Linux-Setup Digest #177, Volume #21 Sun, 6 May 01 17:13:14 EDT
Contents:
Re: Max length of password? ("Duane Healing")
"unsupported partition table.." I'm about to go postal ("Tom")
Re: File System going bad (=?ISO-8859-1?Q?Rasmus_B=F8g_Hansen?=)
USB Visor not working ("Carmine F. Greco")
Re: How to get access to shell during Linux install ?
(=?ISO-8859-1?Q?Rasmus_B=F8g_Hansen?=)
Partition Suggestions and Help (Mike Martin)
Re: Soundblaster 16 PCI setup (=?ISO-8859-1?Q?Rasmus_B=F8g_Hansen?=)
Re: RH7.1 - Can't telnet or SSH in from another host (Aaron Kaase)
Serious problem with the serial port ("Frank St�ckl")
Re: bind 9.1.0-10 problem timing out (=?ISO-8859-1?Q?Rasmus_B=F8g_Hansen?=)
Re: RH7.1 - Can't telnet or SSH in from another host (Aaron Kaase)
Re: Max length of password? (Bill Unruh)
Re: Max length of password? (Bill Unruh)
R: avoiding USB support at start-up ("Massimo Pinto")
----------------------------------------------------------------------------
From: "Duane Healing" <[EMAIL PROTECTED]>
Subject: Re: Max length of password?
Date: Sun, 06 May 2001 12:09:07 -0700
The related line in my /etc/pam.d/passwd looks like this:
password required pam_unix.so nullok obscure min=6 max=12 md5
>From what tests I've done 12 character passwords are not getting passed
when only 8 chars are supplied so it seems to work.
--
-Duane
-DNAware SoftLabs
In a feverish moment of semi-lucidity, "Bj�rn T Johansen"
<[EMAIL PROTECTED]> flailed at the keyboard thusly:
> No, I didn't.
> Is there some options I can add there to increase the password length?
> BTJ
> On Sat, 05 May 2001 10:44:42 -0700, "Duane Healing"
> <[EMAIL PROTECTED]> wrote:
>
>>Did you take a look at /etc/pam.d/passwd?
>
------------------------------
From: "Tom" <[EMAIL PROTECTED]>
Subject: "unsupported partition table.." I'm about to go postal
Date: Sun, 06 May 2001 19:21:40 GMT
I'm about to go postal.
Trying to install Suse 7.1 and when it tries to create the partitions is
says "unsupported partition table..". Ok so I figured I didn't have the
geometry setup correctly for the disk. So I added it to the command line
when I booted Yast2 and still the same problem. Played with the LILO and
linear setting to no avail. Here are the funky things:
1. When booting up I can see Linux identifies all the partitions but Yast2
doesn't see them..it always says there are no partitions.
2. I have partitioned the disk with fdisk and sysed with DOS and was able to
boot off of it (after manually configuring the disk geometry in the BIOS).
So the drive itself seems fine. I even checked with an old version of
Partition Magic and can verify the partitions- although it does not give the
proper geometry of the drive.
3. Tried using Yast1 and fdisk. It still does not see partitions and when I
try to create them it crashes. HOWEVER, LVM does identify the partitions.
4. After the crash I go to "rescue" and run fdisk manually. It says there
are 8 partitions but it only lists 4 and none of them are numbered so I
can't delete them. If I try to change the partition table in anyway it
crashes with a "segmentation fault" error. Funny thing is that the 4
partitions that are there look like they are from the last OS I had
installed - BSD4.2. But how can they still be there!!! I have fdisk /mbr the
drive and have been able to boot to DOS on it. fdisk under Yast1 doesn't see
these partitions
I'm lost...and tired.
