RE: [Leaf-user] ramdisk_size question...

2002-02-28 Thread Adrian Stovall

Woohoo! 

and the answer is...syst_size=32M

sweet...it booted fine (with a 32M root partition).  All I had to do was
copy libnsl.so.1 to /lib and perl was happy.  I am almost ready to
rejoice...

Thanks Manfred (and guitarlynn who almost had it right...the default is
specified in bytes, so syst_size=32 didn't work as expected)

Time to play around a bit :)

-Original Message-
From: Manfred Schuler [mailto:[EMAIL PROTECTED]]
Sent: Thursday, February 28, 2002 3:17 PM
To: Adrian Stovall
Subject: Re: [Leaf-user] ramdisk_size question...


try syst_size=32M
I think you created a 32k ramdisk

Manfred

Adrian Stovall schrieb:
 
 Okay...got one step further.  I found a post from Jaques that mentions
 syst_size
 
 snip
 1/ TMPFS is used instead of /dev/ram
 /dev/ram0 is only used to boot initrd.lrp then the main filesystem is
 pivot_rooted to tmpfs and /dev/ram0 released.
 Three parameters can be introduced in the command line (but defaults
 should be OK in most cases):
 syst_size= max size of LRP filesystem (default SYSTSIZE=6M)
 log_size= max size of /var/log  (default LOGSIZE=2M)
 tmp_size= max size of /tmp (default remaining available memory)
 /snip
 
 (which I was ignorant of before).  I added syst_size=32768 to
syslinux.xfg
 and got
 
 LINUXRC: LEAF Project - Initrd - Bering-Beta4
 Mounting a 32768 TMPFS filesystem...
 Bus Error
 Bus Error
 Bus Error
 Bus Error
 Bus Error
 Bus Error
 Bus Error
 Bus Error
 Bus Error
 Bus Error
 Bus Error
 Bus Error
 Bus Error
 Bus Error
 Bus Error
 Bus Error
 Bus Error
 LINUXRC: Could not mount the boot device. Can't install packages.
 kernel panic: attempted to kill init!
 
 I'm sure this has something to do with the way my syslinux.cfg is
formatted,
 but I sure don't know what to change.
 
 syslinux.cfg reads:
 
 display syslinux.dpy
 timeout 0
 default linux initrd=initrd.lrp init=/linuxrc ramdisk_size=32768
 syst_size=32768 root=/dev/ram0 boot=/dev/hda1,msdos PKGPATH=/dev/hda1
 LRP=root,etc,local,shorwall,perl5
 
 I have tried values of 32768, 16384, and 8192 with radisk_size, and values
 of 32768 and 16384 with syst_size, but I get the same error each time.  Do
I
 have to alter something else in syslinux.cfg (or some other file) to make
 this work?
 
 -Original Message-
 From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]
 Sent: Thursday, February 28, 2002 8:01 AM
 To: Adrian Stovall
 Cc: 'LEAF (E-mail)'
 Subject: Re: [Leaf-user] ramdisk_size question...
 
 This is sort of off  LEAF , but I have had to compile my 2.4.7
 with 32mb ramsize. I really don't know if this is an issue in 2.2
 
 NOW, major contradiction!  In  /usr/src/doc  the ramdisk writeup
 specifically says the ramdisk is dynamic and will grow and shrink
 to accommodate.  This is a major contradiction of several HOWTOs
 and I don't know how to find the truth about how ram disk size works.
 
 Anyone out there knowledgeable ?
 
 Adrian Stovall [EMAIL PROTECTED] on 02/27/2002 05:49:48 PM
 
 To:   'LEAF (E-mail)' [EMAIL PROTECTED]
 cc:(bcc: Phillip Watts/austin/Nlynx)
 
 Subject:  [Leaf-user] ramdisk_size question...
 
 Hi again all...I have a new question...I made sure (repeatedly) that I had
 entered ramdisk_size=32768 in my syslinux.cfg file, however, df -k
reports
 that /dev/root has 6144 blocks allocated (which are 100% used).  How do I
 convince my router that it's supposed to use what I told it to?
 
 I have 256 Meg installed in this machine, so I'm fairly sure that's a
valid
 number to use for ramdisk_size...am I missing something?
 
 I've been searching the 'net for info, and will continue to do so, but
 haven't found anything yet.
 
 TIA
 
 Adrian
 
 ___
 Leaf-user mailing list
 [EMAIL PROTECTED]
 https://lists.sourceforge.net/lists/listinfo/leaf-user
 
 ___
 Leaf-user mailing list
 [EMAIL PROTECTED]
 https://lists.sourceforge.net/lists/listinfo/leaf-user

-- 
Manfred Schuler
E_Mail: mailto:[EMAIL PROTECTED]

___
Leaf-user mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/leaf-user



[Leaf-user] newbie question (Bering/2.4/IDE)

2002-02-26 Thread Adrian Stovall

Hi all...I had successfully finished a previous install with a 2.2.19-IDE
kernel and run from a small IDE HD.  

What I would like to do is repeat this with a 2.4 kernel (currently messing
around with Bering Beta4...no probs running from floppy).  What do I need to
do to make this run from a hard drive? 

I'm hoping for something other than compile a 2.4 kernel with IDE support
enabled, but I'll try to if I have no choice (severe lack of experience
with compiling a kernel on my own).

Is there a 2.4-IDE kernel out there?  Am I stupid, and there's some simple
config option to make the Bering 2.4 kernel boot from my HD?

I'm running this on a Dell PowerApp Web 100 (single PIII-73/256MB/dual
EEPro100) and using Bering Beta4/Syslinux 1.66 on my HD.

