🚀 go-pugleaf

In article <slrn98jr2s.62r.philh@comuno.freeserve.co.uk>
           philh@comuno.freeserve.co.uk "phil hunt" writes:

> When attempting to post email to freeserve, I often get an error message
> from sendmail:

>   Running BAA06205 (sequence 1 of 2)
>   <lgrinzo@stny.rr.com>... Deferred: 
>   Name server: smtp.freeserve.net: host name lookup failure

It's unable to look up the name "stny.rr.com".  The odd thing is the 
reference to smtp.freeserve.net as the name server.

> Note that i can still ping smtp.freeserve.net:

>   philh:~/ftpin> ping smtp.freeserve.net
>   PING relay.pol.net.uk (195.92.195.153): 56 data bytes
>   64 bytes from 195.92.195.153: icmp_seq=0 ttl=248 time=682.883 ms
>   64 bytes from 195.92.195.153: icmp_seq=1 ttl=248 time=1951.710 ms
>   64 bytes from 195.92.195.153: icmp_seq=2 ttl=248 time=960.043 ms
>   64 bytes from 195.92.195.153: icmp_seq=3 ttl=248 time=631.747 ms
>   64 bytes from 195.92.195.153: icmp_seq=4 ttl=248 time=1431.713 ms
>   64 bytes from 195.92.195.153: icmp_seq=5 ttl=248 time=1091.786 ms
>   64 bytes from 195.92.195.153: icmp_seq=6 ttl=248 time=1220.277 ms
>   --- relay.pol.net.uk ping statistics ---
>   8 packets transmitted, 7 packets received, 12% packet loss
>   round-trip min/avg/max = 631.747/1138.594/1951.710 ms    

That's damned slow ping!  But I'm guessing that something else 
is taking place over the link at the same time.  That, too, could 
interfere with name resolution.
-- 
Raj Rijhwani        (umtsb5/16) |  This is the voice of the Mysterons...
raj@rijhwani.org                |  ... We know that you can hear us Earthmen
http://www.rijhwani.org/raj/    |  "Lieutenant Green:  Launch all Angels!"

raj@rijhwani.org (Raj Rijhwani) writes:

> In article <slrn98jr2s.62r.philh@comuno.freeserve.co.uk>
>            philh@comuno.freeserve.co.uk "phil hunt" writes:
> 
> > When attempting to post email to freeserve, I often get an error message
> > from sendmail:
> 
> >   Running BAA06205 (sequence 1 of 2)
> >   <lgrinzo@stny.rr.com>... Deferred: 
> >   Name server: smtp.freeserve.net: host name lookup failure
> 
> It's unable to look up the name "stny.rr.com". The odd thing is the
> reference to smtp.freeserve.net as the name server.

*Hmm*. I'm not convinced you're reading the last line of the message right,
there. To me, based on other sendmail logs here, it looks more like
"whatever my nameserver is, it can't lookup smtp.freeserve.net", not that
it's using smtp.f.n *as* a nameserver.

> > Note that i can still ping smtp.freeserve.net:
> 
> >   philh:~/ftpin> ping smtp.freeserve.net
> >   PING relay.pol.net.uk (195.92.195.153): 56 data bytes
[snip]
> >   64 bytes from 195.92.195.153: icmp_seq=6 ttl=248 time=1220.277 ms
> >   --- relay.pol.net.uk ping statistics ---
> >   8 packets transmitted, 7 packets received, 12% packet loss
> >   round-trip min/avg/max = 631.747/1138.594/1951.710 ms    
> 
> That's damned slow ping! But I'm guessing that something else is taking
> place over the link at the same time. That, too, could interfere with
> name resolution.

True, indeed, it is.

Phil: how big's the `mailq` on that box? If you're on Freeserve as it would
appear, you're using dynamic IP#s too, yeah? How's about restarting bind
and sendmail and kicking off a `sendmail -q'[i]?

(Me, I have my Suspicions about kernels and dynamic IP#s and interfaces
being given new IP#s or even just having the existing one re-established
and whole processes behind a masquerading gateway never being able to send
a packet... I'm sure there's a problem lurking, I don't trust bind & squid
not to hang eternally behind a *Demon* dialup with `noipdefault' enabled.
Mwaaaa, anyone..?)

~Tim

Footnotes: 
[i]  I still miss `runq', when it's sendmail that's installed; older
sendmails had it and exim still has it, too. Nice command.

