Linux-Networking Digest #476, Volume #11 Thu, 10 Jun 99 03:14:35 EDT
Contents:
Re: DHCP and Newbie Linux user (Dann Church)
Re: Samba Help ("Kelvin Barnes")
Re: Problems with icmplog & partial solution ([EMAIL PROTECTED])
PPP-COMPRESS not found ("Charles E. Hill")
Re: Restricted telnet access ("Network Administrator")
Re: networking cable help... (Dann Church)
Re: [Fwd: Using SAMBA on Linux] (Nicholas E Couchman)
Re: Samba and win98 problems (Nicholas E Couchman)
Re: Linux & Cybercafe (Bernd Strieder)
Re: usr sportster 28.8 modem init string for office use HELP (Alan Bettridge)
Re: watching the local network (Walt Boring)
Re: linux <--> win98 via network (Monte Phillips)
Re: How can I tell if it is Linux? (Christopher Mahmood)
Re: Linux can't be a big role...???!!! (Jeremiah)
Squid & diald (Darren Durbin)
Re: PPP nightmare - HELP (Gilford Wimbley)
pppd? ("Jack Zhu")
Gateway question. ("David Syratt")
Re: slow ppp connection on External ISDN modem (Clifford Kite)
Re: Slow telnet and ftp responses (Nicholas E Couchman)
Re: INN and a list of newsgroups ("Gero H. Marten")
Re: Linux Router and xDSL ("Michael Faurot")
Re: user-space NFSd on Linux 2.2.x? (Matthias Meixner)
----------------------------------------------------------------------------
From: Dann Church <[EMAIL PROTECTED]>
Subject: Re: DHCP and Newbie Linux user
Date: Wed, 09 Jun 1999 22:39:02 -0600
Not sure how well dhcpcd parses options, so try the following in the ifup
script:
/sbin/dhcpcd -r -HD .....
and see if that works. Should be the same as running at the command line.
Good luck!
--Dann Church
"Daniel J. Swartz" wrote:
> I just put together a PII 350 system with a 3com 905B Nic and Caldera Open
>
> Linux 2.2. Everything is good so far. NIC works. My problem is this: I have
>
> an ADSL internet service that uses dhcp. I followed the dhcp mini howto to
>
> the letter. It did not work, I could not connect to the internet. Then I
>
> tried the troubleshooting section of the mini howto and tried running
>
> dhcpcd with the -r switch from a terminal window. That worked. But
>
> everytime I reboot I have to open a terminal window and type that command.
>
> The howto also said "If this solves your problem add the "-r" flag to the
>
> boot up scripts ie. instead of /sbin/dhcpcd you will have /sbin/dhcpcd -r"
>
> How do I do this? I tried to edit /etc/sysconfig/network-scripts/ifup
>
> like it said, and I fouund a line that said something like .../sbin/dhcpcd
>
> -HD... and I inserted the lowercase r in front of the HD and rebooted but
>
> it still did not work. I sill have to manually enter dhcpcd -r for my
>
> internet to work.
>
> Please go easy on me as I am just learning Linux.
>
> Thanks in advance
>
> ------------------ Posted via SearchLinux ------------------
> http://www.searchlinux.com
------------------------------
From: "Kelvin Barnes" <[EMAIL PROTECTED]>
Subject: Re: Samba Help
Date: Wed, 9 Jun 1999 11:54:30 -0400
> files as root, using the MS "net use" command. When I ftp or telnet with
my
> modem disconnected, the modem dials out.
There is a settings option on the software that manages your modem dialer
that causes it to fire up as soon as any sort of tcpip operations/clients
start a process. Disable this and your problem will go away.
------------------------------
From: Jonathan Johnson ([EMAIL PROTECTED])
Subject: Re: Problems with icmplog & partial solution
Date: Thu, 10 Jun 1999 04:55:36 GMT
[ Article reposted from alt.os.linux.turbolinux,alt.os.linux.slackware ]
[ Author was [EMAIL PROTECTED] ]
[ Posted on Thu, 10 Jun 1999 01:37:04 GMT ]
Just realized I only posted this to alt.os.linux.turbolinux and not to
alt.os.linux.slackware like I promised! Sorry if you see it twice.
