Linux-Networking Digest #621, Volume #11         Tue, 22 Jun 99 11:13:51 EDT

Contents:
  Re: Why is linux perfomance bad compared to windows? ("Chris L")
  Re: httpd: cannot determine local host name ("Chris L")
  Re: ipchains/mac (sleb)
  How to??? - ADSL w/SuSE 6.1 (Spicoli)
  Setting up several ISP for PPP connections (R. Denoire)
  nfs server woes (again!) ([EMAIL PROTECTED])
  Re: mgetty + ppp server problem ([EMAIL PROTECTED])
  Re: Which is eth0, and which eth1? (Frank Pineau)
  Re: please help me get online (Zoltan Pittner)
  Re: Qmail question (Bob McLaren)
  Re: WIN95 -> Linux box on Serial port! ("Bob Glover")
  help: fetchmail not respecting "is *" (Sim)
  Re: sendmail and NFS... ("Stefan Monnier " 
<[EMAIL PROTECTED]>)
  Re: Mindcraft Retest News (Michael Blakeley)
  Re: How many virtual servers per machine? (Jonathan Guthrie)
  Re: in.ftpd : login failed (peter)
  Re: Could Microsoft Cheat On The New Mindcraft Benchmark? (was: Mindcraft Retest 
News (Philip Brown)
  httpd: cannot determine local host name (dkselich)
  Re: Using ftp in windows to connect to linux ([EMAIL PROTECTED])
  Speeding up PPP? (Thomas Kochak)
  Re: how do i diable remote telnet ftp and other listening/open ports? (Greg de 
Freitas)
  FTP on Linux Redhat ("A. de Vos")
  Re: Modem Sharing (Richard Corfield)

----------------------------------------------------------------------------

From: "Chris L" <[EMAIL PROTECTED]>
Subject: Re: Why is linux perfomance bad compared to windows?
Date: Tue, 22 Jun 1999 09:09:28 -0500

No comeback to the NT uptime <= 49 days that is...

Chris

Yuki Taga wrote in message <[EMAIL PROTECTED]>...
>On 21 Jun 1999 23:29:45 GMT, in article
<[EMAIL PROTECTED]>,
>[EMAIL PROTECTED] (Steve Lamb) wrote:
>
>>  In fact,
>>yesterday I spent 8 hours installing Win98 on my laptop.
>
>This is too funny.  I have no comeback.  You win.  <g>
>I just want to know one thing.  How much did you bill your client?  <vbg>
>
>Yuki ^_^



------------------------------

From: "Chris L" <[EMAIL PROTECTED]>
Subject: Re: httpd: cannot determine local host name
Date: Tue, 22 Jun 1999 08:29:07 -0500

Maybe add the computers name to the hosts file?

Chris
dkselich wrote in message <[EMAIL PROTECTED]>...
>I get this message at boot up ever since I changed my computers name in
>linuxconf
>
>Executing: //etc/rc.d/rc3.d/S85httpd start
>* httpd: cannot determine local host name.
>* Use the ServerName directive to set it manually
>> Starting httpd: httpd
>
>How do I fix this?
>
>Dennis



------------------------------

From: sleb <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.help
Subject: Re: ipchains/mac
Date: Tue, 22 Jun 1999 09:01:09 -0500



Dave Hamilton wrote:

> [[ This message was both posted and mailed: see
>    the "To," "Cc," and "Newsgroups" headers for details. ]]
>
> In article <[EMAIL PROTECTED]>, root <[EMAIL PROTECTED]> wrote:
>
> > i have a mac w/ system 7.5.1.
>
> Upgrade the Mac to System 7.5.3 (at least).  It makes MAJOR changes to
> networking.
>
> > i did try different settings on the mac machine in mactcp, and in the
> > chooser; i tried w/ appletalk on and off.  in mactcp, i tried the
> > ethernet and ethertalk settings, no luck.
> >
> > anyway, if this makes any sense to someone, i sure would appreciate some
> > enlightenment.
> > thanks in advance,
>
> Sure... let's see -- Appletalk makes no difference here, since we're
> dealing with TCP/IP traffic.  You want to manually set your IP address
> (in either MacTCP or the TCP/IP control panels), and then set your
> gateway address to the internal ip address of your linux machine, your
> subnet mask to the same as you have your win98 machines, and dns the
> same as well.  This is much simpler (and more reliable) with Open
> Transport, which was introduced in System 7.5.3.  The update is free
> and available from http://www.apple.com.
>
> Hope this helps!
>
> -Dave Hamilton
> The Mac Observer, http://www.macobserver.com/
> [EMAIL PROTECTED]

