Re: [SLUG] How big is an http request?

[Peter Vogel]

I am trying to figure out why my server receives about 20mb data a day,
even when there's no-one here browsing the web etc.  I can imagine a meg
or two of emails, but the rest I can't account for.  Looking at my
Apache logs I see about 5,000 http requests a day for urls that are not
on my server, mainly porn, gambling, advertsing etc. These return code 404
as expected. I was wondering if these failed reequests could account for
10-20mbytes a day.

On Tue, 4 Feb 2003 19:02:54 +1100
Andrew Bennetts <[EMAIL PROTECTED]> wrote:

> On Tue, Feb 04, 2003 at 06:20:23PM +1100, Robert Collins wrote:
> > On Tue, 2003-02-04 at 17:15, Peter Vogel wrote:
> > > Roughly how many byes of data would I expect to receive for each http
> > > request to my server?
> > 
> > Thats very dependant on the url's you publish.
> > You should expect a minimum of:
> > 10 bytes + the URL to retrieve.
> > 
> > But, will likely see much more:
> > Cookie headers,
> > accept and accept-language headers,
> > host headers.
> 
> As a rough guess completely off the top off my head, I'd guess roughly 500
> bytes per request, considering the headers and whatnot your average HTTP
> clients (i.e. Moz and IE) send.
> 
> "It depends".  How accurately do you need to know? :)
> 
> -Andrew.
> 
> -- 
> SLUG - Sydney Linux User's Group - http://slug.org.au/
> More Info: http://lists.slug.org.au/listinfo/slug


-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] How big is an http request?

[Andrew Bennetts]

On Tue, Feb 04, 2003 at 06:20:23PM +1100, Robert Collins wrote:
> On Tue, 2003-02-04 at 17:15, Peter Vogel wrote:
> > Roughly how many byes of data would I expect to receive for each http
> > request to my server?
> 
> Thats very dependant on the url's you publish.
> You should expect a minimum of:
> 10 bytes + the URL to retrieve.
> 
> But, will likely see much more:
> Cookie headers,
> accept and accept-language headers,
> host headers.

As a rough guess completely off the top off my head, I'd guess roughly 500
bytes per request, considering the headers and whatnot your average HTTP
clients (i.e. Moz and IE) send.

"It depends".  How accurately do you need to know? :)

-Andrew.

-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] disk catastrophe

[David]

thanks. that worked a treat

for those of us not blessed by the almighty.. where should I have looked
to find out information like that? For instance.. .what does "1 4 1 7"
represent?

 On Mon, 3 Feb 2003, Guy Ellis wrote:

> try...
>
> echo >/proc/sys/kernel/printk "1 4 1 7"
>
> At 05:35 pm 03/02/2003 +1100, you wrote:
>
> >
> >How do i redirect the error messages away from the console i'm on.. is it
> >possible?
> >

-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] How big is an http request?

[Robert Collins]

On Tue, 2003-02-04 at 17:15, Peter Vogel wrote:
> Roughly how many byes of data would I expect to receive for each http
> request to my server?

Thats very dependant on the url's you publish.
You should expect a minimum of:
10 bytes + the URL to retrieve.

But, will likely see much more:
Cookie headers,
accept and accept-language headers,
host headers.

See RFC 2616 for the HTTP specification.

On top of that, there will be the TCP overhead...

Why do you ask?

Rob
-- 
GPG key available at: .



signature.asc
Description: This is a digitally signed message part

[SLUG] How big is an http request?

[Peter Vogel]

Roughly how many byes of data would I expect to receive for each http
request to my server?

-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

[SLUG] ---- Virus Detected ----

[noreply]

Dragnet Cleanmail has stopped the following message:

   Message: B000c667e9.0001.mml
   From:[EMAIL PROTECTED]
   To:  [EMAIL PROTECTED]
   Subject: Path Info

Because it believes an attachment to this message contains a virus.
The virus scanning software used was: Vet NT (Ver 10.x)2 

Please clean the file and resend it.
http://www.dragnet.com.au

Re: [SLUG] email server

[Christopher Samuel]

-BEGIN PGP SIGNED MESSAGE-

On Tuesday 04 Feb 2003 12:58 pm, Jeff wrote:

> I have done a netstat -an | grep 25 . Attached is the output.

That shows sendmail listening quite happily to anyone who cares to speak to 
it.  Read my previous email about what I believe is the problem, i.e. the NAT 
box is not forwarding connections on port 25 to the server.

> By the way this is from the machine in question.

Red Dwarf fan by some chance ? :-)

- -- 
Chris SamuelWollongong, NSW

-BEGIN PGP SIGNATURE-
Version: GnuPG v1.0.7 (GNU/Linux)

iQEVAwUBPj8qKI1yjaOTJg85AQGlMwgAp3GbDVPf/5NrG5UJDnAVfmInKXLVxcww
AC4FG8C/e0ORb0qZ746m1O/+yTsTQ+2i5tPSDuRH+p/bW1w/e8ryIXrhLfllStbe
Lbc+W6n/E4MJFs1S+NtHomkJwYoMqoAjf5GNkFu0vpYyn/z1PjWfRPbexQElfKkr
b2iIWu8/BmqgW57iSThaDcdmyiIR3uBQWOg/GI5eCl2fCajsej4SA9KUV1EZqTR6
BH2K9LSAZjdAthI8W/2DMv5ZVVMDhrvRl4Ihv9IpneJioQYTfbPoWELb47oX6anb
6eMOOc6SpBRhr21dJ2Fl+pAa+zQCBrugADP7rlQYP67gasURm4fXBQ==
=FU5P
-END PGP SIGNATURE-

-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

[SLUG] email server

[Jeff]

I have done a netstat -an | grep 25 . Attached is the output.

By the way this is from the machine in question.

Thanks Jeff

tcp0  0 0.0.0.0:25  0.0.0.0:*   LISTEN  
unix  17 [ ] DGRAM1259   /dev/log
unix  3  [ ] STREAM CONNECTED 80625  
unix  3  [ ] STREAM CONNECTED 80025  /tmp/.font-unix/fs7100

Re: [SLUG] How to create an ISP

[Rob B]

At 13:16 4/02/2003, Phillipus Gunawan sent this up the stick:

Greeting SLUG,

I got an assignment, to create almost full documentation (implementation
and pricing) on how to create an ISP in Sydney.
Is there any available article I can lookup which is a real story?