[EMAIL PROTECTED] (Hey! That's me!) wrote:
> Hi, all!
> I've noticed a problem with icmplog. Apparently, sometimes it goes berserk
> and starts flooding the network with icmp requests. I don't know why.
> When it does this, the network slows to a crawl, and humongous logfiles are
> generated, not to mention the tons of messages mailed to root.
> Here's what I can figure out: icmplog is _supposed_ to log icmp events
> happening on the box in question. Sometimes it goes haywire and sends icmp
> requests (?) to possibly the nameserver, or possibly the default gateway.
> I'm not real sure. The machine I noticed this on was apparently sending the
> packets to 192.168.0.1. It sends the packets out so fast (thousands of times
> per second) that the destination machine cannot respond, resulting in error
> messages (see log files in /var/log/) like "last message repeated 1739
> times" with timestamps only one second apart.
> So what I did, was first find the process ID number of icmplog:
> #ps auxww | grep icmplog
> Then I killed it (as root)
> #kill 174 (where 174 is the PID)
> This resulted in a TREMENDOUS decrease in network traffic and hence an
> increase in network throughput.
> To prevent this from happening again, I edited the file:
> /etc/rc.d/init.d/iplog
> and commented out the line which starts icmplog.
> I sure hope there are no other processes which call icmplog and will cause
> it to start up again!
> I would be interested in hearing if anyone else is having similar problems
> and if this fixes it for you.
> If those 'in the know' would e-mail me, I'll try and find out what version
> of icmplog (/sbin/icmplog) is giving the trouble so they can hopefuly come
> up with a fix. Unfortunately, I have neither the talent nor the time to come
> up with a patch for the problem, so I can only disable it for now. If I get
> a chance, I'll try and find some lines from a logfile. These are customer's
> machines, so I can't use them as a test bed.
> --Jon Johnson
> [EMAIL PROTECTED]
> Sutinen Consulting
> P.S.--
> Distribution: TurboLinux 3.1
> Kernel: 2.0.36
> Running Samba and httpd
> EOF
--
[EMAIL PROTECTED]
--
[EMAIL PROTECTED]
------------------------------
From: "Charles E. Hill" <[EMAIL PROTECTED]>
Subject: PPP-COMPRESS not found
Date: Thu, 10 Jun 1999 00:09:32 -0400
I recently installed RH6.0 and all was working fine. Then I felt a
compulsion to screw with it. :-)
I configured, compiled and installed 2.2.9 along with associated modules.
(RH 6.0 comes with 2.2.5) Now PPP connections fail to work.
The modem dials, and attempts to connect, then died with an error "ppp
daemon quit unexpectedly". Checking the logs there are three errors in a
row:
insmod can't locate ppp-compress-21.o
insmod can't locate ppp-compress-26.o
insmod can't locate ppp-compress-24.o
Then it dies.
PPP was not compiled as a module in my 2.2.9, and booting back to the old
kernel doesn't help (same error).
I tried installing a newer PPP.rpm (2.3.8) but that didn't help either.
Can anyone help? I can only log on when booting to non-Linux!
chill
------------------------------
From: "Network Administrator" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.misc
Subject: Re: Restricted telnet access
Date: Wed, 9 Jun 1999 14:36:41 -0400
> One thing, your /etc/passwd file could be vulnerable for password
> cracking if you don't implement shadow passwords.
> It's built-in in the 5.2. YOu only have to enable it but there are
warnings
> of course.
Thanks for the suggestion.
We do use shadow passwords. The problem has been people creating
mailing lists using the usernames from /etc/passwd.
Regards,
--Jesus Alvarez
------------------------------
From: Dann Church <[EMAIL PROTECTED]>
Subject: Re: networking cable help...
Date: Wed, 09 Jun 1999 22:16:47 -0600
With ethernet/twisted pair, you should be able to exend a single segment 180
meters. Beyond that, you'll need to have some kind of repeater to regenerate
a clean signal. The other caveat is that no two machines should be farther
than 5 x 180 meters apart (cablewise). Otherwise you run the danger of late
collisions.
Good Luck!
--Dann Church
cyberjb wrote:
> im trying to run a network cable like... 500ft... hub to pc style... RJ45
> cable
>
> some ppl say just buy 100ft cables and get a cuppluer. but i dont know
> anyone else have someting to help me with?