------------------------------
From: =?ISO-8859-1?Q?Rasmus_B=F8g_Hansen?= <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.help,comp.os.linux.misc,alt.linux.slakware
Subject: Re: File System going bad
Date: Sun, 6 May 2001 21:28:01 +0200
On Sat, 5 May 2001, Scott wrote:
> Took a power hit last night, and reboot's fsck gave me
> 'Deleted inode xxxx has zero dtime. FIXED'
Took a power hit? Do you mean you just turned off the machine while the
filesystem was mounted? Then you ask for FS errors. ext2 is not
journaling. Besides the above message means, that there were no serious
trouble at all.
> Those are the kind of things that got accumulated before,
> without hostile shutdowns.
You mean that you haven't got any errors (yet) with the new RAM? Sounds
like a bad RAM block...
Rasmus
--
-- [ Rasmus 'M�ffe' B�g Hansen ] --------------------------------------
I'm a bastard. I have absolutely no clue why people can ever think
otherwise. Yet they do. People think I'm a nice guy, and the fact is
that I'm a scheming, conniving bastard who doesn't care for any hurt
feelings or lost hours of work if it just results in what I consider
to be a better system.
- Linus Torvalds
========================================= [ Remove 'spam' to reply ] ==
------------------------------
From: "Carmine F. Greco" <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Crossposted-To: comp.os.linux.redhat
Subject: USB Visor not working
Date: Sun, 06 May 2001 19:15:08 GMT
I have RH 7.1 with gnome-pilot and a USB cradle for my Visor. I
configured gnome-pilot to use /dev/pilot, which is a link to
/dev/ttyUSB1, and selected "Handspring USB." Iget to the configuration
screen that tells me to push the HotSync button, and I do, but the PDA
and gnome-pilot don't synchronize. Here's what I see in
/var/log/messages when I press the HotSync button:
May 6 13:09:32 localhost kernel: hub.c: USB new device connect on
bus2/2/3, assigned device number 8
May 6 13:09:32 localhost kernel: usb.c: USB device 8 (vend/prod
0x82d/0x100) is not claimed by any active driver.
May 6 13:09:33 localhost kernel: usbserial.c: USB Serial support
registered for Generic
May 6 13:09:33 localhost kernel: usb.c: registered new driver serial
May 6 13:09:33 localhost kernel: usbserial.c: USB Serial support
registered for Handspring Visor
May 6 13:09:33 localhost kernel: usbserial.c: Handspring Visor
converter detected
May 6 13:09:33 localhost kernel: visor.c: Handspring Visor: Number of
ports: 2
May 6 13:09:33 localhost kernel: visor.c: Handspring Visor: port 1, is
for Generic use and is bound to ttyUSB0
May 6 13:09:33 localhost kernel: visor.c: Handspring Visor: port 2, is
for HotSync use and is bound to ttyUSB1
May 6 13:09:33 localhost kernel: usbserial.c: Handspring Visor
converter now attached to ttyUSB0 (or usb/tts/0 for devfs)
May 6 13:09:33 localhost kernel: usbserial.c: Handspring Visor
converter now attached to ttyUSB1 (or usb/tts/1 for devfs)
It looks like everything is configured right, but it doesn't work. Any
ideas?
Carmine
------------------------------
From: =?ISO-8859-1?Q?Rasmus_B=F8g_Hansen?= <[EMAIL PROTECTED]>
Subject: Re: How to get access to shell during Linux install ?
Date: Sun, 6 May 2001 21:34:29 +0200
On Sun, 6 May 2001, Cedric Chausson wrote:
> I would like to have access rto a shell during install of Linux to use
> the ping command. The RedHat doc says a shell is accessible on virtual
> console 2 which should be accessible with CTRL-ALT-F2. But that doesnt
> work.
When in the installation do you try this? You have to be in 'second
stage' install before the shell on VT2 is started.
Rasmus
--
-- [ Rasmus 'M�ffe' B�g Hansen ] --------------------------------------
Chewbacca is actually not a bad SysAdmin icon. He's large, tends
towards inarticulate grunts, doesn't shave, get's stuck with a lot of
the technical work, and gets blamed and yelled at when it doesn't work.