well,
i'm still stumped.  i installed 7.5.3 and set up tcp/ip in control panels;
when i try to ping the name or ip of the mac machine from the linux box, ping
will just hang.  i can ping everything else.
the mac i have is a performa 631 cd.  i bought an older nic card w/ rj 45.
the nic card is an LC slot card.
i was told that the system software will take care of the nic.  i was
wondering if there might be specific drivers for the card that i could load on
the mac, that way making sure the software for the card is installed.
help
[EMAIL PROTECTED]


------------------------------

From: [EMAIL PROTECTED] (Spicoli)
Subject: How to??? - ADSL w/SuSE 6.1
Date: Tue, 22 Jun 1999 13:31:44 GMT

Hey all,

I have ADSL, which works in my other OS, and I can get the Ethernet
card (it is an Intel Ether Express Pro 100) to be recognized by my
Linux system but I can't find any documentation in the SuSE manual on
getting/installing/configuring TCP/IP for the connection.  Without a
connection to the 'net the system is fun to play with but...

Anybody know where I can find the information on how to get this
bugger working?   

Thanks in advance.

Spicoli

spicoli at aracnet dot com

------------------------------

From: [EMAIL PROTECTED] (R. Denoire)
Subject: Setting up several ISP for PPP connections
Date: Tue, 22 Jun 1999 13:59:36 GMT
Reply-To: [EMAIL PROTECTED]

I am using SuSE 6.0 with updated Kernel 2.2.5 here.

My questions relates the problem that when setting up a ppp connection
to an Internet Service Provider, the IP address of its DNS server
should be entered in the file /etc/rc.config. So when connecting to
the ISP, usually the new connection over ppp is set as defaultroute
and all name resolutions use the IP addresses given in /etc/rc.config.

Now if I am able to connect to different ISPs I have to put the
corresponding IP address of their DNS-Server in this file
/etc/rc.config. But since only three addresses at most can be entered
there, I guess that I can not set up a ppp connection to more than 3
ISPs. Of course, one can edit /etc/rc.config before dialing, but this
is an inconvinience.

I bet that there is someway to achieve setting up more than 3 ISPs and
even overcome the necessity of putting their IP-addresses of their DNS
servers there, since using Windows, one can do this without taking
care of that and it works. (Bear in mind that even just one ISP has
possibly several DNS servers, because not all of them are on duty all
the time, and one would enter all this IP addresses in the
/etc/rc.config).

Does anyone know a way of overcoming this limitation?

If you are not using Suse, how does your distribution handle DNS
servers if your provider and in which configuration files and they
entered?
================


------------------------------

From: [EMAIL PROTECTED]
Subject: nfs server woes (again!)
Date: Tue, 22 Jun 1999 13:14:49 GMT

I have got redhat 6.0 running on an i686 [chekov] and want to get nfs
server working.
I have managed to mount an nfs drive on another machine from chekov,
but I cannot mount a drive from chekov externally.
Apache,  telnet and ftp all work (in both directions)

During bootup,  things seem ok - portmap loads,  pcnfs,  mountd ....

I have set up an exports file (I think correctly)

rpcinfo -p gives info on
rpcbind,  rquatad,  mountd,  pcnfsd,  rstatd.  nfs appears but with
udp and not tcp  (and a reference to portmap doesn't appear)

rpm -qa |grep nfs gives
pcnfsd-2.01sh-6shadow
knfsd-clients-1.2.2-4
knfsd-1.2.2-4


Would appreciate guidance



Eventually I wish the machine to run over tcp/ip and boot up
a set of network computers

Delia Wakelin


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.