> or should i just do that cuppluer thing
------------------------------
From: Nicholas E Couchman <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.questions
Subject: Re: [Fwd: Using SAMBA on Linux]
Date: Wed, 09 Jun 1999 18:26:23 GMT
When you want to mount, use smbmount. This allows you to look at Win9x/WinNT
directories from Linux and read/write/exec. as if they were just another directory
on your linux box. I'm not sure what version of Samba you have, but if you can,
get the latest. If you do get the latest, make sure you enable password
encryption, otherwise samba will do nothing with Wintel machines. Here is how I
use the smbmount command:
smbmount \\\\Server\\share -U user -W workgroup -c 'mount /dir/to/mount/to'
substituting in my values, of course. Since I'm guessing you're new to Samba, you
must, when using smbclient or smbmount, you must use four backslashes before the
server and two before the share. This is the only way it will work because of
Unix/Linux command line issues.
For problem #2, make sure you enable password encryption in Samba (there should be
a line about it in /etc/smb.conf)
Problem #3 is one of those microsoft things. Whenever I transfer data from Win9x
to anywhere (Linux via Samba, NT, etc.), all of my files go to lower case. It has
to do with FAT(32) partitions not supporting case sensitivity. NTFS doesn't do
it.
--Nick
Dan Teodor wrote:
> Hello all,
>
> Here is a problem everybody has run in to but I cannot find any HOWTOs on it,
> strangely enough!!
>
> I have a Windows 95 desktop machine, a windows 95 laptop and a Linux RedHat
> 5.2 desktop machine.
>
> They're all networked together through a 10baseT hub.
>
> I have set up the SAMBA server on my Linux box and when I turn on the Windows
> 95 machines, lo and behold, the Linux machine is there on the Network
> Neighborhood tab at the bottom of windows explorer! Great. Dandy. I can now
> use the windows 95 machines to pull and push files to the drive on the Linux
> box...and if I get really fancy, I can even store my executables there!
>
> Problem #1: How can I get the Linux machine to use the
> SAMBA/Network-Neighborhood connectivity to access the Windows 95 machines? I
> would really like to be able to do something like:
> mount -blabla /dev/win95machine /mnt/machine2/
>
> Is there any way to use samba connectivity to do this? I really cannot afford
> to spend the money to buy an NFS server for the windows 95 machines.
> Hummingbird sells one but it is *very* expensive. How can I access the windows
> 95 machines from the LInux box? This SAMBA thing only seems to go one
> way...and I am sick of FTPing to the 95 machines from the Linux box.
>
> Problem #2: This I suspect is a Windows 98 related problem. Like I said, I
> have a desktop 95 machine and a laptop 95 machine. I upgraded the laptop to
> windows 98 a few weeks ago. When I got on my local network, whenever I tried
> to open up Network Neighborhood and access the drives and printer on the Linux
> machine, WIndows 98 always came back with the dialog box "Please Enter
> password to access resource \\LinuxBoxName\...". I have double and triple
> checked. The account I was using on the Windows 98 machine was valid on the
> Linux box. THe password was correct. On the 95 desktop machine, this account
> and password work without a hitch. However, on the Windows 98 machine, windows
> 98 kept claiming that the password to access the LInux box through Network
> Neighborhood was incorrect.
>
> As proof of the fact that the problem is windows 98 specific, I reloaded
> windows 95 on that laptop after this incident and, after installation and
> account creation...everything went back to normal. I was accessing the Linux
> box through the Network Neighborhood tab like normal!!
>
> Has anybody else run across this? Has anybody found any solutions or done any
> detective work? I would be really interested in hearing about it.
>
> Problem #3: This one qualifies more as a nuisance than a real problem. I use
> text editors very often to edit files directly on the linux box while sitting
> in front of the 95 desktop machine. I simply run the text editor, open the
> file on the LInux box through Network Neighborhood, edit it and save it. No
> problem, works beautifully.
>
> The interesting part comes when I select entire directories on my windows 95
> directory tree and then paste it somewhere on the Linux box directory tree.
> THe transfer goes through just fine and no data is corrupted. However, ALL OF
> THE TRANSFERED FILES HAVE THEIR FILE NAMES CONVERTED TO LOWER CASE!!!