Preferably is a small-medium ISP which using Linux/RedHat as their
servers.


IIRC, there was a document on http://tldp.org along these lines.  Try 
googling for it as well.

cheers,
Rob


--
Any priest or shaman must be presumed guilty until proved innocent.

This is random quote 272 of a collection of 1274

Distance from the centre of the brewing universe:
[15200.8 km (8207.8 mi), 262.8 deg](Apparent) Rennerian

Public Key fingerprint = 6219 33BD A37B 368D 29F5  19FB 945D C4D7 1F66 D9C5

--
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

[SLUG] How to create an ISP

[Phillipus Gunawan]

Greeting SLUG,

I got an assignment, to create almost full documentation (implementation
and pricing) on how to create an ISP in Sydney.
Is there any available article I can lookup which is a real story?

Preferably is a small-medium ISP which using Linux/RedHat as their
servers.

cheers,

Phillip

-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] Paging services

[John Ferlito]

On Tue, Feb 04, 2003 at 12:51:23PM +1100, Brad Thomson wrote:
> 
> Hi All,
> 
> I'm looking for recommendations on a paging service to hook into netsaint
> for alerts.
> What experiences have people had with business offering paging services?
> I also require network coverage out in the sticks, Central West NSW to be
> precise.

Not sure about the coverage in central west but we use Orange for our
paging and have had few problems. I'm fairly sure they have coverage
maps on their website. They have a modem TAP interface as well as a
internet based TAP interface. I have a patgch for sms_client to do TAP
over TCP/IP if you end up going with them.


-- 
John
http://www.inodes.org/
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

[SLUG] Paging services

[Brad Thomson]

Hi All,

I'm looking for recommendations on a paging service to hook into netsaint
for alerts.

What experiences have people had with business offering paging services?

I also require network coverage out in the sticks, Central West NSW to be
precise.


cheers,

Brad
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] Email problem...

[Christopher Samuel]

-BEGIN PGP SIGNED MESSAGE-

Quick abstract for Jeff:

1) Your NAT box is not forwarding port 25 connections to the mail server.
That's why you're not getting email.

2) All email for the domain should be being queued on 
evnsrv2.computernerds.com.au, your backup mailserver.

Now onto why that is so:

On Tuesday 04 Feb 2003 9:52 am, Peter Rundle wrote:

> You're right, it was late and I was having a poke (hence the emoticon
> in the original posting).

:-)

> Back to the problem,
>
> Jeff,
>
> The server (203.219.50.38) is accessible to the outside world for port
> 110 but not 25. Until this is fixed there is no hope of getting mail
> inbound to it.

Correct. The question then is where is it being blocked.

I believe that it is the NAT box not forwarding on SMTP connections to the 
mail server correctly.  Here's why...

Using tcptraceroute to the server for port 110 shows:

Tracing the path to 203.219.50.38 on TCP port 110, 30 hops max
 1  dialup.pacific.net.au (203.9.190.192)  167.371 ms  118.381 ms  102.842 ms
 2  Ethernet0.wol001.pacific.net.au (61.8.31.4)  112.793 ms  106.762 ms  
104.911 ms
 3  lns-1.pacific.net.au (203.9.190.190)  145.711 ms  192.781 ms  189.793 ms
 4  f0-0-100.syd004.pacific.net.au (210.23.140.244)  158.827 ms  189.537 ms  
155.813 ms
 5  ge-2-0-0-500.ar1.SYD1.gblx.net (203.192.130.81)  168.869 ms  178.804 ms  
158.613 ms
 6  TPG-Internet.ar1.SYD1.gblx.net (203.192.130.222)  174.769 ms  200.697 ms  
167.887 ms
 7  syd4-7206.tpgi.com.au (203.12.160.80)  187.863 ms  181.823 ms  206.817 ms
 8  * * *
 9  * * *
10  free-tpg-038.tpgi.com.au (203.219.50.38)  406.823 ms  405.758 ms  235.680 
ms
11  free-tpg-038.tpgi.com.au (203.219.50.38) [open]  222.988 ms  414.935 ms  
265.545 ms

Using tcptraceroute to the server for port 25 shows:

Tracing the path to 203.219.50.38 on TCP port 25, 30 hops max
 1  dialup.pacific.net.au (203.9.190.192)  107.497 ms  111.405 ms  103.809 ms
 2  Ethernet0.wol001.pacific.net.au (61.8.31.4)  105.829 ms  103.746 ms  
116.931 ms
 3  lns-1.pacific.net.au (203.9.190.190)  204.764 ms  213.834 ms  205.727 ms
 4  f0-0-100.syd004.pacific.net.au (210.23.140.244)  166.904 ms  159.575 ms  
178.625 ms
 5  ge-2-0-0-500.ar1.SYD1.gblx.net (203.192.130.81)  183.867 ms  261.649 ms  
170.784 ms
 6  TPG-Internet.ar1.SYD1.gblx.net (203.192.130.222)  200.845 ms  151.841 ms  
154.756 ms
 7  syd4-7206.tpgi.com.au (203.12.160.80)  161.791 ms  190.722 ms  169.843 ms
 8  * * *
 9  * * *
10  free-tpg-038.tpgi.com.au (203.219.50.38) [closed]  370.107 ms  337.598 ms  
401.727 ms

Now, notice that there are 10 hops to find that the SMTP port is closed, but 
11 to find that the POP3 port is open.  This indicates that the system doing 
NAT for the server is forwarding port 110, but not forwarding port 25. Thus I 
suspect it is the NAT box returning the RST (rather than an "administratively 
prohibited" ICMP message you may expect from a firewall, or just disappearing 
into a black hole), not the email server.

> However a reverse lookup of that server returns
>
>38.50.219.203.in-addr.arpa  name = free-tpg-038.tpgi.com.au.
>
> And a subsequent MX search of the domain tpgi.com.au returns three
> servers each with 3 addresses overlapping each other but none of
> which are the above server.

tpgi.com.au is the ISP that he is connected to.

