializing connection
[removed some (for me) not existing ng]
inetquestion wrote:
>
> I'm running redhat 7.1 and was reading an article about different
> methods of starting sshd. One problem I noticed is that the
> connection to the server is very slow. The article mentioned
> that this could be because the services is restarting evertime a
> connection is attemped instead of running all the time.
You're not running sshd from (x)inetd, are you?
> Can anyone one offer an performance tuning to sshd or maybe
> some other areas I should look into.
Mostly those problems are due to wrong/missing DNS config, try
connecting
with IP, setup /etc/hosts on both machines and check /etc/nsswitch.conf
for the right resolution order.
Good luck
Michael Heiming
------------------------------
From: Darren Wyn Rees <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux
Subject: Change NIC (Smoothwall/Redhat)
Date: Sat, 02 Jun 2001 09:59:56 +0100
How do I change the NIC in a box running Smoothwall ?
I installed Smoothwall and it recognised my 3COM509 ISA NIC, but
now I would like to install a D-Link DFE-830TX PCI NIC (which
is an RL 8139 chipset card).
I have succesfully changed the NIC in an OpenBSD box, with a little help,
but I do not know how to do this using Linux.
--
Darren Wyn Rees [EMAIL PROTECTED]
ASK your ISP to ADD the NEW england.* Newsgroups
http://www.england.news-admin.org/accessfaq.html
------------------------------
Date: Sat, 02 Jun 2001 11:04:05 +0200
From: Michael Heiming <[EMAIL PROTECTED]>
Subject: Re: Problem with Apache and SUSE 7.0
Syco wrote:
>
> So, I can try squid.
> Is it available somewhere ( I don't see it in my distro ), and is it free ?
Where did you look? Sure it's free.
squid 2.2.STABLE5 comes with SuSE 7.0, start yast and install squid.
Michael Heiming
------------------------------
From: "Jan =?iso-8859-1?q?Oberl=E4nder=22?= <[EMAIL PROTECTED]>
Subject: UDMA drives and linux DMA support
Date: Sat, 02 Jun 2001 10:56:45 +0200
Hi,
I have a TX board with Intel PIIX4 UDMA33 support, and a bunch of UDMA
capable hard drives. My linux kernel (2.2.19) has support for dma
transfers, but when I try to turn on DMA support with hdparm -d1, I get
this:
hdc: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hdc: dma_intr: error=0x84 { DriveStatusError BadCRC }
hdc: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hdc: dma_intr: error=0x84 { DriveStatusError BadCRC }
hdc: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hdc: dma_intr: error=0x84 { DriveStatusError BadCRC }
hdc: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hdc: dma_intr: error=0x84 { DriveStatusError BadCRC }
hdc: DMA disabled
ide1: reset: success
But DMA should work... what is wrong? Or, does this not mean anything and
really it still uses udma?
Greetings,
Jan
------------------------------
From: "mkb" <[EMAIL PROTECTED]>
Subject: Can't get out of READONLY
Date: Sat, 2 Jun 2001 05:15:36 -0500
Hi, things were rolling along fine with my install of RH7.1 on a Thinkpad
i-1200 series laptop (no floppy drive). Then I hit my first major obstacle.
My PCMCIA card causes initialization to hang when it calls for the the
pcmcia files. THATS OK I found a fix, and need to change
/etc/sysconfig/pcmcia and some other files...
Problem is, when I do a text login as 'linux single' I can't get out of
READONLY mode to write my changes back to my Linux native drive! I've tried
logging in as root, chmoding on the file I want to change before editing,
heck, I even tried unmounting my drive and mounting it with the '-w' option.
NO GO. I've tried all editors and none will let me write.
I do notice this during boot: VFS:Mounted Root (ext2 filesystem) readonly
I need to ask for help at this point - any would be appreciated.