-- Mike Sphar
========================================= [ Remove 'spam' to reply ] ==
------------------------------
From: Mike Martin <[EMAIL PROTECTED]>
Subject: Partition Suggestions and Help
Date: Sun, 06 May 2001 19:50:03 GMT
I'm just about to move onto a new hard drive and need some advice on
partitioning schemes. It is RedHat 6.2 used by a handful of users for
mail and two users for surfing and learning Linux. It will be going
onto a 2 Gig drive.
Any advice or pointers to FAQs and web sites are greatly appreciated.
------------------------------
From: =?ISO-8859-1?Q?Rasmus_B=F8g_Hansen?= <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.networking,comp.os.linux.misc,comp.os.linux.help
Subject: Re: Soundblaster 16 PCI setup
Date: Sun, 6 May 2001 21:50:53 +0200
On Sat, 5 May 2001, Scott wrote:
> I rebuilt the kernel with ensoniq 1371 loaded. I DO get sound,
> but it's kinda staticy
>
> However, 'cat /dev/sndstat' gives me Device not found.
/dev/sndstat is an OSS (OpenSoundSystem) feature. es1371 is a non-OSS
driver and thus does not support /dev/sndstat (nor /proc/sound).
> It was there when I had tried building the kernel with SB16 support.
> But now it's not there...
You should be able to get it by somehow configuring OSS - but I'm not
100% sure.
I have a SB PCI 64, and I have never used /dev/sndstat nor /proc/sound
but my card works fine anyway.
Rasmus
--
-- [ Rasmus 'M�ffe' B�g Hansen ] --------------------------------------
Drink wet cement: Get Stoned.
========================================= [ Remove 'spam' to reply ] ==
------------------------------
From: [EMAIL PROTECTED] (Aaron Kaase)
Subject: Re: RH7.1 - Can't telnet or SSH in from another host
Date: Sun, 06 May 2001 20:08:39 GMT
On Sun, 06 May 2001 12:53:19 GMT, [EMAIL PROTECTED] (James Rose)
wrote:
>Great! You figured out one of your problems on your own. You can be sure that
>the telnet server is now working (at least over the loopback device.)
>
>>HOWEVER, I still can't access the system from another computer and I
>>have no idea why. Any ideas? I can telnet OUT of the system just fine,
>>just not IN to the computer.
>
>You can also be sure that the networking stuff is loaded correctly because you
>can telnet/ping out to somewhere else.
Exactly. I can telnet, ftp, ping, traceroute out... Works fine. I
can't do anything in.
>What happens when you try to telnet from another machine? Any error messages?
It times out. No error message. And this same thing happens even
before I activated it.
>Where is the machine you are attempting to connect to, in relation to the
>machine you are initiating the connection from? Can you use another service
>(ftp, http, ssh, etc.) to connect to this machine successfully?
They are on the same LAN in the same subnet.
System A (Linux) = 192.168.123.199 (255.255.255.0)
System B (Win2k) = 192.168.124.191 (255.255.255.0)
They are behind a SMC Barricade router which acts as the gateway
(192.168.123.254). Both machines get their TCP/IP configuration
through DHCP.
I have a hunch this problem has to do with TCP wrappers that everyone
is talking about... these /etc/hosts.allow and /etc/hosts.deny files.
But it also might have to do with this xinetd, which I understand
replaces inetd in previous RedHat releases, and is also causing a lot
of confusion over how to enable services that are locked out by
default.
If someone could just tell me how to connect System B to System A
through telnet and what modifications I need to make to System A to do
this, it would provide a great baseline learning experience for me to
at least know where to start.
--
Aaron Kaase
St. Paul, MN
------------------------------
From: "Frank St�ckl" <[EMAIL PROTECTED]>
Subject: Serious problem with the serial port
Date: Sun, 6 May 2001 22:03:17 +0200
Hello,
I am having a serious problem accessing the serial port with wvdial.
