Linux-Misc Digest #737, Volume #26 Sat, 6 Jan 01 21:13:01 EST
Contents:
Re: How do I get my NIC:s started on RH6.2 boot??? ("Sloan Miller")
Re: Sequencing large amounts of files. ("Dan White")
Re: joining two lines (Tim Haynes)
Re: ftp but not telnet ("Sloan Miller")
looking for libc-5.3.12-31.i386.rpm (Ronald Haynes)
Re: Rlimit problem? ("Dan White")
Re: Best Dual Processor board and processor ([EMAIL PROTECTED])
Re: at daemon fails to startup ("Dan White")
Re: kernel or disk error ("Dan White")
Re: How to trap output logs from xdm, kdm, startx, etc? ("Dan White")
Re: Software Inventory (Linux User)
Re: port forwarding for external machines ("Dan White")
Re: DHCPD questions ("Dan White")
Re: Red Hat dead/dying? (John A. Stewart)
Re: Safe Upgrade to glibc 2.2? (Bob Schreibmaier)
----------------------------------------------------------------------------
From: "Sloan Miller" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.networking
Subject: Re: How do I get my NIC:s started on RH6.2 boot???
Date: Sat, 6 Jan 2001 16:02:44 -0800
Reply-To: "Sloan Miller" <[EMAIL PROTECTED]>
I too think setting up these configs throught linuxconf is a great idea. Do
this then after look into /etc/sysconfig/network-scripts/ cat the file
ifcfg-eth0 and see why you were better off letting linuxconf do the job.
then cd up one dir and cat network this would be easy to write but let
Linux do it for you.
An alternative would be to add the commands to the end of the
/etc/rc.d/rc.local file this is the last file read upon boot and the config
could happen there.
Bo Berglund <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> I have two NIC:s (3C509 combo) in my RH6.2 server. Since it is a
> server install I have no KDE or any other GUI tools at hand :-(
> I have had to add the two cards to mu /etc/conf.modules file in order
> to be able to use the cards at all. But every time I start up Linux I
> have to do the following as root:
> ifconfig eth0 192.168.1.1 netmask 255.255.255.0 up
> ifconfig eth1 192.168.0.1 netmask 255.255.255.0 up
> route add default gw 192.168.0.129
>
> Only at this point does the network function.
> It is becoming rather tedious work and I want it all done as a part of
> the startup. How can this be done?
> Every time I install a Linux before I did it as a workstation and I
> could use the KDE tools for this. Then it was somehow entered into the
> system permanently, but not now. I tried adding ifcfg-eth0 and
> ifcfg-eth1 files to the network-scripts directory but then the
> ethernet driver totally refused to start so I had to remove them..
>
> I have read Linux Networking HOWTO and Ethernet HOWTO and several
> other docs but I can't find a tip on this problem....
>
>
>
> Bo Berglund
> [EMAIL PROTECTED]
------------------------------
From: "Dan White" <[EMAIL PROTECTED]>
Subject: Re: Sequencing large amounts of files.
Date: Sun, 07 Jan 2001 01:02:00 GMT
A few questions.
Is the directory structure flat? do you have the image directories in one
directory, each of which only contains files and not subdirectories?
> If I had files numbered 0-603 and there were a couple
> files missing in the middle of that sequence (because they were no
> longer needed and deleted), the program still added them to the end of
> the sequence instead of filling the gaps.
This implies that you have files located in a temporary directory that
you want to add to a certain directory. Is that right? How do you
determine which directory the files get put in?
Or are you just wanting to resequence the whole thing at once? or are you
talking about a batch script that will be used in the future to add new
files?
- Dan White
In article <jIy56.1055$[EMAIL PROTECTED]>, "Greg Hains"
<[EMAIL PROTECTED]> wrote:
> Hi,
>
> I am about to move a large amount of files from my Windows system to a
> Linux box - largely graphic files. There are thousands of files in
> hundreds of directories. I need to name them all sequentially with a
> specific naming convention/mask/pattern. I am fairly familiar with DOS
> but was not satisfied with the methods that were offered to me in
> sequencing these files - all sorts of reasons. I have only just touched
> upon the power of scripting in the Shell of Linux (specifically
> Mandrake) and would like to carry out this task on this server.
>
> My requirements: eg. A directory called 010B050A with the files called
> 010B050A_001.JPG etc - note the leading zeros and use of hexadecimal in
> the numbering. (I also want to vary the amount of leading zeros
> depending on the directory). I also want to leave the file extensions as
> they are.
>
> One major problem that was not easily overcome (by an OS batching to
> date) was the ability of the program to be able to pick up missing
> sequences. eg. If I had files numbered 0-603 and there were a couple
> files missing in the middle of that sequence (because they were no
> longer needed and deleted), the program still added them to the end of
> the sequence instead of filling the gaps. It doesn't matter if the
> numbering of each of the files change - it also reduces the amount of
> time the program takes to renumber and re-sort the files - instead of
> doing 600-odd files it only changes two or three incorrect ones.