Look at the POP3 banner - it says:

+OK POP3 lister.crdc.com.au v2001.78rh server ready

Thus his domain is likely to be crdc.com.au - now using dig for that domain 
shows:

;; QUESTION SECTION:
;crdc.com.au.   IN  MX

;; ANSWER SECTION:
crdc.com.au.2303IN  MX  10 mail.crdc.com.au.
crdc.com.au.2303IN  MX  20 
evnsrv2.computernerds.com.au.

;; AUTHORITY SECTION:
crdc.com.au.2303IN  NS  ns0.computernerds.com.au.
crdc.com.au.2303IN  NS  ns1.computernerds.com.au.

;; ADDITIONAL SECTION:
mail.crdc.com.au.   2303IN  A   203.219.50.38
ns0.computernerds.com.au. 1746  IN  A   203.41.103.170
ns1.computernerds.com.au. 1746  IN  A   203.41.103.171

Notice that the A record for the first MX (mail.crdc.com.au) points to the IP 
address we've been wondering about.

Secondly, this shows that all his mail should be being queued on his secondary 
MX, evnsrv2.computernerds.com.au, where it will probably sit until the NAT is 
fixed or it decides to give up and return the mail.

- -- 
Chris SamuelWollongong, NSW

-BEGIN PGP SIGNATURE-
Version: GnuPG v1.0.7 (GNU/Linux)

iQEVAwUBPj8Dy41yjaOTJg85AQEYaAgAhG8qPWHSwUqXn5GjgXVru0Eg2xhZBPEP
QHkl0gqBkP7g77/AkgxkU7aKdbF4DEThLmOB58N4ry1jGg8K9PlCv2Wwrolt93TM
mME/WAzx0cKD8AdSLPdvJBwaz92Eeo9F2VaxBjy6XCpqS+9JTP6E4hsv4YV3EyLY
GIo0kJZPlh39+Wku8ZQQ0IQ3ibOgT9Qxo5nXX3XKsGBc2HkWeOGLNRTh0nDHlXiu
N9E2ESHHE0w94qIHkQ/otyReKtIZjTcxxom59KfPsw7W4CIn22PHbU5unpopeC0r
SnPyLiU0enBMT5WgK758uN8rMDbFZNAN8JYaj/N/h0wKtdxRgQ/MeQ==
=DYAR
-END PGP SIGNATURE-

-- 
SLUG - S

Re: [SLUG] RH8 installation probs

[Terry Collins]

mark davis wrote:
snip
> I dont understand if
> you buy a original copy of red hat which i have done,
> it should atleast install properly :(

If you brought an original copy, then exercise the support that you paid
for. Don't pay out good money and waste it.

Of course, if what you really have is a copy of a boxed CDrom, then I
would hazard a guess that whomever copied the CDroms for you actually
copied them on a non-linux system and some of the filenames caused the
copy program to barf and they were changed by the person copying without
realising their implications.

I've had similar problem ins CDs for RH7.3 and RH8.0. Don't ever
remember seeing it before. Alternatively, could be cheap 700Mb CD's etc.
(Okay, I'm just fudding cheapies from K-Mart, Strathfield Car Radios,
etc {:-).


-- 
   Terry Collins {:-)}}} email: terryc at woa.com.au  www:
http://www.woa.com.au  
   Wombat Outdoor Adventures 

 "People without trees are like fish without clean water"
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] Calculating physical memory from /proc/meminfo

[John Clarke]

On Tue, Feb 04, 2003 at 10:08:03AM +1100, [EMAIL PROTECTED] wrote:

> I've also found 256K missing...
> 
> Wondering about that now... but I'm not sure if the answer is really that 
> important or not... :)

Probably not.  511.75MB is close enough to 512MB for me.  This is even
closer:

[root@dropbear ~]# ls -l /proc/kcore 
-r1 root root 536612864 Feb  4 10:04 /proc/kcore

Only out by 252kB, and it doesn't rely on log files not being rotated
away :-)


Cheers,

John
-- 
whois [EMAIL PROTECTED]
GPG key id: 0xD59C360F
http://kirriwa.net/john/
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] Dial on demand

[Phil Scarratt]

There is an option (demand I think) for ppp to dial on demand - just add it to
the pppd command. CHeck the man page. I used to do that before I went to adsl.

Fil



Quoting Clint Gaddes <[EMAIL PROTECTED]>:

> 
> 
> Hi Guys,
> 
> I am looking for a simple way to configure the dial on demand for squid.
> 
> I have configured red hat 7.3 and squid. Manually configured ppp to dial
> local ISP.
> 
> THe problem I have run across is the ability for squid to be able to
> dial on demand. I have looked at Diald, but I was wounding if there was
> a better \ easier way to achieve demand on dial.
> 
> You would think there would be a better way than Diald?
> 
> Thanks
> 
> -- 
> SLUG - Sydney Linux User's Group - http://slug.org.au/
> More Info: http://lists.slug.org.au/listinfo/slug
> 


-
Phil Scarratt
It Consultant
0403 531 271

-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] Dial on demand

[Phil Scarratt]

There is an option (demand I think) for ppp to dial on demand - just add it to
the pppd command. CHeck the man page. I used to do that before I went to adsl.

Fil



Quoting Clint Gaddes <[EMAIL PROTECTED]>:

> 
> 
> Hi Guys,
> 
> I am looking for a simple way to configure the dial on demand for squid.
> 
> I have configured red hat 7.3 and squid. Manually configured ppp to dial
> local ISP.
> 
> THe problem I have run across is the ability for squid to be able to
> dial on demand. I have looked at Diald, but I was wounding if there was
> a better \ easier way to achieve demand on dial.
> 
> You would think there would be a better way than Diald?
> 
> Thanks
> 
> -- 
> SLUG - Sydney Linux User's Group - http://slug.org.au/
> More Info: http://lists.slug.org.au/listinfo/slug
> 


-- 
Phil Scarratt
IT Consultant
0403 531 271

-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] Calculating physical memory from /proc/meminfo

[mkraus]

Thanks heaps John...

Found it!

I've also found 256K missing...

Wondering about that now... but I'm not sure if the answer is really that 
important or not... :)

Thanks again...