------------------------------

From: [EMAIL PROTECTED]
Subject: Re: mgetty + ppp server problem
Date: Tue, 22 Jun 1999 13:21:11 GMT

This is an excerpt from my mgetty.ttyS2 log file:

06/22 14:57:25 yS2  waiting for ``RING''
06/22 14:57:25 yS2   got: [0d][0a]RING ** found **
06/22 14:57:25 yS2  send: ATA[0d]
06/22 14:57:25 yS2  waiting for ``CONNECT''
06/22 14:57:25 yS2   got: [0d]
06/22 14:57:25 yS2    CND: RING[0a]A[00]T&F&C1&D2[0d]
06/22 14:57:35 yS2    CND: A[0d][0a]OK[0d]
06/22 14:57:35 yS2    CND: OKATA[0d]
06/22 14:57:45 yS2    CND: ATA[0d][00][0a]CONNECT ** found **
06/22 14:57:55 yS2  send:
06/22 14:57:55 yS2  waiting for ``_''
06/22 14:57:55 yS2   got:  115200~T[ff]}#[c0]!}!}$}
}7}"~[00][ff]}#[c0]!}!}'}}4}"[16]T[01][00]/[f0][f0][03],[00][ff][ef][0a]
06/22 14:58:25 yS2    CND: CONNECT 115200~T_}#_!}!}$} }7}"~
06/22 14:58:25 yS2    CND: found: 115200~T_}#_!}!}$} }7}"~ ** found**
06/22 14:58:25 yS2   waiting for line to clear (VTIME), read:
[10]/[00][00]
06/22 14:58:25 yS2    looking for utmp entry... (my PID: 164)
06/22 14:58:26 yS2   tio_set_flow_control( HARD )
06/22 14:58:26 yS2   print welcome banner (/etc/issue)
06/22 14:58:26 yS2   getlogname (AUTO_PPP),
read:[16][00][01][00]/[f0][f0][03]
06/22 14:58:35 yS2   getlogname (AUTO_PPP), read:,[00][ff][ef][0a]
06/22 14:58:35 yS2   tio_get_rs232_lines: status: RTS CTS DSR DTR DCD
06/22 14:58:35 yS2    login: use login config file
/usr/local/etc/mgetty+sendfax/login.config
06/22 14:58:35 yS2   match: user='o', key=''
06/22 14:58:35 yS2   match: user='o', key=''
06/22 14:58:35 yS2   match: user='o', key='/AutoPPP/'
06/22 14:58:35 yS2   match: user='o', key='*'*** hit!
06/22 14:58:35 yS2   calling login: cmd='/bin/login', argv[]='login o'
06/22 14:58:35 ##### data dev=ttyS2, pid=164, caller='none',
conn='115200~Tÿ}#À!}!}$} }7}"~', name='', cmd='/bin/login',user='o'
06/22 14:58:35 yS2   setenv: 'CALLER_ID=none'
06/22 14:58:35 yS2   setenv: 'CONNECT=115200~T_}#_!}!}$} }7}"~'

After that the modem hangs up.


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.

------------------------------

From: [EMAIL PROTECTED] (Frank Pineau)
Subject: Re: Which is eth0, and which eth1?
Date: Tue, 22 Jun 1999 08:58:02 -0400
Reply-To: [EMAIL PROTECTED]

Count yourself lucky that you could even get the eepro to compile.  When i
ran any of the make config's for 2.2.10, the eepro driver selection is
always greyed out.  I'm too much of a linux newbie to figure out how to
compile it in by hand, so it never makes the eepro.o driver.