>
> This is extremely aggravating as I have a number of development projects whose
> source code I keep in directory trees and whose files contain some capitals
> (the most egregious example being the individual "Makefile"s contained oin
> each subdirectory). Obviously doing a make causes all kinds of file to not be
> found because of all the lost capitals. I have resorted to tarring files up,
> transfering the tarred file and then untarring them on the LInux box. However,
> WHY THE HELL DOES A TRANSFER OVER A SAMBA SERVER REMOVE THE CAPITALS FROM ALL
> THE FILE NAMES??!!?? Who is the nincampoop that wrote that functionality in
> there? Please help!! Is there an option somewhere in the smbd/nbmd setup files
> that I do not know about that will turn this behavior off?
>
> Dan Teodor
> [EMAIL PROTECTED]
>
> PS: When posting a reply, please CC it to my e-mail address as well. I read my
> e-mail more often than I do the newsgroups.
------------------------------
From: Nicholas E Couchman <[EMAIL PROTECTED]>
Subject: Re: Samba and win98 problems
Date: Wed, 09 Jun 1999 18:28:00 GMT
Actually, WinNT uses encryption unless you go mess around w/ your registry
(like Win98).
Dirk Riebesell wrote:
> Win95 and NT are transfering plain passwd. Win98 are using a security
> protocol. You have to change this in your Win98 Registry (Don't know the
> full path)
>
> Greetings
> Dirk Riebesell
------------------------------
From: Bernd Strieder <[EMAIL PROTECTED]>
Crossposted-To: alt.uu.comp.os.linux.questions,comp.os.linux.development.system
Subject: Re: Linux & Cybercafe
Date: Wed, 09 Jun 1999 20:27:22 +0200
David Knight wrote:
>
> Maurice Kemmann <[EMAIL PROTECTED]> writes:
>
> > MicroNg schrieb:
>
> > > however, how to limit the access that so the user can only access the
> > > browser but NOT any other
> > > program ? ( to access other program, for eg for the webmaster to shutdown
> > > the computer, an passcode
> > > is required).
>
> > You can configure the windowmanager without any xterm. So the users are
> > not able to start any other program. On the desktop you offer only the
> > browser. Login should be via xdm !
> > That's it i think !
>
> That it itself probably wont work, just put in netscape
>
> telnet://localhost
>
> and it will open an xterm running the telnet session, from which they
> would then be able to start other programs.
>
> David
>
I would try:
restrict execution rights, restrict the PATH environment variable to the
minimum you require, create a special bin-directory with only those apps
symlinked to you need and set PATH only to this path.
Give the users a restricted shell as login shell if you would like to
give them a shell.
unconfigure anything like this telnet://localhost in netscape if
possible. make all configuration data but bookmarks of netscape
unwritable. Let the login-Script refresh the login-directory of those
guest users.
Have daemons that check your system regularly if nothing has changed by
checksums.
Restrict the rights to access daemons of your system by wrappers even
for users of it.
HTH
Bernd Strieder
------------------------------
From: Alan Bettridge <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.misc
Subject: Re: usr sportster 28.8 modem init string for office use HELP
Date: Wed, 09 Jun 1999 23:07:30 -0600
Terry Moore wrote:
>
> Greetings,
>
> I have usr sportster 28.8 (is not PNP )
>
> at home on a POTS line ATDT123-1234 works fine.
>
> when i took it to the office it requirs 9 -123-1234
>
> ATDT9123-1234 fails with no dial tone
>
> i have tries many variations of , ; \ to get it to work.
>
> does anyone have a string that works for this application ??
>
> Thanks in advance,....
>
> linux 2.0.34 slackware...............
>
> at diald sting would work too..
>
> Terry
Your message doesn't indicate if you have an analog phone
line or not. First, verify that you do. If you don't then
the modem will not detect a dial tone. Second, hook up an
analog phone into the "phone" (other) jack on the modem and
see if you get a dial tone. If so, then dial a 9 and listen
for what happens. Do you get a second dial tone? If you do
then try using AT DT 9 W 123 45678 (the spaces are just for
readability).
Try using AT$ and AT&$ for online help from the modem.
good luck.