Mike
---
Michael S. E. Kraus
Administration
Capital Holdings Group (NSW) Pty Ltd
[EMAIL PROTECTED]
phone (02) 9955 8000 fax (02) 9955 8144




John Clarke <[EMAIL PROTECTED]>
Sent by: [EMAIL PROTECTED]
04/02/2003 09:56 AM

 
To: [EMAIL PROTECTED]
cc: 
Subject:Re: [SLUG] Calculating physical memory from /proc/meminfo


On Tue, Feb 04, 2003 at 09:42:27AM +1100, [EMAIL PROTECTED] 
wrote:

> # grep kernel /var/log/dmesg
> Freeing unused kernel memory: 264k freed
> 
> As  you can see I don't get the same result. It seems that dmesg hasn't 
> stored information as far back as on the other machine.
> 
> Is there another place where this information would be stored?

If you haven't rotated the log out of existence, it'll be in one of
/var/log/messages*:

[root@dropbear ~]# grep -i memory /var/log/messages*
/var/log/messages.4:Jan  8 10:50:50 dropbear kernel: Memory: 
511988k/524032k available (1012k kernel code, 9480k reserved, 
844k data, 80k init, 0k highmem)
/var/log/messages.4:Jan  8 10:50:50 dropbear kernel: Freeing unused 
kernel memory: 80k freed

Note that 524032k is slightly (256k) less than the 512MB I have in this
machine.


Cheers,

John
-- 
whois [EMAIL PROTECTED]
GPG key id: 0xD59C360F
http://kirriwa.net/john/
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug



-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] Calculating physical memory from /proc/meminfo

[Peter Hardy]

On Tue, 4 Feb 2003 09:42:27 +1100, [EMAIL PROTECTED] wrote:
> Using `free` I encounter the same problems as when examining
> /proc/meminfo

> However, I searched the archives as you suggested and found that
> examining dmesg should show physical memory. dmesg had scrolled way
> past that point (and hence lost) on my machines so I did a `grep
> kernel /var/log/dmesg` to get:
*snip*
> Is there another place where this information would be stored?

Well, if you want to get *really* ugly about it, /dev/mem is an image of
physical memory (as opposed to /dev/kmem which is virtual memory), so:

root@marvin:~# dd if=/dev/mem of=blahblah
262144+0 records in
262144+0 records out
134217728 bytes transferred in 23.054214 seconds (5821831 bytes/sec)
root@marvin:~# ls -l blahblah
-rw-r--r--1 root peter134217728 Feb  4 09:49 blahblah

There's also /proc/kcore which seems to be total physical + virtual
memory;
peter@marvin:~$ ls -l /proc/kcore 
-r1 root root 134221824 Feb  4 10:01 /proc/kcore

But I'm not sure if physical memory only is represented in /proc
anywhere.

Me, I usually just make an educated guess based on the output of
free. peter@marvin:~$ free
 total   used
Mem:127588 123136

is close enough to 128MB for me. :-)

-- 
Pete
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] Email problem...

[Peter Rundle]

Chris,

You're right, it was late and I was having a poke (hence the emoticon
in the original posting).

Back to the problem,

Jeff,

The server (203.219.50.38) is accessible to the outside world for port
110 but not 25. Until this is fixed there is no hope of getting mail
inbound to it.

However a reverse lookup of that server returns

  38.50.219.203.in-addr.arpa  name = free-tpg-038.tpgi.com.au.

And a subsequent MX search of the domain tpgi.com.au returns three
servers each with 3 addresses overlapping each other but none of
which are the above server. All three respond on port 25 listing
sendmail as the SMTP agent.

So..you have two issues.

 1. If tpgi.com.au is the mail domain in question then, your DNS is
pointing in-bound mail to other IP addresses,

 2. the IP address that you wish to receive inbound mail on is
not accessible from the internet on port 25.

The lack of access on port 25 to the outside world could either be
a blocking at your adsl firewall, or possibly that port is blocked
by iptables on the server itself. Yes I know local traffic can get
through but imagine that you have a rule which says allow inbound
to all ports if source address = 192.168.0.0 then a following rule
which denys access to all. This would let your local users sendmail
but make the server inaccesible to the outside world. I suspect that
this is the case because the telnet to port 25 is being denied not
dropped.

BTW, I know that posting ip's to this list feels like a bad thing to
do from a security point of view, but if you have a mail server then
you advertise that ip in the DNS, so once you make the decision to
host e-mail you are already asking for it. So you haven't really
compromised security by letting the ip out of the bag.

HTH

P.



--
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] Calculating physical memory from /proc/meminfo

[John Clarke]

On Tue, Feb 04, 2003 at 09:42:27AM +1100, [EMAIL PROTECTED] wrote:

> # grep kernel /var/log/dmesg
> Freeing unused kernel memory: 264k freed
> 
> As  you can see I don't get the same result. It seems that dmesg hasn't 
> stored information as far back as on the other machine.
> 
> Is there another place where this information would be stored?

If you haven't rotated the log out of existence, it'll be in one of
/var/log/messages*:

[root@dropbear ~]# grep -i memory /var/log/messages*
/var/log/messages.4:Jan  8 10:50:50 dropbear kernel: Memory: 
511988k/524032k available (1012k kernel code, 9480k reserved, 
844k data, 80k init, 0k highmem)
/var/log/messages.4:Jan  8 10:50:50 dropbear kernel: Freeing unused 
kernel memory: 80k freed

Note that 524032k is slightly (256k) less than the 512MB I have in this
machine.


Cheers,

John
-- 
whois [EMAIL PROTECTED]
GPG key id: 0xD59C360F
http://kirriwa.net/john/
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] Calculating physical memory from /proc/meminfo

[mkraus]

G'day...

Using `free` I encounter the same problems as when examining /proc/meminfo

However, I searched the archives as you suggested and found that examining 
dmesg should show physical memory. dmesg had scrolled way past that point 
(and hence lost) on my machines so I did a `grep kernel /var/log/dmesg` to 
get:

Memory: 53904k/57344k available (1170k kernel code, 3056k reserved, 332k 
data, 260k init, 0k highmem)
Freeing unused kernel memory: 260k freed

Which showed me the true physical memory  - 57344k which is 56Mg, which 
all makes sense. (Memory is almost always in multiples of 8Mg or similar 
depending on the motherboard and configuration.)

However, when I go to one of our other servers and do the same search:

# grep kernel /var/log/dmesg
Freeing unused kernel memory: 264k freed

As  you can see I don't get the same result. It seems that dmesg hasn't 
stored information as far back as on the other machine.

Is there another place where this information would be stored?

Thanks...