FP



On Fri, 18 Jun 1999 23:15:39 GMT, John Hovell <[EMAIL PROTECTED]> typed
this with two fingers:

>Why don't you just compile the drivers into the kernel?  I know you can control
>the order in which they are found through an append command in lilo.conf -- not
>to mention they will both be recognized -- you might want to use 2.2.5-22 -- as
>its specifically for your RedHat release.  You might want to see
>http://metalab.unc.edu/linux/HOWTO/Ethernet-HOWTO-10.html#ss10.2
>
>I think it suggests putting "alias eth0 tulip" and "alias eth1 other-driver".
>
>Also, Donald Becker's multi-card how-to:
>http://cesdis.gsfc.nasa.gov/linux/misc/multicard.html
>
>You can definitely get them both working if you want to.  I have one machine w/
>3 different types of ether cards in it.
>
>Write back if you need more specifics,
>
>Hope this helps,
>John
>
>PS... Kinda weird that when I respond to people in newsgroups all over the world
>you run into some people right near by.  (I live in Darien).
>
>jay nospam beatty wrote:
>
>> I just upgraded from 2.2.5-15 to 2.2.10. I have 2 PCI NICs: a eepro on the
>> m/b and a PCI DEC tulip.  My conf.modules has eth0 as tulip and eth1 as
>> eepro. No 'ether' line.
>>
>> When I boot 2.2.5-15 it finds the DEC tulip as eth0 - which is connected to
>> the cable modem.  I've not been able to get it to find eepro - connected to
>> my home lan - but that's OK.
>>
>> When I boot 2.2.10  *with the same conf.modules* it finds eth0 as eepro. It
>> doesn't find the DEC tulip. Then it assigns all to eepro the eth0 ip info -
>> which should bind to the DEC tulip card. None of the NIC drivers are built
>> in the kernel.
>>
>> How do I control which is eth0 and eth1 ( I thought it was conf.modules! ).
>> I've care less about getting both cards to work ( actually, I've given up ).
>> I just don't want ot have to switch cables each time I get a new kernel.
>>
>> Thanks.
>>
>> Jay


-- 
<*><*><*><*><*><*><*><*><*><*><*><*><*><*><*><*><*>
Frank Pineau  ------------>

        [EMAIL PROTECTED]
   http://bbs.industrynet.net/html/pineaus/

Use your powers for good, not for evil.  Well, you
can use them for evil if you want, but it voids
your warranty.
<*><*><*><*><*><*><*><*><*><*><*><*><*><*><*><*><*>
PGP Public Key:
http://bbs.industrynet.net/html/pineaus/fpineau.asc

No mail from web-based mail systems such as Hotmail
or Yahoo will be accepted.

------------------------------

From: Zoltan Pittner <[EMAIL PROTECTED]>
Subject: Re: please help me get online
Date: Tue, 22 Jun 1999 10:34:27 -0400

the easyest way from redhat 5.2 is through X. Fire up your X windows with
startx, in linuxconf configure your ppp/slip to a new ppp connection, fill in
the phone no, username, password, choose PAP authentification. Once you did
that, you should get out of X and reboot. Maybe there are other ways without
reboot,
init "something", just I do not know what is the something. After reboot, get
back to X and click on your menu -> networking -> Usernet and click on "ppp0"
if you configured this connection as ppp0 in Linuxconf. This has to start to
dial you ISP.Sure I suppouse, you have a modem installed, and it is not a
Winmodem...

Any more questions, E-mail me.
Zoltan.

anorexorcist wrote:

> if someone is using linux (specifically the redhat distribution) and
> concentric as there dial up isp (or for that matter any dial up isp) please
> contact me...i'm having difficulty getting online with linux.


------------------------------

From: Bob McLaren <[EMAIL PROTECTED]>
Subject: Re: Qmail question
Date: Mon, 21 Jun 1999 17:43:24 GMT

I found that if you have an established account for steve you won't be able
to create an alias for him in the /var/qmail/alias directory.
What I ended up doing is editing the ~steve/.qmail file.  This file should
have one line in it that will either say ./Maildir/ or ./mbox/ or something
to that effect.  All you have to do is remove that line so the messages no
longer go to those directories.  Then add a line specifying the email
address you want to forward to.