>
> Can anybody demonstrate to me a script that will allow me to do the
> above?
>
> Greg Hains
>
>
>
------------------------------
From: Tim Haynes <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.admin
Subject: Re: joining two lines
Date: 06 Jan 2001 23:56:58 +0000
Reply-To: Tim Haynes <[EMAIL PROTECTED]>
Anthony Ewell <[EMAIL PROTECTED]> writes:
> Thank you! (You know, one of these days I
> am going to have to learn perl. It seems like you can manipulate anything
> with it!)
It would help if I actually got it right, of course.
[snip]
> > You could always pipe it through:
> > perl -npe '$c++ || chomp'
> > if you want :8)
That should be
perl -npe '$c++&1 || chomp'
so that it alternates every 2 lines. Hmmm, oops.
~Tim
--
Clouds cross the black moonlight, |[EMAIL PROTECTED]
Rushing on down to the sound |http://piglet.is.dreaming.org
of a turning world |
------------------------------
From: "Sloan Miller" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.networking
Subject: Re: ftp but not telnet
Date: Sat, 6 Jan 2001 16:18:10 -0800
Reply-To: "Sloan Miller" <[EMAIL PROTECTED]>
tcp-wrappers will allow you to regulate who can access your ftp server
behind a firewall. This is /etc/hosts.allow and /etc/hosts/deny.
datagram <[EMAIL PROTECTED]> wrote in message
news:JAv56.32120$[EMAIL PROTECTED]...
> Hi,
> is it possible to stop people accessing ftp vas telnet ?
>
> I know the answer is yes ... how ?
> links ? docs ? any help is greately appreciated.
>
> Thanks.
>
>
------------------------------
From: Ronald Haynes <[EMAIL PROTECTED]>
Subject: looking for libc-5.3.12-31.i386.rpm
Date: Sat, 06 Jan 2001 17:22:19 -0800
Reply-To: [EMAIL PROTECTED]
Hi, I am looking for a copy of this rpm file, the redhat site lists it
but when I try to download
it (by clicking on the link provided) it says file not found. A pointer
in the right direction
would be great.
Thanks,
R Haynes
------------------------------
From: "Dan White" <[EMAIL PROTECTED]>
Subject: Re: Rlimit problem?
Date: Sun, 07 Jan 2001 01:24:14 GMT
In article <934prn$l24$[EMAIL PROTECTED]>, [EMAIL PROTECTED] wrote:
> I am running a VAlinux system and I am trying to install the latest
> Legato Client. After the install, the nsrexecd will not start. If I
> run and strace on the command "nsrexecd -s <hostname>" the last 3 lines
> show:
>
> getrlimit(RLIMIT_NOFILE, {rlim_cur=32*1024, rlim_max=1024}) = 0
> --- SIGSEGV (Segmentation fault) ---
> +++ killed by SIGSEGV +++
>
> It seems like it faults right after getrlimit. Has anyone seen this
> before?
Segfaults suggest a library problem. Make sure you have the required
glibc version (or the version for your glibc).
- Dan White
------------------------------
Crossposted-To:
alt.linux,comp.os.linux.hardware,comp.os.linux.networking,comp.os.linux.setup
Subject: Re: Best Dual Processor board and processor
From: [EMAIL PROTECTED]
Date: Sun, 07 Jan 2001 01:27:26 GMT
"Joshua Butcher" <[EMAIL PROTECTED]> writes:
> What is the best dual process motherboard, and processor to buy for a =
> small - medium sized web/mysql server It is for my home, I have started =
> a business and I am running it from home, and have no exp yet with dual =
> processor combinations? I will be running RedHat 7.0. I have the =
> machine up and running now, but its an older AMD K6-3 450 with a promise =
> 66 raid controller...
Do you need CPU power? A WWW server isn't exactly CPU-intensive
unless you run lots of server-side scripts, and proper SQL queries
will keep MySQL from blocking too badly.
The things that'll help you the most are memory and disk: the former
for a big cache, the latter for when the cache isn't quite big
enough. You might consider spending your money on a nice fast SCSI
RAID setup and a 512MB stick of RAM instead of a new board and pair of
CPUs.
--
Eric McCoy <[EMAIL PROTECTED]>
------------------------------
From: "Dan White" <[EMAIL PROTECTED]>
Subject: Re: at daemon fails to startup
Crossposted-To: comp.os.linux.setup
Date: Sun, 07 Jan 2001 01:28:28 GMT
In article <J1P56.925$[EMAIL PROTECTED]>, "Ted Troccola"
<[EMAIL PROTECTED]> wrote:
> I recently duplicated a working Red Hat 6.1 system.