One I was using SuSE Linux 6.4 and I used wvdial to dial up
to my ISP. Then I decided to build a new installation right out
of the sources because I was not satified with my installation because
it was so "canned" and I wanted to learn more about it, get rid of things
I did not really know what they did and I wantet to get rid of the package
management. So I got the "Linux from Scratch"-book and started the
installation.
Everything worked perfectly and I got a nice system. Then I wantet to
install the
internet access and I compiled wvdial but wvdial said "ttyS0: Device or
Resource busy".
I began to analyze the problem and found out the following:
1. It was not a problem of the kernel. I used various kernels: mine, the
one of BigSlack and the one supplied by SuSE. They all did not work
in my installation but in the unzipped Bigslack they worked perfectly.
Once I tried using chroot to make sure it is the same kernel and only
the other files but the problem persisted
2. The device-nodes are installed correctly: I moved them to Bigslack and
back to my installation. They worked in Bigslack but not in mine.
3. It is not a hardware or interrupt conflict. I looked up all the setserial
and stty settings in Bigslack and made sure they were all the same in my
system.
There is no difference but they do not work. Even more strange is that if I
do a
cat > /dev/ttyS0
and type anything, for example ATZ, the modem reacts.
4. I used the same binary of wvdial in both systems but the result was the
same:
Bigslack worked, mine not.
I have now read any HOWTO I could get but all of them did not help. They
said that
this problem was caused by a interrupt conflict (which is not possible in my
case)
or another program is using ttyS0 (which i also made sure is not the case).
Right now I am pretty desperate and I have no idea what the problem could
be.
I begin to think that the whole thing is just laughting at me. I hope you
can
give me any hint.
------------------------------
From: =?ISO-8859-1?Q?Rasmus_B=F8g_Hansen?= <[EMAIL PROTECTED]>
Subject: Re: bind 9.1.0-10 problem timing out
Date: Sun, 6 May 2001 22:10:37 +0200
On Sun, 6 May 2001, Shore Linux Solutions wrote:
> I am running RedHat 7.1 and I can't seem to get bind/named running
> properly. Each time I try to start it (/etc/rc.d/init.d/named start) it
>
> says that iit is starting ok however it apparently immediately times
> out. Prior to running 7.1 I ran 6.2 and I had bind v8 running fine. My
>
> second question at this point would be does anyone know of any quick or
> easy tutorials/howtos etc etc on the needed file structure and syntax of
>
> bind v9. I have looked at the man pages and the docs that came with the
>
> package but I can't seem to figure why it is timing out. The only
> examples I have been able to find are sample named.conf files. I think
> the problem might be somewhere in the zones or zone files. Any and all
> assistance in solving this problem would be greatly appreciated.
Bind 9 has a nasty habit of not starting (after telling it is starting
fine), if its files are not owned by the correct uid/gid...
Make sure /var/named, /var/log/named and /var/lock/named and all files
therein are all owned by user named, group named. You probably created
some files owned by root or something like that.
Rasmus
--
-- [ Rasmus 'M�ffe' B�g Hansen ] --------------------------------------
There are two major products that come out of Berkeley: LSD and BSD.
We don't believe this to be a coincidence.
-- Jeremy S. Anderson
========================================= [ Remove 'spam' to reply ] ==
------------------------------
From: [EMAIL PROTECTED] (Aaron Kaase)
Subject: Re: RH7.1 - Can't telnet or SSH in from another host
Date: Sun, 06 May 2001 20:30:06 GMT
On Sun, 06 May 2001 14:49:08 +0200, Michael Heiming
<[EMAIL PROTECTED]> wrote:
>Telnet is mostly controlled via tcp wrappers (/etc/hosts.allow and
>/etc/hosts.deny)
>check "man tcpd" and "man 5 hosts_access" for details, may be there is
>your problem.