Example: Forwarding [EMAIL PROTECTED] to [EMAIL PROTECTED]
Edit ~steve/.qmail file to contain the following line
[EMAIL PROTECTED]

That should do the trick.  Good luck to you.

[EMAIL PROTECTED] wrote:

> Got a question about Qmail. I have two domains whose MX record points
> to the same server (modelprinting.com, webheadshots.com). Anyway, is it
> possible in Qmail to set up aliasing so that anything sent over to, say
> [EMAIL PROTECTED] automatically gets forwarded to
> [EMAIL PROTECTED]? At the same time, any mail sent over the
> [EMAIL PROTECTED] stays put and stays in that mailbox?
>
> I tried setting an alias file in /var/qmail/alias called .qmail-
> [EMAIL PROTECTED], but it doesn't seem to forward.
>
> What is the best tool to do what I want to do? Procmail?
>
> - Steve
>
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.

--
Bob McLaren
Network Administration
Financial Statement Services, Inc.
HTTP://WWW.FSSI-CA.COM



------------------------------

From: "Bob Glover" <app1rtg_at_air.ups.com>
Crossposted-To: comp.os.linux.misc
Subject: Re: WIN95 -> Linux box on Serial port!
Date: Tue, 22 Jun 1999 13:33:01 +0100

Windoughs expects a modem to be attached to the other end of the serial
port.  I looked into this myself before I broke down and bought a
network-in-a-box.  You have a couple of options open to you.

Option 1.
There is a protocol that Windoughs calls "Direct Cable Connection".  You
should find that in your Win menus.  This is not PPP and if you go this
route, you will need to re-compile support for that protocol into the Linux
kernel.

Option 2.
Find the configuration file for your modem in Windoughs.  I found it once
upon a time on my Wintendo, but I can't, for the life of me, remember where.
Try using "find" to look for a file containing your modem's name.  When you
find this file you may be able to tweak it so that it doesn't insist on
dialing, but then again you may not be able to.  Good luck, you'll need it.

Option 3.
On the Linux side, make a program or script that behaves like a modem would.
Easier said than done, but also easier than option 2.

Option 4.
This might seem obvious, but you could get a modem.

Ferdinand V. Mendoza wrote in message <[EMAIL PROTECTED]>...
>Folks,
>Today I got a Toshiba notebook to play with my
>Linux box. I tried to connect it to one of the
>unused COM ports and I configured mgetty to
>test a terminal log-in. It works.
>Now, I want to use PPP over the same serial port
>(no modem of course) and plan to do some
>benchmarks. I can run a script to fire up PPP
>once the login is successful. My problem now is
>how do I configure the Windows
>side to do a null modem networking.
>Someone can help, please.
>TIA.
>
>
>Ferdinand
>
>



------------------------------

From: [EMAIL PROTECTED] (Sim)
Subject: help: fetchmail not respecting "is *"
Date: Mon, 21 Jun 1999 20:04:09 GMT

My .fetchmail rc reads 

user "sdkdgsdwrhj" with password "dwxjfsdjf" is * options keep

For some reason fetchmail is passing everything over as
postmaster@localhost whereas is * is supposed to prevent this and pass
whatever is before the @.

In fact the mails are addressed either simon.hampton or my 2 aliases.
What could be wrong.  I'm using version 5.0.4

Simon

------------------------------

From: "Stefan Monnier <[EMAIL PROTECTED]>" 
<[EMAIL PROTECTED]>
Subject: Re: sendmail and NFS...
Date: 21 Jun 1999 16:16:09 -0400

>>>>> "srix" == srix  <[EMAIL PROTECTED]> writes:
> I've set up a sendmail server to use NFS mounts for /var/spool/mqueue
> and /var/spool/mail. It works, but if I try to send 100+ mails in quick

I don't know how to solve your specific problem, but I'd recommend just
not to do that.  For one, I really can't see why you'd want the mqueue
to be accessed via NFS.  And for the spool, it's better to export it via NFS
from the mail-server than to import it to the mail-server.
Especially since you can then easily switch to another access method like POP
or IMAP which doesn't suffer from the same problems that plague NFS
(mostly locking, which is either absent (leading to corrupted mailboxes
or lost email) or slow and buggy in most other cases).