>
> I duplicated the entire filesystem on a second hard drive in the same
> box
> ( so no hardware changed ), updated the fstab to point to new partitions
> and
> tried to boot to the new root partition ( using LILO: linux
> root=/dev/hda7 where hda7 is my new root partition)
>
> I see everything start correctly, except for the "at daemon". This
> fails on the new system, but works fine on the old.
Make sure the permissions are right on /tmp /var/run, /var/run/utmp etc.
- Dan White
------------------------------
From: "Dan White" <[EMAIL PROTECTED]>
Subject: Re: kernel or disk error
Date: Sun, 07 Jan 2001 01:32:36 GMT
In article <[EMAIL PROTECTED]>, "Patrick Lamb" <[EMAIL PROTECTED]>
wrote:
> I've got a pile of lines in /var/log/message like:
>
> Jan 3 09:54:41 dwarfdog kernel: attempt to access beyond end of device
> Jan 3 09:54:41 dwarfdog kernel: 03:01: rw=0, want=8549516,
> limit=8056597 Jan 3 09:54:41 dwarfdog kernel: dev 03:01 blksize=512
> blocknr=17099031
> +sector=17099031 size=512 count=1
>
> Just started late last week, apparently. Can anyone tell me what it
> means, and how to fix it?
>
> TIA!
Do you happen to be running a tape device on the system?
- Dan White
------------------------------
From: "Dan White" <[EMAIL PROTECTED]>
Subject: Re: How to trap output logs from xdm, kdm, startx, etc?
Date: Sun, 07 Jan 2001 01:38:55 GMT
In article <[EMAIL PROTECTED]>, "Jason Bond"
<[EMAIL PROTECTED]> wrote:
> How does one trap the output from xdm or kdm or startx or startkde? It
> is failing for some reason and without documentation, I am a loss to
> figure out why. Thanks much.
>
> Jason
>
>
>
Try:
startx 2>&1 > /tmp/xerrors.txt
- Dan White
------------------------------
From: Linux User <[EMAIL PROTECTED]>
Subject: Re: Software Inventory
Date: Sat, 6 Jan 2001 17:39:24 -0800
Dan White wrote:
> In article <[EMAIL PROTECTED]>, "Lisa Johnson" <[EMAIL PROTECTED]>
> wrote:
>
> > Is there anything out there that will query a Linux, Unix or Irix server
> > to determine what software is running on it?
>
> If the system uses a package manager, then you can use it to list the
> software installed. On RPM based systems:
>
> rpm -qa | less
>
> On Debian based systems:
>
> dpkg -l "*" | less
I guess I was assuming from where she said "query a Linux, Unix or Irix
server" that she needed something that would do remote queries.
Jim H
------------------------------
From: "Dan White" <[EMAIL PROTECTED]>
Subject: Re: port forwarding for external machines
Date: Sun, 07 Jan 2001 01:42:54 GMT
In article <[EMAIL PROTECTED]>, "Alina Zukich" <[EMAIL PROTECTED]>
wrote:
> I run a ip-masquerading network with my cable modem. I've set up a web
> server on the linux machine(apache).
>
> On my linux box, eth0 is 192.168.0.1 and eth1 is my cable modem. My
> three other machines are running windows and are 192.168.0.x The network
> with the cable modem works just fine.
>
> Now I would like a web server so that external machines can see what's
> on the web page that I create (a web page using mysql and php stuff)
>
> I've added this to my rc.firewall: ipmasqadm portfw -f ipmasqadm portfw
> -a -P tcp -L [eth1 address] 8080 -R [192.168.0.1] 8080
> (I choose port 8080 so that I don't have to run apache as root)
>
> I also gave the server name in httpd.conf file as 192.168.0.1 (and made
> sure to change the port to 8080).
One problem is that 192.168.0.1 is not visible to systems outside your
network, you should change that to the external ip or name of your linux
system.
------------------------------
From: "Dan White" <[EMAIL PROTECTED]>
Subject: Re: DHCPD questions
Crossposted-To: comp.os.linux.networking
Date: Sun, 07 Jan 2001 01:46:49 GMT
In article <[EMAIL PROTECTED]>, [EMAIL PROTECTED]
wrote:
> I am running a Cobalt RAQ2 with Cobalt Linux. The distribution is so
> similar to Red Hat that the differences are trivial! Though, it should
> be mentioned, that this is a very lean distribution.
>
> I was avised by Cobalt that this machine will not support a DHCP server.
> I have DHCPD running. No sweat! The problem is, I have no way to start
> it upon startup in the scripts. I tried using <chkconfig
> --add>. It didn't work.