I am sure you are correct, it has something to do with the TCP
wrappers. I did glance over those man pages. But a few newbie
questions linger in my mind:
Do I edit the /etc/hosts.allow text file to insert my parameters? I
assume this is the case. I put ALL:ALL in there (after all the #
comments), restarted the xinenetd service (typed '/sbin/service xinetd
restart'). No improvement at all, telneting in from my other system
still times out.
>Direct root login is disabled per default on most distros for good
>reasons.
You're right, however I can't even get to the point where I can log in
with any User ID, let alone root.
With other distros, I've had no problems telneting into the system.
I'm looking for assistance from someone experienced who has installed
RH7.1, and has had this exact same problem, and to tell me exactly
what changes I need to fix this problem.
This, combined with the man pages, will teach me a lot more than the
man pages themselves.
>On the other hand you're just asking for trouble if you run telnet,
>there are no
>so called "security enthusiasts", it's just plain stupid to use telnet,
>like buying a modern locking system for your home and then leave the key
>below the doormat...So, use ssh & scp, there is really no need to run
>telnet or
>any kind of r* services.
I'm having the same problems trying to SSH in, too. Times out. Again,
didn't have this problem with other distros.
I'm on a private network with a firewall, plus I have nothing to hide,
so I'm not too concerned with the security issues of telnet. But I
think once I'm able to telnet, I should also be able to SSH.
--
Aaron Kaase
St. Paul, MN
------------------------------
From: [EMAIL PROTECTED] (Bill Unruh)
Subject: Re: Max length of password?
Date: 6 May 2001 20:38:29 GMT
>No, I didn't.
>Is there some options I can add there to increase the password length?
Did not read the original post. The Unix crypt(3) passwords are set at 8
bytes. That is it. No way to change that.
The MD5 based passphrases can be of arbitrary length.
------------------------------
From: [EMAIL PROTECTED] (Bill Unruh)
Subject: Re: Max length of password?
Date: 6 May 2001 20:41:28 GMT
In <[EMAIL PROTECTED]> "Duane Healing"
<[EMAIL PROTECTED]> writes:
>The related line in my /etc/pam.d/passwd looks like this:
>password required pam_unix.so nullok obscure min=6 max=12 md5
>From what tests I've done 12 character passwords are not getting passed
>when only 8 chars are supplied so it seems to work.
crypt(3) is defined to use 8 bytes. That is the length of a DES
password, which is how the crypt(3) hash function uses the password.
It cannot be changed. It is written in immutable bits, with chattr +i
forever set.
If you want longer passwords use the MD5 based password scheme.
------------------------------
From: "Massimo Pinto" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.hardware
Subject: R: avoiding USB support at start-up
Date: Sun, 06 May 2001 20:51:07 GMT
Right. Sounds like a simple idea, but what if USB came as a module with the
pre-built kernel 2.2.16 that came with RedHat 7?
More generally, I would like to ask how to manage the modules that are to be
loaded at boot time, if modules are available.
Another question comes to my mind. I have compiled the kernel 2.4.2 a few
times now, looking carefully in the properties of each "element" of the
config stage, that I do via xconfig. Still there are a few features of my
pre-built 2.2.16 that I cannot reproduce with the newer kernel as I compile
it. Is there a way, given a kernel, to trace all the config options that
were given to compile it in the way it is? In this way, I could check what
do I need to include in kernel 2.4.2 to make it work better than my 2.2.16.
Thanks again
Massimo
Duane Healing <[EMAIL PROTECTED]> wrote in message
[EMAIL PROTECTED]
> Build a new kernel with no USB support.
>
> --
> -Duane
> -DNAware SoftLabs
>
> In a feverish moment of semi-lucidity, "Massimo Pinto"
> <[EMAIL PROTECTED]> flailed at the keyboard thusly:
> [yadda, yadda, yadda]
I liked that.
> > I would like to remove USB support
> > completely at start-up.
---
Massimo Pinto
Ph.D. Student
Gray Laboratory Cancer Research Trust
http://www.graylab.ac.uk/usr/pinto
------------------------------
** 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
******************************