Mike
---
Michael S. E. Kraus
Administration
Capital Holdings Group (NSW) Pty Ltd
[EMAIL PROTECTED]
phone (02) 9955 8000 fax (02) 9955 8144




Peter Hardy <[EMAIL PROTECTED]>
Sent by: [EMAIL PROTECTED]
03/02/2003 01:00 PM

 
To: [EMAIL PROTECTED]
cc: 
Subject:Re: [SLUG] Calculating physical memory from /proc/meminfo


On Mon, 3 Feb 2003 12:42:43 +1100 [EMAIL PROTECTED] wrote:
> What is the process from calculating physical memory from `cat 
> /proc/meminfo` ?

free(1) parses the output of meminfo into something a little more
readable.

> In the memtotal line is that including swap memory?

Comparing the output from free with the contents of /proc/meminfo, it
seems that the memtotal line doesn't include swap.

> When I calculate the number of megabytes, its never quite right - so I
> was wondering what techniques others use.

free. :-)

There's been quite a lot of discussion on interpreting memory usage in
the past. Perusing the list archives at http://slug.org.au/archives.html
should help out.

-- 
Pete
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug



-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

[SLUG] Samba PDC and Win2k

[Phil Scarratt]

Hi all

This is possibly a little off-topic, but just wondering what people are doing re
group policies in Win2k when using a Samba PDC. Are you just modifying the NT4
system policies to suit w2k or is there a way of implementing group policies
properly without installing MoneySoft's w2k server.

Fil
-- 
Phil Scarratt
IT Consultant
0403 531 271

-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] RH8 installation probs

[Dave Airlie]

sounds like a CD speed issue or something.. and poor quality CDs.. but
I've no idea how to get around it .. you might be able to use hdparm to
set the CD speed at the  bash prompt.. I think MS discs used to do
that and is why stuff took ages to install...

Dave.

On Tue, 4 Feb 2003, Phil Scarratt wrote:

> I had a very similar problem except the error would happen with random packages,
> not always the same one. I had downloaded the iso images and burnt to CD. The
> media verification passed OK also. I ended up using a friends copy which worked.
>
> Fil
>
>
> Quoting mark davis <[EMAIL PROTECTED]>:
>
> > Hello how are you?
> > I am begining to learn about rh8 and bought my first
> > copy today. Unfortunately it wont even load up :(
> > it boots up on the disk and you can go through all its
> > procedures and selection desk top set up (the first
> > selection on the top) and then enter your passwords
> > etc. but when it starts to load everythinng it always
> > comes up with a fault and then resets! :(  the fault
> > says this.  "The
> > file/mnt/sysimage/var/tmp/zlib-1.1.44.i386.rpm can not
> > be opened. this is due to a mising file, a bad
> > package, or bad media.. press return to try again"
> > but it doesnt try again, it just resets everything.
> > I did a media check and its past, I dont understand if
> > you buy a original copy of red hat which i have done,
> > it should atleast install properly :(
> > If anyone can help me, I would realy appreciate it.
> > [EMAIL PROTECTED]  or if ya wanna phone
> > 9862-6683 im in(sydney).
> > Many thanks
> > Darren
> >
> > __
> > Do you Yahoo!?
> > Yahoo! Mail Plus - Powerful. Affordable. Sign up now.
> > http://mailplus.yahoo.com
> > --
> > SLUG - Sydney Linux User's Group - http://slug.org.au/
> > More Info: http://lists.slug.org.au/listinfo/slug
> >
>
>
> -
> Phil Scarratt
> It Consultant
> 0403 531 271
>
>

-- 
David Airlie, Software Engineer
http://www.skynet.ie/~airlied / [EMAIL PROTECTED]
pam_smb / Linux DecStation / Linux VAX / ILUG person


-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] RH8 installation probs

[Phil Scarratt]

I had a very similar problem except the error would happen with random packages,
not always the same one. I had downloaded the iso images and burnt to CD. The
media verification passed OK also. I ended up using a friends copy which worked.

Fil


Quoting mark davis <[EMAIL PROTECTED]>:

> Hello how are you?
> I am begining to learn about rh8 and bought my first
> copy today. Unfortunately it wont even load up :(
> it boots up on the disk and you can go through all its
> procedures and selection desk top set up (the first
> selection on the top) and then enter your passwords
> etc. but when it starts to load everythinng it always
> comes up with a fault and then resets! :(  the fault
> says this.  "The
> file/mnt/sysimage/var/tmp/zlib-1.1.44.i386.rpm can not
> be opened. this is due to a mising file, a bad
> package, or bad media.. press return to try again" 
> but it doesnt try again, it just resets everything.  
> I did a media check and its past, I dont understand if
> you buy a original copy of red hat which i have done,
> it should atleast install properly :(
> If anyone can help me, I would realy appreciate it.
> [EMAIL PROTECTED]  or if ya wanna phone
> 9862-6683 im in(sydney).
> Many thanks
> Darren
> 
> __
> Do you Yahoo!?
> Yahoo! Mail Plus - Powerful. Affordable. Sign up now.
> http://mailplus.yahoo.com
> -- 
> SLUG - Sydney Linux User's Group - http://slug.org.au/
> More Info: http://lists.slug.org.au/listinfo/slug
> 


-
Phil Scarratt
It Consultant
0403 531 271

-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] RH8 installation probs

[Adam Hewitt]

Not that there is any problem in asking the list, as there are plenty of
people on here who could help you, but isn't the point of purchasing a
copy of RH so that you get the RH support?? If so maybe you should
contact them and find out if they maybe released a bad batch of CD's or
something.

Then again, I use Debian so I probably have no idea...

Adam.