        Stefan

------------------------------

Date: Mon, 21 Jun 1999 13:29:27 -0700
From: [EMAIL PROTECTED] (Michael Blakeley)
Crossposted-To: comp.infosystems.www.servers.unix
Subject: Re: Mindcraft Retest News

In article <[EMAIL PROTECTED]>, Mark Evans
<[EMAIL PROTECTED]> wrote:

>R. Denoire <[EMAIL PROTECTED]> wrote:
>
>> -If the server used has to cope with more than one NIC, NT is better
>> here; supposedly Linux developers admit this weakness in current
>> kernel versions and promised to work on this issue.
>
>Is this refering to a single or multiple processor machine?
>
>Note that WRT to the original Mindcraft test rather specific "fiddles"
>were used. Which rely on the the number of NIC's being the same as
>the number of processors.  Also, IIRC, the required functionality
>in NT is new and possibly not even available to any real customer.

Isn't this a red herring in any case? How many web servers are flooding
their 100BaseT NICs? And if they are, why not work on 1000BaseT?

Multiple NICs were important with 10BaseT LANs, but most of the
interesting work nowadays is in the app server space. In that space, CPU
and disk are much more likely to be gating factors than NICs are.

--
Michael Blakeley       [EMAIL PROTECTED]     <http://www.blakeley.com/>
            Performance Analysis for Internet Technologies

------------------------------

From: Jonathan Guthrie <[EMAIL PROTECTED]>
Subject: Re: How many virtual servers per machine?
Date: 21 Jun 1999 21:05:12 GMT

Adam Long <[EMAIL PROTECTED]> wrote:
> My qeustion is how many virtual servers would you recommend I should run on
> each machine at a max.

Till the machines bog down.  I've got about 3 dozen on my Cyrix 5x86-120
here.
-- 
Jonathan Guthrie ([EMAIL PROTECTED])
Brokersys  +281-895-8101   http://www.brokersys.com/
12703 Veterans Memorial #106, Houston, TX  77014, USA

------------------------------

From: [EMAIL PROTECTED] (peter)
Subject: Re: in.ftpd : login failed
Date: Mon, 21 Jun 1999 21:24:17 GMT

In article <[EMAIL PROTECTED]>, [EMAIL PROTECTED] 
says...
> Not to ask a stupid question but is the user your logging in part of
> group FTP? Defined in the /home/ftp/etc/passwd file? You need to read
> your ftp man's and make sure your user has access. It sounds like you
> missed something along the way. Users do not have to be in ftp passwd
> file as long as they are defined as part of ftp group and not
> non-allowed via ftpaccess.
> HTH
> David Previti
> 

user is member of the ftp-group and I dont find any information about 
/home/ftp/etc/passwd - file. this file even dont exist in wu-ftpd-
2.4.2b18-2
just to make sure I created it and add the user to the passwd-file. no 
success. 



peter

=================
pilsl@
ANTISPAM
goldfisch.atat.at

------------------------------

From: [EMAIL PROTECTED] (Philip Brown)
Crossposted-To: 
omp.os.ms-windows.nt.advocacy,comp.os.linux.advocacy,comp.infosystems.www.servers.unix,comp.os.linux.misc
Subject: Re: Could Microsoft Cheat On The New Mindcraft Benchmark? (was: Mindcraft 
Retest News
Reply-To: [EMAIL PROTECTED]
Date: 21 Jun 1999 21:28:45 GMT

On Fri, 18 Jun 1999 01:25:34 GMT, [EMAIL PROTECTED] wrote:
>On 17 Jun 1999 20:58:38 GMT in comp.os.linux.networking,
>[EMAIL PROTECTED] (Philip Brown) uttered the following
>profound gem of wisdom:
>
>...
>>So if there is something that is highly [non-FP] CPU-based, an intel+solaris
>>box is still more cost effective than a sparc box.
>
>Really depends upon what you want it for.