-- 
The sun is melting over the hills,           |piglet@stirfried.vegetable.org.uk
All our roads are waiting / To be revealed   |http://spodzone.org.uk/

In article <slrn98jr2s.62r.philh@comuno.freeserve.co.uk>, one of infinite monkeys
	at the keyboard of philh@comuno.freeserve.co.uk (phil hunt) hath written:
> When attempting to post email to freeserve, I often get an error message
> from sendmail:
> 
>   Running BAA06205 (sequence 1 of 2)
>   <lgrinzo@stny.rr.com>... Deferred: 
>   Name server: smtp.freeserve.net: host name lookup failure

[ NOTE - you just gave the lgrinzo address, if it's genuine, to a lot
  of spam harvesters ]

[ vague guessing waffle follows - would take too long to check ]

Have you set smtp.freeserve.net as your relay?  A quick look at my
smtproutes shows smtp.freeserve.co.uk for places outside my own
namespace.  If smtp.freeserve.net were just a CNAME, some software
might complain.

[ other thought that bears no relation to the above ]

Haven't looked recently, but didn't Freeserve once reroute port 25
forcing outgoing smtp through their relay?  If you aren't configured
to use the relay explicitly, that could lead to misleading error
messages.

-- 
Nick Kew

Is your site a lawsuit waiting to happen?
See <URL:http://valet.webthing.com/intranet/> before it's too late.

On Wed, 14 Feb 2001 04:49:03 GMT, Raj Rijhwani <raj@rijhwani.org> wrote:
>In article <slrn98jr2s.62r.philh@comuno.freeserve.co.uk>
>           philh@comuno.freeserve.co.uk "phil hunt" writes:
>
>> When attempting to post email to freeserve, I often get an error message
>> from sendmail:
>
>>   Running BAA06205 (sequence 1 of 2)
>>   <lgrinzo@stny.rr.com>... Deferred: 
>>   Name server: smtp.freeserve.net: host name lookup failure
>
>It's unable to look up the name "stny.rr.com".  The odd thing is the 
>reference to smtp.freeserve.net as the name server.

I thought it sent the emails to smtp.freeserve.net, which then forwarded
them to the ultimate destination.

>> Note that i can still ping smtp.freeserve.net:
>
>>   philh:~/ftpin> ping smtp.freeserve.net
>>   PING relay.pol.net.uk (195.92.195.153): 56 data bytes
>>   64 bytes from 195.92.195.153: icmp_seq=0 ttl=248 time=682.883 ms
>>   64 bytes from 195.92.195.153: icmp_seq=1 ttl=248 time=1951.710 ms
>>   64 bytes from 195.92.195.153: icmp_seq=2 ttl=248 time=960.043 ms
>>   64 bytes from 195.92.195.153: icmp_seq=3 ttl=248 time=631.747 ms
>>   64 bytes from 195.92.195.153: icmp_seq=4 ttl=248 time=1431.713 ms
>>   64 bytes from 195.92.195.153: icmp_seq=5 ttl=248 time=1091.786 ms
>>   64 bytes from 195.92.195.153: icmp_seq=6 ttl=248 time=1220.277 ms
>>   --- relay.pol.net.uk ping statistics ---
>>   8 packets transmitted, 7 packets received, 12% packet loss
>>   round-trip min/avg/max = 631.747/1138.594/1951.710 ms    
>
>That's damned slow ping!  But I'm guessing that something else 
>is taking place over the link at the same time.

nntp and http traffic, mostly.

-- 
*****[ Phil Hunt ***** philh@comuno.freeserve.co.uk ]*****
"An unforseen issue has arisen with your computer. Don't worry your
silly little head about what has gone wrong; here's a pretty animation
of a paperclip to look at instead." -- Windows2007 error message

               

On 14 Feb 2001 10:27:52 +0000, Tim Haynes <piglet@stirfried.vegetable.org.uk> wrote:
>raj@rijhwani.org (Raj Rijhwani) writes:
>
>> In article <slrn98jr2s.62r.philh@comuno.freeserve.co.uk>
>>            philh@comuno.freeserve.co.uk "phil hunt" writes:
>> 
>> > When attempting to post email to freeserve, I often get an error message
>> > from sendmail:
>> 
>> >   Running BAA06205 (sequence 1 of 2)
>> >   <lgrinzo@stny.rr.com>... Deferred: 
>> >   Name server: smtp.freeserve.net: host name lookup failure
>> 
>> It's unable to look up the name "stny.rr.com". The odd thing is the
>> reference to smtp.freeserve.net as the name server.
>
>*Hmm*. I'm not convinced you're reading the last line of the message right,
>there. To me, based on other sendmail logs here, it looks more like
>"whatever my nameserver is, it can't lookup smtp.freeserve.net", not that
>it's using smtp.f.n *as* a nameserver.

smtp.freeserve.net is an smtp server, I would imagine.

If it was a nameserver, it would be given as an IP address, not a DNS
name, I imagine.