On Mon, 2003-02-03 at 22:30, mark davis wrote:
> Hello how are you?
> I am begining to learn about rh8 and bought my first
> copy today. Unfortunately it wont even load up :(
> it boots up on the disk and you can go through all its
> procedures and selection desk top set up (the first
> selection on the top) and then enter your passwords
> etc. but when it starts to load everythinng it always
> comes up with a fault and then resets! :(  the fault
> says this.  "The
> file/mnt/sysimage/var/tmp/zlib-1.1.44.i386.rpm can not
> be opened. this is due to a mising file, a bad
> package, or bad media.. press return to try again" 
> but it doesnt try again, it just resets everything.  
> I did a media check and its past, I dont understand if
> you buy a original copy of red hat which i have done,
> it should atleast install properly :(
> If anyone can help me, I would realy appreciate it.
> [EMAIL PROTECTED]  or if ya wanna phone
> 9862-6683 im in(sydney).
> Many thanks
> Darren
> 
> __
> Do you Yahoo!?
> Yahoo! Mail Plus - Powerful. Affordable. Sign up now.
> http://mailplus.yahoo.com
> -- 
> SLUG - Sydney Linux User's Group - http://slug.org.au/
> More Info: http://lists.slug.org.au/listinfo/slug



-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

[SLUG] Re:email server

[Jeff Hutchinson]

The server sits on  a private network and is being NATed to the outside.
The adsl router is doing this Netcomm NB1400. Inbound I have let 110/tcp
and 25/tcp through. I have told the router which machine to resolve
these packets to which is my mail server. Anything can get out.

Sorry about being a bit security consicene. But if you would like to try
the ip of the machine in question (well the IP I have being assigned)
203.219.50.38. You well get through on port 110 but not 25.

I was doubtful of the DNS. There is another worry in that I can send
mail to this address and it seems to disappear into a black hole there
is no bounce!

Thankyou for all of your replies.I will try the suggestions tomorrow. 

Jeff





-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

[SLUG] RH8 installation probs

[mark davis]

Hello how are you?
I am begining to learn about rh8 and bought my first
copy today. Unfortunately it wont even load up :(
it boots up on the disk and you can go through all its
procedures and selection desk top set up (the first
selection on the top) and then enter your passwords
etc. but when it starts to load everythinng it always
comes up with a fault and then resets! :(  the fault
says this.  "The
file/mnt/sysimage/var/tmp/zlib-1.1.44.i386.rpm can not
be opened. this is due to a mising file, a bad
package, or bad media.. press return to try again" 
but it doesnt try again, it just resets everything.  
I did a media check and its past, I dont understand if
you buy a original copy of red hat which i have done,
it should atleast install properly :(
If anyone can help me, I would realy appreciate it.
[EMAIL PROTECTED]  or if ya wanna phone
9862-6683 im in(sydney).
Many thanks
Darren

__
Do you Yahoo!?
Yahoo! Mail Plus - Powerful. Affordable. Sign up now.
http://mailplus.yahoo.com
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

[SLUG] Dial on demand

[Clint Gaddes]

Hi Guys,

I am looking for a simple way to configure the dial on demand for squid.

I have configured red hat 7.3 and squid. Manually configured ppp to dial
local ISP.

THe problem I have run across is the ability for squid to be able to
dial on demand. I have looked at Diald, but I was wounding if there was
a better \ easier way to achieve demand on dial.

You would think there would be a better way than Diald?

Thanks

-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] Email problem...

[Christopher Samuel]

-BEGIN PGP SIGNED MESSAGE-

On Monday 03 Feb 2003 9:37 pm, Peter Rundle wrote:
> Jeff,
>
> Tsk, Tsk, Look what happens when I telnet to port 80 on your
> server.
>
>   telnet 203.51.63.160 80

Strange, I don't recall seeing him post the IP address of the server, only the 
client.

> So your e-mail should really begin,
>
>"I have setup a MicroSoft email server which."
[...]

The logs he posted clearly showed ipop3d activity from a UNIX box of some 
sort, with that IP address collecting mail from it.

Chris
- -- 
Chris SamuelWollongong, NSW

-BEGIN PGP SIGNATURE-
Version: GnuPG v1.0.7 (GNU/Linux)

iQEVAwUBPj5NnY1yjaOTJg85AQG0TAf+PJiQb4qOf+u+FaG3fpl+qIduHcnOdGf/
cBy9ds96F5Z5ZUzUdnhTvgjKH8WHl2H/ATKweZvIymL562Qhrg4OMDlIC3MlV4uD
KbeNJ9FTwyjgdjPqKYhdKtLOvYWQdD05WvXFDV60YgqkPy/7RLuGfgsM6kJxi201
d2swM9mwwHKjkuxjAXymyyhkG4D7OoC7bd0cLWzXXGAq5BQ0c224PATis4GLrQ7Z
w07WdGFXQ+QCch791SETC2tj7t67JsPEdA671NzxJXIsPg70o6pa1fAes8hvO8JP
qNEBRdJUlyJj2vYYmNBQEBJpixviNaO3hvCsRCRoHfJZJg5Ayy/cDw==
=33fx
-END PGP SIGNATURE-

-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] Mixing Memory Brands

[Heracles]

On Friday 31 Jan 2003 8:57 am, Jonathan Kelly wrote:
> Hi,
>
> does anyone have any knowledge/experience of there being problems
> mixing *brands* of memory. The memory is PC2100 (266MHz) Reg Ecc. I
> currenty have one 512M stick of Crucial installed, and need another,
> but there seems to be upto about 25% difference in price depending on
> brand (and of course Crucial seems the most expensive).
>
> Also, any recomendations for good people to buy from?

Mixing brands can cause problems when there are slight differences in the 
speed - and there are sometimes.
Try Q-Flow at Eastwood  or Westgate Technology at Rydlemare for the best 
prices. (Westgate also sell at the North Rocks Market on Sundays)

HTH
Stay well and happy
Heracles
-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

RE: [SLUG] Email problem...

[Kevin Saenz]

Does the mail server have 2 nics?

> Sorry, sendmail.
> 
> If the server was not configured correctly it would not be sending and
> recieving internally.??
> 
> jeff
> 
> On Mon, 2003-02-03 at 20:12, Kevin Saenz wrote:
> > on the machine if you do
> > netstat -an | grep 25 
> > 
> > what happens
> > 
> > it sounds like you may not have configured your mail server
> > correctly
> > 
> > what mail server are you using? postfix, qmail, or sendmail?
-- 
Kevin Saenz <[EMAIL PROTECTED]>

-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] Email problem...

[Peter Rundle]

Jeff,

Tsk, Tsk, Look what happens when I telnet to port 80 on your
server.

 telnet 203.51.63.160 80
 Trying 203.51.63.160...
 Connected to 203.51.63.160.
 Escape character is '^]'.
 get
 HTTP/1.1 400 Bad Request
 Server: Microsoft-IIS/5.0
 Date: Mon, 03 Feb 2003 10:51:55 GMT
 Content-Type: text/html
 Content-Length: 87

 ErrorThe parameter is
 incorrect. Connection closed by foreign host.