You just uttered the standard "sun-vs-intel" retort, in knee-jerk fashion,
without reading what I wrote.

I specifically said CPU constrained tasks. and disqualified floating point,
which sensible people know sparc has an edge on.

If you have something that is CPU based, it will pretty much sit in memory,
and stay in L2 cache. So, reliability of disks, etc, etc, etc do not come into
play. In which case, a 500mhz intel box is faster than an ultra, and just
as reliable.

Still assuming the constraints above, a simple 1cpu vs 1cpu box, will have
intel way in front, for price vs cpu I'm not sure how well the scaling will
affect things, but a four=way 500mhz pentium box will be a lot cheaper than a
4-way ultra box, and equivalent or better integer performance.

some price points:

http://www.gateway.com/prod/sb_e5250550xl_prodinfo.shtml

a 2-way Xeon 550 mhz computer, for $6499, NO DISCOUNT

a 4-way Xeon 550mhz computer, $10,800  [512k L2 cache]
a 4-way Xeon 550mhz computer, $17,700  [1-meg L2 cache]
a 4-way Xeon 550mhz computer, $28,400  [2-meg L2 cache]



compare to sun's cheapest 4-way box, an E450:

4-way 250mhz E450, $20,900 list price [1-meg L2 cache]
4-way 300mhz E450, $30,000 list price [2-meg L2 cache]
4-way 300mhz E450, $36,500 list price [4-meg L2 cache]

The 250mhz isn't particularly competative when it comes to CPU power.


I'd like to see some integer comparisons between the best xeon box, and the
best E450 box. But good luck trying to get a hold of 4x400mhz ultra-cpus right
now.


I HATE this. I much prefer sun boxes. But you'll certainly pay the premium for
them.


-- 
[Trim the no-bots from my address to reply to me by email!]
[ Do NOT email-CC me on posts. Pick one or the other.]
 --------------------------------------------------
The word of the day is mispergitude


------------------------------

From: dkselich <[EMAIL PROTECTED]>
Subject: httpd: cannot determine local host name
Date: Tue, 22 Jun 1999 06:21:01 -0700

I get this message at boot up ever since I changed my computers name in
linuxconf

Executing: //etc/rc.d/rc3.d/S85httpd start
* httpd: cannot determine local host name.
* Use the ServerName directive to set it manually
> Starting httpd: httpd

How do I fix this?

Dennis

------------------------------

From: [EMAIL PROTECTED]
Subject: Re: Using ftp in windows to connect to linux
Date: Mon, 21 Jun 1999 20:14:37 GMT

How are you ftp'ing?
Have you tried opening an MS-Dos prompt and
ftp'ing at the Windows prompt to the ip address of
one of the linux boxes (c:\windows>ftp #.#.#.#).
This works for me. I found I had to switch the IP
address from dynamic to static on the win98 box
and keep it in the same domain and subnet mask as
the linux boxes. I don't know how this will effect
your gateway though (which is something I will be
attempting to set up here soon.)

Now, maybe you can help me with something.
I am trying to ftp TO the win98 box from the linux
boxes. Any suggestions? I had it working fine on
'95 w/ onnet's '93 version of ftpsrv. '98 doesn't
seem to offer any type of mode of ftp to listen or
act as host and of coarse the '93 dos version
doesn't run on '98. Ideally I just skip the
ftp'ing at a dos-prompt and have them networked in
a real fashion. Any suggestions there? What are
you using for your gateway? And how do you
transfer from the linux boxes to the win98 box?

Hope MY advice was helpful to you. I listed a
little more help below.



In article <7kljoh$2q2s9$[EMAIL PROTECTED]>,
  "bv" <[EMAIL PROTECTED]> wrote:
> I have 2 linux boxes and a win98 box. But when i
try to connect with ftp
> from windows to linux, windows comes up with a
dial-up-window. When I cancel
> i can't connect but at the moment i have a
connection to internet the ftp
> connects (the same for telnet).