>> That's damned slow ping! But I'm guessing that something else is taking
>> place over the link at the same time. That, too, could interfere with
>> name resolution.
>
>True, indeed, it is.
>
>Phil: how big's the `mailq` on that box? 

At the time, there were 2 messages in it.

>If you're on Freeserve as it would
>appear, you're using dynamic IP#s too, yeah? 

That's how it's set up in kppp, yes.

(I'm using SuSE6.4, BTW).

>How's about restarting bind

How would I do that?

>and sendmail and kicking off a `sendmail -q'[i]?

I normally manually execute ``sendmail -q -v'' to post emails; that's
what I did above.

>(Me, I have my Suspicions about kernels and dynamic IP#s and interfaces
>being given new IP#s or even just having the existing one re-established
>and whole processes behind a masquerading gateway never being able to send
>a packet... I'm sure there's a problem lurking, I don't trust bind & squid
>not to hang eternally behind a *Demon* dialup with `noipdefault' enabled.
>Mwaaaa, anyone..?)

-- 
*****[ Phil Hunt ***** philh@comuno.freeserve.co.uk ]*****
"An unforseen issue has arisen with your computer. Don't worry your
silly little head about what has gone wrong; here's a pretty animation
of a paperclip to look at instead." -- Windows2007 error message

               

philh@comuno.freeserve.co.uk (phil hunt) writes:

> >*Hmm*. I'm not convinced you're reading the last line of the message
> >right, there. To me, based on other sendmail logs here, it looks more
> >like "whatever my nameserver is, it can't lookup smtp.freeserve.net",
> >not that it's using smtp.f.n *as* a nameserver.
> 
> smtp.freeserve.net is an smtp server, I would imagine.
> 
> If it was a nameserver, it would be given as an IP address, not a DNS
> name, I imagine.

That too. Not even FS are all that likely to get their numbers in that much
of a tizz, it's only the bean-counting department that lost the plot...

> >Phil: how big's the `mailq` on that box? 
> 
> At the time, there were 2 messages in it.

Hmmm. OK. Not much traffic then.

> (I'm using SuSE6.4, BTW).

Bless you.

> >How's about restarting bind
> 
> How would I do that?

Something to the effect of `/etc/init.d/bind restart' would be favourite
(and compliant); given the SuSE effect, it could well be /sbin/init.d/ and
it might well be `named' not `bind'.

If you're not running your own nameserver, of course, it's a bit difficult
to restart it, so just pass along please ;)

> >and sendmail and kicking off a `sendmail -q'[i]?
> 
> I normally manually execute ``sendmail -q -v'' to post emails; that's
> what I did above.

Hmm. OK. Stop & restart the daemon first, after dialling up, btw.

~Tim
-- 
`Do you mean a Lenox heating/                |piglet@stirfried.vegetable.org.uk
refrigeration unit?', he asked.              |http://spodzone.org.uk/

Nick Kew wrote:

> Haven't looked recently, but didn't Freeserve once reroute port 25
> forcing outgoing smtp through their relay?  If you aren't configured
> to use the relay explicitly, that could lead to misleading error
> messages.

Indeed they did. I believe they still do. An unpleasant habit IMO but
you can sort of see why they do it.

Ian

In article <3A8AFEC9.9269F9FC@house-from-hell.demon.co.uk>, one of infinite monkeys
	at the keyboard of Ian Northeast <ian@house-from-hell.demon.co.uk> hath written:
> 
> Indeed they did. I believe they still do.

Seems likely, but I can't be ***ed to investigate.

>	 An unpleasant habit IMO but
> you can sort of see why they do it.

ISTM that if it gives them a decent handle on spam, then it's well worth
it - and in any case, they're perfectly within their technical rights
(the SMTP) to force their users to use a relay.  Do you have a problem
with it?

-- 
Nick Kew

Is your site a lawsuit waiting to happen?
See <URL:http://valet.webthing.com/intranet/> before it's too late.

nick@fenris.webthing.com (Nick Kew) writes:

> > An unpleasant habit IMO but you can sort of see why they do it.
> 
> ISTM that if it gives them a decent handle on spam, then it's well worth
> it - and in any case, they're perfectly within their technical rights
> (the SMTP) to force their users to use a relay. Do you have a problem
> with it?

Hmmm, IAN(ian), but actually I'll go out on a limb and say I can certainly
see the advantages in doing it, as long as they don't get their smarthosts
in ORBS/RBL by mistake as well :8]

(If you're going to have a smarthost at all, I'd prefer the chance to
forget its existence and plod on happily - that said, I've not fallen foul
of idiots using the DUL in the last 6 months or so, I dunno why.)