[EMAIL PROTECTED]
------------------------------
From: julian <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Crossposted-To: comp.os.linux.x
Subject: new monitor X Problem
Date: Sat, 02 Jun 2001 20:03:06 +1000
Hello
I've replaced mt 13" monitor with a 17" one. The problem is that I
have had to alter the resolution down to get a viewable image. I'm not
quite sure why. The Video card, a Trio 3D/2x has 8 Mb of RAM, but the
config only has 4 listed. How do I change this? I'm running Mandrake
7.2 Another thing is that the colour in the icons in Netscape has
gone!!
If any one has any ideas, I'm all ears
Julian
------------------------------
Crossposted-To:
comp.os.linux.misc,comp.os.linux.x,linux.redhat,linux.redhat.install,linux.redhat.misc,redhat.kernel.general
Subject: Re: 2GB File size limitation
From: [EMAIL PROTECTED]
Date: 02 Jun 2001 06:19:24 -0400
Aniartia <[EMAIL PROTECTED]> writes:
> [EMAIL PROTECTED] wrote:
>> My advice: just don't. Why do you need big files, anyway?
>
> 2Gb = ~7:17 min of 32 channel audio So when recording say a 2 hour
> session I expect to eat the greater part of a 40Gb hard drive.
Wow--impressive! Do you do that with Linux? Using what sort of software?
Len.
--
Modularity is not a hack.
-- Dan Bernstein
------------------------------
From: "Peter T. Breuer" <[EMAIL PROTECTED]>
Subject: Re: Can't get out of READONLY
Date: Sat, 2 Jun 2001 12:02:48 +0200
mkb <[EMAIL PROTECTED]> wrote:
> Problem is, when I do a text login as 'linux single' I can't get out of
> READONLY mode to write my changes back to my Linux native drive! I've tried
Why not? man mount! Pay attention to the -o remount and -w options.
> logging in as root, chmoding on the file I want to change before editing,
> heck, I even tried unmounting my drive and mounting it with the '-w' option.
What do you mean "even"? This is the only thing to try, and the
only thing to do in order to change the filesystem mode to readwrite.
But "umount"? How? I suspect you are talking about /, and you can't
umount /!
> NO GO. I've tried all editors and none will let me write.
EEEEEh? And so what? Editing is not a test of anything! If you can make
a file, then the fs is readwrite. Full stop.
> I do notice this during boot: VFS:Mounted Root (ext2 filesystem) readonly
That's correct, and normal.
> I need to ask for help at this point - any would be appreciated.
You need to state what you are doing, and the responses it provokes.
I don't see why I should spend time guessing. As it happens, I have a
pretty good idea of what you ARE doing, and it is not what you SAY you
are doing. So I'm going to wait until you come up with a
problem statement, and error log, before offering you an answer that
I could offer you now, with 70% certainty of it being correct.
There's enough info for you above, if YOU feel like doing the work of
researching it.
Peter
------------------------------
Crossposted-To:
comp.os.linux.misc,comp.os.linux.x,linux.redhat,linux.redhat.install,linux.redhat.misc,redhat.kernel.general
Subject: Re: 2GB File size limitation
From: [EMAIL PROTECTED]
Date: 02 Jun 2001 06:27:31 -0400
David Schwartz <[EMAIL PROTECTED]> writes:
>
> No, I'm being serious. I was, in fact, unable to unzip the zone file
> for 'com' and had to pipe the unzip to split.
Eek. unzip must be one of the programs which won't handle 64-bit file
offsets. FTP, Samba and gunzip are also unable to cope with such large
files.
When I received a customer database almost 10GB in size, I spent several
days upgrading glibc, the kernel, and various utilities--but in the
end, the only way to copy it to my machine was to run it through a
Win32 split utility on the source box, and copy the pieces.
(Win32 can handle big files--but many Win32 utilities can't. The situation
is just as grim in M$ country.)
I suspect it'll take years to sort out, because (1) few people bother with
big files, (2) integers as file offsets are *everywhere*, and (3) there are
*almost always* alternatives to giant files, and intelligent designers will
avoid them by not encoding open-ended data stores as single files.