So your e-mail should really begin,

  "I have setup a MicroSoft email server which."

Sympathy ends, before I provide any more support would you please
indicate your prefered payment method and acknowledge the $500
hourly rate. ;-)

P.


--
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] Email problem...

[Peter Rundle]

Jeff,

If I try to telnet to your e-mail server on port 25 this is
what happens,

 $ telnet 203.51.63.160 25
 Trying 203.51.63.160...
 telnet: connect to address 203.51.63.160: Connection refused

Until this is fixed there is zero possibility to receive inbound
mail. I cannot tell if your server is rejecting the connection
or if it is the firewall that is doing it. How is your firewall
configured? Is it allowing tcp through on port 25 to that address.
Does your mail server have a real world ip address or is it being
Nat'd by the firewall? In this case what is the firewall? An appliance
like a Cisco or are you using Linux/iptables?

When you say that you can telnet to the server and it responds how
are you testing this? I certainly can't telnet to it on ports 110
25 or 23. But you are visible to me via ping.

  PING 203.51.63.160 from 192.168.1.10 : 56(84) bytes of data.
  64 bytes from 203.51.63.160: icmp_seq=0 ttl=112 time=70.914 msec
  64 bytes from 203.51.63.160: icmp_seq=1 ttl=112 time=58.177 msec

Are bigpond firewalling further upstream that you aren't aware of?
Your log reverse dns's you as

  CPE-203-51-63-160.nsw.bigpond.net.au

So I assume you have some sort of connection / service with them,
what is it and are you "allowed" in Telstra speak to have a server?

Once you have fixed the access to port 25 from outside then you
should start receiving e-mail.

HTH

Pete

--
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

RE: [SLUG] Email problem...

[Jeff Hutchinson]

Sorry, sendmail.

If the server was not configured correctly it would not be sending and
recieving internally.??

jeff

On Mon, 2003-02-03 at 20:12, Kevin Saenz wrote:
> on the machine if you do
> netstat -an | grep 25 
> 
> what happens
> 
> it sounds like you may not have configured your mail server
> correctly
> 
> what mail server are you using? postfix, qmail, or sendmail?


-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] Email problem...

[Jeff Hutchinson]

On Mon, 2003-02-03 at 20:02, Christopher Samuel wrote:
> -BEGIN PGP SIGNED MESSAGE-
> 
> On Monday 03 Feb 2003 6:47 pm, Jeff Hutchinson wrote:
>
> 1) The MX record for the domain points to an address record for the mail 
> server. It must *not* point to a CNAME and it must *not* be an IP address.
> 
> i.e.  mydomain.org.au IN  MX  5   mailhost.mydomain.org.au
>   mailhost.mydomain.org.auIN  A   10.1.1.1
> 
> 2) Confirm that from a third party such as http://www.samspade.org/t/
> 
> 3) Use tcpdump (as above) to confirm incoming packets.
> 
>Yes, I have an MX record. I can do a host -t mx mydomain.com and it returns with
my mail server.
Thanks for your input I will try what you suggest tomorrow.

Jeff



-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

RE: [SLUG] Email problem...

[Kevin Saenz]

on the machine if you do
netstat -an | grep 25 

what happens

it sounds like you may not have configured your mail server
correctly

what mail server are you using? postfix, qmail, or sendmail?

all are configured differently



> On Mon, 2003-02-03 at 19:46, Ben de Luca wrote:
> > Try telneting to port 25 on your email server from external to the mail
> > server. If you get in type
> > 
> > 
> > HELO
> > 
> > 
> > Then type help and try and send an email the fun way
> > 
> > 
> I can't login. 
> telnet: Unable to connect to remote host: Connection refused.
> 
> Can you tell me what this means for my email server? More to the point
> why is it sending and receiving internally but not externally?
> 
> Thanks,
> 
> Jeff 
-- 
Kevin Saenz <[EMAIL PROTECTED]>

-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] Email problem...

[Christopher Samuel]

-BEGIN PGP SIGNED MESSAGE-

On Monday 03 Feb 2003 7:23 pm, Jeff Hutchinson wrote:

> I can't login.
> telnet: Unable to connect to remote host: Connection refused.
>
> Can you tell me what this means for my email server? More to the point
> why is it sending and receiving internally but not externally?

OK, are you typing:

telnet myserver.mydomain 25

or just:

telnet myserver.mydomain 

?

Because you'll need to do the first to connect to the SMTP mail port on your 
server. If that is being rejected then that's probably a good reason why you 
can't receive email!

Chris
- -- 
Chris SamuelWollongong, NSW

-BEGIN PGP SIGNATURE-
Version: GnuPG v1.0.7 (GNU/Linux)

iQEUAwUBPj40L41yjaOTJg85AQE4swf3aHGJgGtjWzu39WZ/u1zt2YfNfWplPHXr
wc1m36VWiYG4uNkDkNO6y+4H41+nhEhloIK7IOS7MhV0+P0XtLcXP71wU9DFdVr8
8vfwLdvsz0jUJeXcPQjfrFvXMtWzwSw8HDDRuCJswumSmYpJJuvGYCBEixrkM10F
J7u0ghDvPGbwX5wA2MLGVwg7i0pa85lQQCCWMGqenX7uJ0pVqHnxi3XeKM6ezj08
jtv0wWhawm1uC8LFrHIxDBbto7VIoCGdCBbib2TNVKZswfdPgILFrniscT8Gn/l3
yOjnsdSZWHzrEmYfqTauBh0UIXIE02pHFFyrQ3okG/luXn4pPUIO
=m/X8
-END PGP SIGNATURE-

-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Re: [SLUG] Email problem...

[Christopher Samuel]

-BEGIN PGP SIGNED MESSAGE-

On Monday 03 Feb 2003 6:47 pm, Jeff Hutchinson wrote:

> I have setup an email server which stands behind a firewall. Internally
> we are able to sent and receive email. I fact we can send email
> anywhere. The problem lies with the incoming email --- none.

Obvious first question - does your primary MX point to the A record for the 
mail server, or are there no MX records and just an A record (either should 
work, but an MX record is always good just for luck).  Then, if mail is sent 
to other MX's, do those have the correct rules to forward onto you ?