> This sounds like the program you are using for
  ftp has an address conflict w/ the dial-up and
  the dial-up apparently over-rides since it's
  98 driven. Or maybe you don't have a protocol
  set up for it. If you go into network properties
  there should be 2 tcp/ip protocols set up. One
  for your dial-up adapter and one for your lan
  card. If there are 2 then try changing the one
  for the lan card to be static by going into
  properties.




> I also have a gateway at the windows-box. Both
linux computers uses it. Do i
> have to change anything in the route.conf?
Becaus the gateway must only be
> used when i try to connect a internet-ip from
linux. I also don't have a
> dns-server for my LAN. Is this nessecary for
good networking, and is it hard
> to set up?
>




Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.

------------------------------

From: Thomas Kochak <[EMAIL PROTECTED]>
Subject: Speeding up PPP?
Date: Mon, 21 Jun 1999 16:54:02 -0500

I just got a used V2 to go with my TNT card. I've always played Quake 3
Test in windows and got a ping around 165-195. When I played it with my
V2 in linux I got a ping around 300-350!! I am using a Diamond
SupraExpress 56e (external). What position does the mtu & mru values in
the /etc/ppp/options file do? I believe they were both set at 550 or
something like that. If I increase/decrease them will that speed it up
at all? And is there any other way to speed it up?


------------------------------

From: Greg de Freitas <[EMAIL PROTECTED]>
Subject: Re: how do i diable remote telnet ftp and other listening/open ports?
Date: Mon, 21 Jun 1999 22:19:25 +0100
Reply-To: [EMAIL PROTECTED]

Richard wrote:
> 
> what command can i use to dissable the ports on my system connected to
> my lan here at work so that i can lock down security?
1) edit /etc/inetd.conf
        add a '#' at the beginning of the line for each service you don't need
2) kill -HUP <pid>
        ,where <pid> is that of inetd process, found as follows:

        # ps -ef | grep inetd
:-)

--
Ciao 4 now, Greg.
# Email     :  mailto:[EMAIL PROTECTED]   #
# Email     :  mailto:[EMAIL PROTECTED]    #
#  To Live, To Love, To Learn, To Leave A Legacy.    #



------------------------------

From: "A. de Vos" <[EMAIL PROTECTED]>
Subject: FTP on Linux Redhat
Date: Tue, 22 Jun 1999 15:26:27 +0200

I cannot FTP from Windows to Linux through the LAN (ethernet). But I am able
to connect to the Apache server through Explorer.

How is FTP server started on Linux? And how can I reach it?

Arne de Vos




------------------------------

From: Richard Corfield <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup,tw.bbs.comp.linux,uk.comp.os.linux
Subject: Re: Modem Sharing
Date: 22 Jun 1999 15:21:29 +0100

[EMAIL PROTECTED] (Gilford Wimbley) writes:

> Yes.  I did this for a while.  The hassle is how do tell the linux
> machine to connect to the internet?  If your NT users don't mind
> telnetting in to the linux server to bring up the connection, then
> it's pretty painless.  If you want it to be all automatic, (that is,
> so the linux machine dials on demand and hangs up after a lull) this
> requires a program called diald.  Diald can be a bit troublesome to
> set up, but I succeeded so it can't be that hard ;-)
> 

There's another program called (IIRC) MasqDialier which has a Windows
client allowing your users to bring the link up on demand.

 - Richard.

-- 
   _/_/_/  _/_/_/  _/_/_/ Richard Corfield <[EMAIL PROTECTED]>
  _/  _/    _/    _/      Web Page:       http://www.littondale.freeserve.co.uk
 _/_/      _/    _/       Dance (Ballroom, RnR), Hiking, SJA, Linux, ... [ENfP]
_/  _/  _/_/    _/_/_/    PGP2.6 Key ID: 0x0FB084B1     PGP5 Key ID: 0xFA139DA7

------------------------------


** 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
******************************

Reply via email to