Alan
------------------------------
From: Walt Boring <[EMAIL PROTECTED]>
Subject: Re: watching the local network
Date: Wed, 09 Jun 1999 11:36:48 -0700
Roberto Nunnari wrote:
> Hi there,
>
> I need to watch what's going on the local network.
> Can any kind soul explain to me how this can be done using
> my linux (slackware 4.0) box? I especially need to watch
> what a particular machine puts on the ethernet.
>
Try tcpdump for starters. If that doesn't do it for you, try iptraf or
ksniffer.
Walt
------------------------------
From: [EMAIL PROTECTED] (Monte Phillips)
Subject: Re: linux <--> win98 via network
Date: Thu, 10 Jun 1999 04:40:44 GMT
You didn't mention your lmhosts files! they must be present on both
machines as well.
g'Luk
On Thu, 10 Jun 1999 00:26:23 -0400, Andrew King <[EMAIL PROTECTED]>
wrote:
>I realize this may have been answered many times before, so please
>forgive any possible redundancy. I'm trying to get my Linux box and a
>Win98 laptop to ping one another on a network via a 10/100 ethernet hub,
>yet all my config attempts have been exhausted to no avail. Can any of
>you who have done this before provide me with the details?
>
>Here's a brief synopsis:
------------------------------
From: Christopher Mahmood <[EMAIL PROTECTED]>
Subject: Re: How can I tell if it is Linux?
Date: 09 Jun 1999 22:11:42 -0700
i usually use http://www.netcraft.co.uk/cgi-bin/Survey/whats
-ckm
------------------------------
Reply-To: [EMAIL PROTECTED]
From: [EMAIL PROTECTED] (Jeremiah)
Subject: Re: Linux can't be a big role...???!!!
Crossposted-To: comp.os.linux.setup
Date: Thu, 10 Jun 1999 05:40:59 GMT
In article <7jnhnk$arl$[EMAIL PROTECTED]>,
"Jack Zhu" <[EMAIL PROTECTED]> spake thusly:
>
>> MS provides limited support to their customers too... and
>>somehow this is enough?
>>
> I'm not saying I like MS, just try to say Linux is still filled with
> disadvantages.
At the moment, you're right... but to say that it will
always be that way is a stretch. Remember Win3.1? MS has improved
a lot since then. Linux will continue to improve and will be a viable
competitor on the desktop within a year. Whether it will take over
is another topic...
>> Puh-leaze. What about companies with little money that need a
>>stable OS? Are they going to use NT? Hahahahaha. Seriously, Linux
>>entered the corporate world because IT people were given budgets and
>>told to make it work... Linux was the cost-effective choice. Your
>>"simple as this" is neither simple nor true.
>>
>
> I know Merrill Lynch use Oracle under NT.
Heh... so ML fall under the category of those companies that
don't have enough money to use Unix? :)
> Basiclly I agree with you. But can u give me more tip about my questions?
No, sorry... I haven't had use to use PPP since I got a
cable modem. My guess is that either it's a permissions problem,
or they've changed the location of the pap-secret (to a different
file or something). The documentation of PPP should say... (I'm
pretty sure the PPP HOWTO won't answer your question)
Best of luck,
Brian
--
email to bmeloon1 at twcny dot rr dot com. evilquaker is a spam collector.
------------------------------
From: Darren Durbin <[EMAIL PROTECTED]>
Subject: Squid & diald
Date: Wed, 09 Jun 1999 20:02:56 +0100
Hi,
I'm trying to get Squid to work in conjunction with diald, but I'm
running into a major problem.
diald itself works fine - if the link is down and I try to ping a host,
the line comes up and packets start returning after a few seconds,
similar results with telnet and ftp.
Squid also works fine *if the line is already up*. If I force the line
up then try a connection through Squid everything works fine.
However, if the line is down, when I connect to the proxy the line is
brought up successfully but Squid retrieves no data and times out.
If, during the timeout period, I wait until I am certain the line has
been brought up, click the stop icon in the browser and reenter the
address, it all works fine as the line is up at the time of the request.
This is causing problems with users who sit and wait until the timeout
occurs, look at the screen, re enter the address, by which time the line
has dropped, and round and round they go.