Any info is *greatly* appreciated.

TIA

Adrian

___
Leaf-user mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/leaf-user



[Leaf-user] packet re-assembly/content inspection...

2002-01-07 Thread Adrian Stovall

Hi all...I'm looking for any pointers on where to look for either
works-in-progress, FAQ's, ruminations, etc. on re-assembling packets
travelling into a leaf-based device (e-mail messages, http responses, etc)
so I can do filtering based on content.  I have a watchguard firewall that
does this, and I'm wondering if anyone knows anything about the mechanics of
this process.  Are there modules for this?  Is this a gigantic undertaking?
Any response is appreciated.



Adrian M. Stovall
Senior Systems Engineer
PFK Business Systems, Inc.
[EMAIL PROTECTED]

___
Leaf-user mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/leaf-user



CORRECTION: [Leaf-user] logging directive...where do I configure this? (ipmasqadm question)

2001-12-13 Thread Adrian Stovall

Actually, I had set my syslog.conf correctly...(tested my syslog machine by
throwing some broadcasts at the router and started seeing messages).  

Does ipmasqadm have an option to log hits against it?  I've been testing
using grc.com's shields-up, and I was hoping I'd be able to send messages to
syslog when a packet matched my autofw rule.

I added:

ipmasqadm autofw -F
ipmasqadm autofw -A -r tcp 1 65535 -h non.existent.ip.address
ipmasqadm autofw -A -r tcp 1 65535 -h non.existent.ip.address

to the end of my rc.firewall file, and while it successfully stealths my
protected machines, I never get notified of the rules being used.



-Original Message-
From: Adrian Stovall [mailto:[EMAIL PROTECTED]]
Sent: Thursday, December 13, 2001 4:48 PM
To: LEAF (E-mail)
Subject: [Leaf-user] logging directive...where do I configure this?


I've been hopping all over FAQ's, Guides, etc., and can't seem 
to find where
I specify a logging host.  any quick answers?

Anything that can go wrong, will go wrong.
  -Finagle's Law

If there are two or more ways to do something, and one of 
those ways can
result in a catastrophe, then someone will do it. 
  -Edward A. Murphy, Jr.

Murphy Was an optomist
   -O'Toole's commentary on Murpy's Law


Adrian M. Stovall
Senior Systems Engineer
PFK Business Systems, Inc.
[EMAIL PROTECTED]

___
Leaf-user mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/leaf-user


___
Leaf-user mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/leaf-user



SUMMARY: [Leaf-user] eepro100.o module troubles...(HD install)

2001-12-11 Thread Adrian Stovall

Thanks (Charles and Matt).  Sorry for the simple question (this is my first
attempt not using a basic coyote install).  Added the pci-scan module and
updated all the ip_masq_xxx modules...everything is running smooth.

-Original Message-
From: Matt Schalit [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, December 11, 2001 12:13 AM
To: [EMAIL PROTECTED]
Subject: Re: [Leaf-user] eepro100.o module troubles...(HD install)


Adrian Stovall wrote:

 I can still boot, but now when the network moduled try to load, I get
 
 eepro100 - /lib/modules/eepro100.o: unresolved symbol 
acpi_set_pwr_state
 /lib/modules/eepro100.o: unresolved symbol pci_drv_unregister
 /lib/modules/eepro100.o: unresolved symbol pci_drv_register
 
 Anybody have any pointers on eepro100 modules and 
IDE-enabled kernels?


Most frequently asked question these days.
As Charles said, uncomment pci-scan and be sure it
is loaded before eepro100 in /etc/modules.conf.

Cheers,
Matthew



___
Leaf-user mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/leaf-user


___
Leaf-user mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/leaf-user



[Leaf-user] eepro100.o module troubles...(HD install)

2001-12-10 Thread Adrian Stovall

Hi all,
I've managed to get my linux router to boot from a hard drive in a Dell
Powerapp Web 100 (IDE, simpler than SCSI). This machine has dual on-board
EtherExpress Pro 100's, which come up just fine using the floppy I made with
Coyote.

My problem comes into play when changing things to boot from the IDE drive.
I syslinux'ed the HD, copied the linux-2.2.19-3-LEAF-normal-IDE.zImage.upx
kernel to the HD from a floppy and renamed it, and copied everything (except
ldlinux.sys and linux) from the floppy to the hard disk. Unfortunately,
neither of the network interfaces will start, and all modules complain
loudly about not being the same version as the kernel.

I realized after looking at messages during boot that none of the ip modules
(and the eepro100.o module) had version numbers that matched the kernel
version, so as a test, I downloaded the eepro100.o module from
http://lrp.steinkuehler.net/files/kernels/2.2.19-3-normal/ (same place I got
the IDE-enabled kernel) , put it on a floppy, copied it to /lib/modules, and
backed up the boot image.
I can still boot, but now when the network moduled try to load, I get 

eepro100 - /lib/modules/eepro100.o: unresolved symbol acpi_set_pwr_state
/lib/modules/eepro100.o: unresolved symbol pci_drv_unregister
/lib/modules/eepro100.o: unresolved symbol pci_drv_register

Anybody have any pointers on eepro100 modules and IDE-enabled kernels?

Anything that can go wrong, will go wrong.
  -Finagle's Law

If there are two or more ways to do something, and one of those ways can
result in a catastrophe, then someone will do it. 
  -Edward A. Murphy, Jr.

Murphy Was an optomist
   -O'Toole's commentary on Murpy's Law


Adrian M. Stovall
Senior Systems Engineer
PFK Business Systems, Inc.
[EMAIL PROTECTED]

___
Leaf-user mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/leaf-user