Also, does the firewall log any attempted inbound SMTP connections, successful 
or not ?   If it's not configured to, then set it to do so so you'll have 
some idea of there are any incoming packets.

If you can't do that do:

tcpdump -n port 25

on the mailhost to watch SMTP transactions.

> I have checked the following,
>
> 1. Allowed port 110/tcp and 25/tcp through the firewall. From outside
> the firewall you can telnet to the server and it responds.

OK, so if no mail is getting to you then it's highly likely it's a DNS issue.

Check:

1) The MX record for the domain points to an address record for the mail 
server. It must *not* point to a CNAME and it must *not* be an IP address.

i.e.mydomain.org.au IN  MX  5   mailhost.mydomain.org.au
mailhost.mydomain.org.auIN  A   10.1.1.1

2) Confirm that from a third party such as http://www.samspade.org/t/

3) Use tcpdump (as above) to confirm incoming packets.

4) From the outside send an email using telnet to port 25.

> 2. I'm fairly sure DNS is working but I have the FEATURE( accept_
> unresolved_domains) turned on ( not my favorite option ).
>
> I have attached part of the maillog.
> The system is Redhat 8.

Erm, that log file only had POP traffic, no sendmail log entries at all.
- -- 
Chris SamuelWollongong, NSW

-BEGIN PGP SIGNATURE-
Version: GnuPG v1.0.7 (GNU/Linux)

iQEVAwUBPj4wOo1yjaOTJg85AQF3sQf7BgbElSmSJTON5A8MvvnFv5pN56IQh3+a
FV1MIZp6PTVmipRUrFS1LgHoSSj0+9SCmX0atGO6xvG1PlA480v++wnC4Fb72s2q
wJgJozs39delVQoKrLotjxhYvpTipBJhEB4XPnVIUP4z8G9THu915KoY7Rr9JYNU
AItJUZp740qTcEXH9Ep9o2iYHaBc1ZX/vrI29uuHjmFMYIAnxm6hSpEIMnlOSDn9
Gb4ZJa0MX0emLfEuEaWltGw3LxF4y8CaSI2/QzeNC4qBNxm70ZAh4pkHxSPqb8Ax
A0Wxu2crj26BTc8CxL3NtOOCW63UolhQGTJiqf1l0wlmRMv4xXvGSg==
=30FU
-END PGP SIGNATURE-

-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

RE: [SLUG] Email problem...

[Jeff Hutchinson]

On Mon, 2003-02-03 at 19:46, Ben de Luca wrote:
> Try telneting to port 25 on your email server from external to the mail
> server. If you get in type
> 
> 
> HELO
> 
> 
> Then type help and try and send an email the fun way
> 
> 
I can't login. 
telnet: Unable to connect to remote host: Connection refused.

Can you tell me what this means for my email server? More to the point
why is it sending and receiving internally but not externally?

Thanks,

Jeff 

-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

RE: [SLUG] Email problem...

[Ben de Luca]

Try telneting to port 25 on your email server from external to the mail
server. If you get in type


HELO


Then type help and try and send an email the fun way



-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] On Behalf
Of Jeff Hutchinson
Sent: Monday, 3 February 2003 5:47 PM
To: [EMAIL PROTECTED]
Subject: [SLUG] Email problem...


I have setup an email server which stands behind a firewall. Internally
we are able to sent and receive email. I fact we can send email
anywhere. The problem lies with the incoming email --- none.

I have checked the following,

1. Allowed port 110/tcp and 25/tcp through the firewall. From outside
the firewall you can telnet to the server and it responds.

2. I'm fairly sure DNS is working but I have the FEATURE( accept_
unresolved_domains) turned on ( not my favorite option ).

I have attached part of the maillog.
The system is Redhat 8.

Its probably something stupid I am doing.
Any help appreciated,

Jeff 



-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

[SLUG] Email problem...

[Jeff Hutchinson]

I have setup an email server which stands behind a firewall. Internally
we are able to sent and receive email. I fact we can send email
anywhere. The problem lies with the incoming email --- none.

I have checked the following,

1. Allowed port 110/tcp and 25/tcp through the firewall. From outside
the firewall you can telnet to the server and it responds.

2. I'm fairly sure DNS is working but I have the FEATURE( accept_
unresolved_domains) turned on ( not my favorite option ).

I have attached part of the maillog.
The system is Redhat 8.

Its probably something stupid I am doing.
Any help appreciated,

Jeff 



Feb  3 14:13:44 lister ipop3d[2614]: pop3 service init from 203.51.63.160
Feb  3 14:13:44 lister ipop3d[2614]: Login user=jeff 
host=CPE-203-51-63-160.nsw.bigpond.net.au [203.51.63.160] nmsgs=0/0
Feb  3 14:13:44 lister ipop3d[2614]: Logout user=jeff 
host=CPE-203-51-63-160.nsw.bigpond.net.au [203.51.63.160] nmsgs=0 ndele=0
Feb  3 14:14:24 lister ipop3d[2736]: pop3 service init from 203.51.63.160
Feb  3 14:14:25 lister ipop3d[2736]: Login user=jeff 
host=CPE-203-51-63-160.nsw.bigpond.net.au [203.51.63.160] nmsgs=0/0
Feb  3 14:14:25 lister ipop3d[2736]: Logout user=jeff 
host=CPE-203-51-63-160.nsw.bigpond.net.au [203.51.63.160] nmsgs=0 ndele=0
Feb  3 14:14:28 lister ipop3d[2737]: pop3 service init from 203.51.63.160
Feb  3 14:14:28 lister ipop3d[2737]: Login user=jeff 
host=CPE-203-51-63-160.nsw.bigpond.net.au [203.51.63.160] nmsgs=0/0
Feb  3 14:14:28 lister ipop3d[2737]: Logout user=jeff 
host=CPE-203-51-63-160.nsw.bigpond.net.au [203.51.63.160] nmsgs=0 ndele=0
Feb  3 14:14:33 lister ipop3d[2738]: pop3 service init from 203.51.63.160
Feb  3 14:14:33 lister ipop3d[2738]: Login user=jeff 
host=CPE-203-51-63-160.nsw.bigpond.net.au [203.51.63.160] nmsgs=0/0