The timeout is coming from Squid, and not the browser ( IE4.0 in this
case )
Any ideas on how I can get Squid working in harmony with diald ?
Thanks,
Darren
--
Darren Durbin 'I know that this is vitriol, no solution,
[EMAIL PROTECTED] spleen venting, but I feel better having
screamed on you..." - R.E.M
------------------------------
From: [EMAIL PROTECTED] (Gilford Wimbley)
Subject: Re: PPP nightmare - HELP
Date: Thu, 10 Jun 1999 06:08:24 GMT
On Sat, 5 Jun 1999 18:25:14 +0300, "Socrates Charalambous"
<[EMAIL PROTECTED]> wrote:
>I have serious problem with my ppp connection.
>
>I connect to my ISP with minicom, I get out of minicom without reseting the
>line and start pppd.
>It stays connected for about 20 seconds and then the connection terminates.
>
>This is what pppd logs in the /usr/log/debug
>
>Jun 5 03:21:56 seaquest pppd[137]: sent [LCP ConfReq id=0x1]
>Jun 5 03:22:15 seaquest last message repeated 6 times
>Jun 5 03:25:26 seaquest pppd[153]: sent [LCP ConfReq id=0x1]
>Jun 5 03:25:47 seaquest last message repeated 7 times
>Jun 5 03:27:40 seaquest pppd[172]: sent [LCP ConfReq id=0x1]
>Jun 5 03:28:01 seaquest last message repeated 7 times
>Jun 5 03:29:45 seaquest pppd[180]: sent [LCP ConfReq id=0x1]
>Jun 5 03:30:06 seaquest last message repeated 7 times
>
>I use kernel 2.3.0 and pppd 2.3.4
>
>Please HELP.
>
>
>
I personally think you are doing the right thing using minicom so you
can see what is going on. I wonder if you need to use chap or pap
authentication? Is it possible to call your isp and ask whether and
what type of authentication is used? Also, there are *many* options
that can be passed to pppd. You might need to set some option or
other.
What is the command line you use to invoke pppd? What do you see, if
anything, on the screen when you use minicom?
anyway, keep at it!
regards,
GW
------------------------------
From: "Jack Zhu" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup
Subject: pppd?
Date: Thu, 10 Jun 1999 01:26:51 -0400
I used Redhat 5.0(Kernel 2.0.32) + pppd 2.2.0 to dialup to my ISP. It works
well. Then I upgrade kernel to 2.2.1, obviously it doesn't work with pppd
2.2.0. (This point is very hard to be found in these Kernel documents). So I
also upgrade my pppd to 2.3.6 by following the instructions.
The nightmare is: I use the same scripts, same configuration and same dialup
number. This time the combination of kernel and pppd doesn't work. When I
try to call "pppd", the error message is:
"/usr/sbin/pppd: peer authentication required, but no suitable secret(s)
found
/usr/sbin/pppd: for authenticating any peer to us.(abc)"
"abc" is my logname to my ISP.
Why this happen? I do have a pap-secret file, and everything works just well
in the earlier version.
[EMAIL PROTECTED]
------------------------------
From: "David Syratt" <[EMAIL PROTECTED]>
Subject: Gateway question.
Date: Thu, 10 Jun 1999 16:16:52 +1000
Have RH5.2 with 2 Net cards.
Using PPP on dialup to Inet.
I want the the default route for the PPP connection to use eth1 as I have
the internal net using eth0.
I have been working with the routing table but have been unable to get both
eth1 and PPP working at the same time.
Can you point me in the right direction please.
Thanks David
------------------------------
From: kite@NoSpam.%inetport.com (Clifford Kite)
Subject: Re: slow ppp connection on External ISDN modem
Date: 9 Jun 1999 14:10:32 -0500
Darren Durbin ([EMAIL PROTECTED]) wrote:
: However, I think ( and this is just a guess ) that the problem may be
: that it only talks sync PPP, which older version of pppd didn't. I
: believe the new versions have an option for this so I'm going to try it
: when I get some free time...
AFAIK the ordinary PC serial port hardware (16550A UART) doesn't support
syncronous. If so, then you're not going to succeed without something
to translate from asyncronous to syncronous.
--
Clifford Kite <kite@inet%port.com> Not a guru. (tm)
/* The signal-to-noise ratio is too low in many [news] groups to make
* them good candidates for archiving.