Len.
--
I'll be happy to accept a lottery ticket as a gift - but I'll never
buy one.
-- Warren Buffett, 1985
------------------------------
From: [EMAIL PROTECTED] (Uwe Malzahn)
Subject: Re: configuring the modem
Date: Sat, 2 Jun 2001 12:32:05 +0200
In article <[EMAIL PROTECTED]>,
[EMAIL PROTECTED] (Victor S. Miller) writes:
> More information:
>
> I'm running RH 7.0 with kernel 2.2.19-7.0.1.
>
> setserial -a /dev/ttyS1 yields
>
> /dev/ttyS1, Line 1, UART: 16550A, Port: 0x02f8, IRQ: 3
> Baud_base: 115200, close_delay: 50, divisor: 0
> closing_wait: 3000
> Flags: spd_normal skip_test
>
> and ls -l /dev/ttyS1
>
> crw-rw---- 1 root uucp 4, 65 Aug 24 2000 /dev/ttyS1
^^^^
You should be in the group uucp.
Cheers,
Uwe
--
"Early to rise, early to bed, makes a man healthy, wealthy and dead."
-- (Terry Pratchett, The Light Fantastic)
------------------------------
From: "Stephen Kirby" <[EMAIL PROTECTED]>
Subject: Permission denied
Date: Sat, 2 Jun 2001 21:35:12 +1000
I am stuck on a problem with access permissions. The file access
permissions on files and directories seem to be correct. I log in as root
OK but if I try to
change to another user I get permission denied.
What else is controlling the permissions if it is not the file and directory
permissions?
Have I inadvertantly changed a config file somewhere?
eg
su auser
I get
permission denied /bin/bash
I have tried
su nobody
and get
permission denied /bin/sh
ls -l /
shows
drwxr-xr-x root root .... bin
ls -l /bin/sh
shows
rwxr-xr-x root root ..... sh
ls -l /bin/bash
shows
rwxr-xr-x root root ..... bash
------------------------------
From: Steve Martin <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: Printer stopped working with eth1 installed
Date: Sat, 02 Jun 2001 08:13:39 -0400
Sasan Iman wrote:
>
> My printer (HP laserjet) stopped working after I added my second
> ethernet card. Also, my eth1 (second card) does not come up
> automatically during boot. Any idea what may be going wrong? I am
> running Redhat 6.2 on an intel machine.
My off-the-cuff guess would be that the second ethernet card is
causing a hardware resource conflict with the printer port.
Double-check IRQs and I/O addresses.
As for making two ethernet cards work, there is some good info
on that in the Ethernet-HOWTO, available (probably) on your
system or install media, or online at www.linuxdoc.org.
------------------------------
From: [EMAIL PROTECTED] (Bit Twister)
Subject: Re: Where did the program go?
Reply-To: [EMAIL PROTECTED]
Date: Sat, 02 Jun 2001 12:24:07 GMT
On Fri, 01 Jun 2001 23:04:41 -0400, Tsugi Tanaka <[EMAIL PROTECTED]> wrote:
> Hello. I am wondering where the program went after installing with
>"rpm -i" command.
>Is there any way to put the program name on the program menu?
>
Helps if you provide what distro and release level you are using
when you post questions to the news groups. Different distros
have different commands, files, and links to files. Even happens
between release levels of the same distribution.
If window manger problem/question give it's name (gnome, kde, sawmill,...)
So, assuming Redhat or Mandrake, as root run
updatedb This runs for a while to build the file location database
now you can do a
locate -i file_2_find_here
example to locate program about network:
locate -i net | grep bin/
example to locate network howto:
locate -i net | grep -i howto
The warranty and liability expired as you read this message.
If the above breaks your system, it's yours and you keep both pieces.
Practice safe computing. Backup the file before you change it.
Do a, man command_here or cat command_here, before using it.