>
> Help would be much appreciated! Thanks in advance!
>
> Glenn
Did you compile it, or install it from RPM? Either way you should be able
to add it to /etc/rc.d/rc.local as a hack.
- Dan White
------------------------------
From: [EMAIL PROTECTED] (John A. Stewart)
Crossposted-To: comp.os.linux.advocacy,comp.unix.solaris
Subject: Re: Red Hat dead/dying?
Date: 7 Jan 2001 01:41:55 GMT
In <Xdu56.19180$[EMAIL PROTECTED]> [EMAIL PROTECTED] writes:
>John A. Stewart <[EMAIL PROTECTED]> wrote:
>> In <[EMAIL PROTECTED]> "Aaron R. Kulkis" <[EMAIL PROTECTED]> writes:
>>>Replace Windows with Linux....
>>>
>>>Seriously. I recently did a contract with EDS. We supported over
>>>5,000 Unix workstations at GM sites scattered all over Michigan and
>> I'm a Unix system administrator but I can see that most of our users are
>[...]
>> Our experience is that Windows machines are easier to support in large
>> numbers. We can simultaneously drop a new image on hundreds of machines
>I'm an admin in a group which supports a smaller number of machine's
>than Aaron's (somewhere between 500 and 1000) and I'm more in
>agreement with him. I can load many Sun's and/or HP's with jumpstart
>or Ignite-UX. It's quite easy.
That's fine if you are supporting a large number of machines. We are
supporting a small number of machines spread across six different subnets
so Jumpstart isn't of any practical use to us.
>For smaller updates I wrote some perl scripts years ago that can do
>almost anything I want.
>All of our desktop machines are quite similar, almost identical within
>a few classes. This does make life much easier.
Ours range from antiquated Sparc 2 machines to recent model Ultra 10
machines and just about everything in between.
>> difficult to manage because of the tendancy for Unix systems to be
>> configured to provide services and for machines to be kept in service much
>Don't do that. All common services should be provided by servers in
>locked server rooms, on UPS's and with diesel backup and mirrored
>disks. If possible fail-over is good.
I'm in full agreement there. We've had some success in consolidating
services on our central servers but some departments insist they must
have their own departmental mail/file server. Money is always a problem
in a university environment and that is why there was a tendancy to get
desktop machines to perform double duty as servers. The ultimate silliness
was our Civil Engineering department where almost every desktop machine
was serving home directories using whatever space wasn't required on the
disk drive for the operating system.
>Then when a clent machine fails, repair/replacement is trivial.
>> email and file service on a few large well maintained servers, you find
>> many departments attempting to support several different operating systems
>> and associated servers with only one or two staff members.
>I see this, and I'm not bothered, and don't try to stop it, but I
>won't provide any support (maybe a few words of advice if I'm not
>busy). If some group sets up a rogue domain, great, but they have full
>responsibility.
A serious drawback of this distributed support model is that it is
impractical to implement a firewall. There are so many departmental
machines that have to be accessible from the Internet that attempting to
develop rules to control access would be a mammoth job and the rule sets
would require frequent changes. As a result, we have to properly secure
every machine and maintain a current set of patches on all machines.
--
John Stewart -- Computing and Communications Services, Carleton University
Internet: [EMAIL PROTECTED] 613-520-2600x3707
"you are incompetent bloodlusting idiot pirates. -- CCS student feedback"
------------------------------
From: [EMAIL PROTECTED] (Bob Schreibmaier)
Subject: Re: Safe Upgrade to glibc 2.2?
Date: Sun, 07 Jan 2001 02:12:27 GMT
In article <[EMAIL PROTECTED]>,
Peter T. Breuer <[EMAIL PROTECTED]> wrote:
>muzh <[EMAIL PROTECTED]> wrote:
>> AFAIK upgrading basic libraries in a running Linux installation is
>> dangerous at best, and impossible at worst. You would have to boot into
>
>No, it's never impossible. It's dangerous in the sense that the damage
>you can do is great. So you have to decrease the expected damage by
>being Very Careful when you do do it.
>
>Peter
Thanks to you and everyone else who responded. I may just wait a while
and update the whole system. Some programs, like fetchmail 5.6.2, won't
compile with 2.1.3. So, eventually, I may want to go to glibc 2.2.
Probably don't need to do it yet.
Oh yes, one person asked about not being able to login. What happened
is that I got the "login:" message, but it wouldn't even let me type
a password. It immediately went back to the "login:" message.
Thanks again, gang!
Bob
--
+------------------- \-\-\-\ ----------------------------+
| Bob Schreibmaier K3PH | E-mail: [EMAIL PROTECTED] |
| Kresgeville, PA 18333 | http://www.qsl.net/k3ph |
+--------------------------------------------------------+
------------------------------
** 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
******************************