* --- Mike Moraes, Answers to FAQs about Usenet */
------------------------------
From: Nicholas E Couchman <[EMAIL PROTECTED]>
Subject: Re: Slow telnet and ftp responses
Date: Thu, 10 Jun 1999 04:12:03 GMT
Make sure that the computer you are trying to connect from is registered w/
Linux. This means that it is present in the /etc/hosts file. If it isn't,
run linuxconf. Go to network -> misc ->info. about other hosts. Add the
computer here. You should see better results.
--Nick
J Zhang wrote:
> I have a AMD k6-2 chip powered machine on a LAN. The machines has a
> Netgear FX310 card. I have been running Redhat 5.2 for a long time,
> everything has worked perfect. However, when I try telnet and ftp the
> linux host from other machines in the LAN, it takes more than one minute
> to get a response.
>
> Apache server on the same linux machine works great. When I check the
> inetd.conf, I found that that the default setting is to let inetd start
> tcpd, then tcpd check the access control files.
>
> I am wondering how can I get the responses faster.
>
> Thanks
>
> Jeff
------------------------------
From: "Gero H. Marten" <[EMAIL PROTECTED]>
Subject: Re: INN and a list of newsgroups
Date: Thu, 10 Jun 1999 07:04:21 +0200
It should be under /usr/lib/news/bin
--
Gero H. Marten
<http://www.provi.de/gmarten/index.html>
--
------------------------------
From: "Michael Faurot" <[EMAIL PROTECTED]>
Subject: Re: Linux Router and xDSL
Date: 10 Jun 1999 04:06:35 GMT
Gabe O'Brien <[EMAIL PROTECTED]> wrote:
: Here's my problem...
: I've just gotten a DSL connection at home, and I want to use Linux on my old
: PC at home to run a router and set up a LAN. My provider (Bell Sympatico in
: Ottawa) uses DHCP to assign IP addresses to the machines on the line, so the
: NIC attached to the Internet will need to use DHCP, but I'm planning on
: using static addresses internally for simplicity's sake. Ideally I'd like
: to take advantage of the extra NICs I have laying around and avoid buying a
: hub by putting several NICs in the Linux box and having a direct cross-over
: line to each of the PCs on the LAN. Is this a good idea?
Depending upon how many NICs you're really talking about, you'll likely
be giving yourself a headache trying to resolve the IRQ, DMA and IO port
conflicts if you're talking about more than three NICs. Hubs aren't
that expensive. On a recent trip to San Jose, I picked up a 5 port hub
for $25 at Fry's. I'm sure you can find similar deals from either local
or online vendors.
--
==============================================================================
Michael | mfaurot | An avocado-tone refrigerator would look good on
Faurot | atww.org | your resume.
------------------------------
From: [EMAIL PROTECTED] (Matthias Meixner)
Crossposted-To: comp.protocols.nfs
Subject: Re: user-space NFSd on Linux 2.2.x?
Date: 9 Jun 1999 07:26:21 GMT
Billy Nadeau ([EMAIL PROTECTED]) wrote:
> Hi; Trying to avoid Linux-AIX nfs incompatibilities, I tried recompiling
> a linux 2.2.x kernel without NFS server support to use "only" the nfsd
> user-space app, but then mountd fails with an unsupported function
> call. Is there a way on 2.2.x linux kernels to downgrade to user-space
> nfs server?
You also have to install the user-space NFS server and start it instead of
the kernel NFS server (rpc.nfsd instead if rpc.knfsd and rpc.mountd instead
of rpc.kmountd).
We had it running here with 2.2.7 + Suse 6.0, but it was really unstable
(probably RPC bugs in glibc2) so we gave it up in the end.
(Kernel NFS also had some bugs in serving Solaris machines, so it was also
no option.)
- Matthias Meixner
--
Matthias Meixner [EMAIL PROTECTED]
Technische Universit�t Darmstadt
Rechnerbetriebsgruppe Telefon (+49) 6151 16 6670
Wilhelminenstra�e 7, D-64283 Darmstadt, Germany Fax (+49) 6151 16 4701
------------------------------
** 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.networking) 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-Networking Digest
******************************