------------------------------
From: "Peter T. Breuer" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.x
Subject: Re: new monitor X Problem
Date: Sat, 2 Jun 2001 14:16:06 +0200
In comp.os.linux.setup julian <[EMAIL PROTECTED]> wrote:
> I've replaced mt 13" monitor with a 17" one. The problem is that I
> have had to alter the resolution down to get a viewable image. I'm not
> quite sure why. The Video card, a Trio 3D/2x has 8 Mb of RAM, but the
Whay aren't you sure? I'm not sure why you aren't sure since I think
you should be sure, being the only person sure of your actions and
capabilities ;-).
> config only has 4 listed. How do I change this? I'm running Mandrake
You edit the config file and stipulate that you have 8MB, if you do.
> 7.2 Another thing is that the colour in the icons in Netscape has
> gone!!
Normal. That means you are running now in 24 bit mode: 3 bytes per
pixel. With 4MB, You should be able to do 1152x768 resolutions easily.
Peter
------------------------------
From: [EMAIL PROTECTED] (Bit Twister)
Crossposted-To:
comp.os.linux.misc,comp.unix.questions,linux.redhat,linux.redhat.install
Subject: Re: sshd - slow initializing connection
Reply-To: [EMAIL PROTECTED]
Date: Sat, 02 Jun 2001 12:45:12 GMT
On Sat, 02 Jun 2001 00:19:52 -0400, inetquestion
> <[EMAIL PROTECTED]> wrote:
>I'm running redhat 7.1 and was reading an article about different
>methods of starting sshd. One problem I noticed is that the
>connection to the server is very slow.
>The article mentioned
>that this could be because the services is restarting evertime a
>connection is attemped instead of running all the time.
The "running all the time" would mean the target box has sshd daemon
running just waiting for a ssh connection.
A box would have to be maxed out to not be able to spin up
a ssh session pretty quick.
>
>Can anyone one offer an performance tuning to sshd or maybe
>some other areas I should look into.
what might be happing is it is taking a while to resolve node names.
Between my two boxes I have each other's box names and ip addresses in
each box's /etc/hosts file and I have DNS search local host first by
having localhost before my isp's domain in /etc/resolv.conf. Example:
cat /etc/resolv.conf
search localhost grlnd1.tx.home.com
nameserver x.x.x.x
nameserver x.x.x.x
------------------------------
From: Dan Jacobson <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux.mandrake
Subject: power up in wee hours, cron jobs run twice
Date: 01 Jun 2001 05:16:59 +0800
I don't know it this happens in your linux distribution, but every day
say I power up at 1:00 AM. and a few minutes later anacron runs the
appropriate jobs. At about 4AM they get rerun by the main cron, etc.
Today is the first of the month so it is running the cron.monthly so
it is even more ridiculous, with about an whole hour of wear and tear
on my hard disk.
# grep monthly /var/log/syslog
Jun 1 01:35:42 localhost anacron[711]: Will run job `cron.monthly' in 15 min.
Jun 1 01:50:43 localhost anacron[711]: Job `cron.monthly' started
Jun 1 01:50:46 localhost anacron[1128]: Updated timestamp for job `cron.monthly' to
2001-06-01
Jun 1 02:13:17 localhost anacron[711]: Job `cron.monthly' terminated
Jun 1 04:42:00 localhost CROND[8423]: (root) CMD (run-parts /etc/cron.monthly)
Jun 1 04:42:01 localhost anacron[8426]: Updated timestamp for job `cron.monthly' to
2001-06-01
They should have designed this stuff better: "ran so recently, don't
run again." They assume if one powers up once a day, it is after 5
AM. By the way, if one shutdowns the computer during one of these
jobs, often the job running doesn't take care to only overwrite the
thing it was updating after final success, so one ends up with a hole
instead of just an non updated version, e.g., makewhatis.
Seen in mandrake 7.2.
--
http://www.geocities.com/jidanni Tel886-4-25854780 e-mail:restore .com.
------------------------------
** 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.setup.
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
******************************