~Tim
-- 
   10:57pm  up 52 days,  1:05, 12 users,  load average: 0.84, 0.91, 0.97
piglet@stirfried.vegetable.org.uk |You take your message to the waters,
http://piglet.is.dreaming.org     |And you watch the ripples flow

Tim Haynes <piglet@stirfried.vegetable.org.uk> writes:

> see the advantages in doing it, as long as they don't get their smarthosts
> in ORBS/RBL by mistake as well :8]

If anyone drops a .net.uk host into a blacklist, they're being a bit
silly aren't they?  Then again, DALnet and OPN have both previously
ALine'd btinternet rather than tidying the abuse up properly...

-- 
MJR
Member of the Anglian Linux User Group, UK
Meets every month or so and OPN #alug on Mondays at 2000 UK
http://www.anglian.lug.org.uk/

MJ Ray <markj@cloaked.freeserve.co.uk> writes:

> Tim Haynes <piglet@stirfried.vegetable.org.uk> writes:
> 
> > see the advantages in doing it, as long as they don't get their smarthosts
> > in ORBS/RBL by mistake as well :8]
> 
> If anyone drops a .net.uk host into a blacklist, they're being a bit
> silly aren't they?  

And this would prevent it how? ;)

> Then again, DALnet and OPN have both previously ALine'd btinternet rather
> than tidying the abuse up properly...

Hmmmm. Oops :)

~Tim
-- 
   12:20pm  up 52 days, 14:27, 12 users,  load average: 0.58, 0.91, 1.09
piglet@stirfried.vegetable.org.uk |And the past is sudd'nly the part
http://piglet.is.dreaming.org     |Of life you throw away
                              |

In article <87wvas6vf6.fsf@cloaked.freeserve.co.uk>, MJ Ray wrote:
>Tim Haynes <piglet@stirfried.vegetable.org.uk> writes:
>
>> see the advantages in doing it, as long as they don't get their smarthosts
>> in ORBS/RBL by mistake as well :8]
>
>If anyone drops a .net.uk host into a blacklist, they're being a bit
>silly aren't they?  Then again, DALnet and OPN have both previously
>ALine'd btinternet rather than tidying the abuse up properly...

In DALnet's case, they were being abused by a large number of
btinternet.com hosts and getting no response from abuse@. If an ISP is
unwilling to do anything about its users not playing nicely with other
systems, it seems pretty reasonable to stop them from being able to do any
damage.

Mind you, I've also seen cert at cam.ac.uk refuse to take any action
against a compromised machine that was being used to attack DALnet. Maybe
security people just don't understand IRC?

-- 
Matthew Garrett | mjg59-uk.comp.os.linux@srcf.ucam.org

Matthew Garrett  <mjg59-uk.comp.os.linux@srcf.ucam.org> wrote:
>Mind you, I've also seen cert at cam.ac.uk refuse to take any action
>against a compromised machine that was being used to attack DALnet. Maybe
>security people just don't understand IRC?

That seems extremely unlikely. As a Departmental Computer Officer, I've
been contacted by Cambridge CERT about such a compromised machine, and it
only consists of two people.

Are you sure their reply simply wasn't willing to discuss what action
would be taken with you?
-- 
David/Kirsty Damerell, damerell@chiark.greenend.org.uk.  All Hail Discordia!
|   | And then they came and took me out, The men of doom and malice:  |   |
|---|Destroyed my life, removed my sense, Gave me the poisoned chalice.|---|
| | | My betrayal's life to me...                Elder Sign: Treachery | | |

In article <sUk*0NVOo@news.chiark.greenend.org.uk>, David Damerell wrote:
>Matthew Garrett  <mjg59-uk.comp.os.linux@srcf.ucam.org> wrote:
>>Mind you, I've also seen cert at cam.ac.uk refuse to take any action
>>against a compromised machine that was being used to attack DALnet. Maybe
>>security people just don't understand IRC?

(snip)

>Are you sure their reply simply wasn't willing to discuss what action
>would be taken with you?

In this case, the reply was to a DALnet server admin. The machine in
question happened to be in my college, so the admin asked me to get in
touch with my local COs who had heard nothing from cert@cam.
Unsurprisingly they wanted it removed from the network pretty quickly. I'm
honestly completely unsure of why nothing was done at the university
level, and in most other cases where I've had to deal with them they've
taken action quickly.

-- 
Matthew Garrett | mjg59-uk.comp.os.linux@srcf.ucam.org

In article <86zofpr23b.fsf@potato.vegetable.org.uk>
           piglet@stirfried.vegetable.org.uk "Tim Haynes" writes:

> *Hmm*. I'm not convinced you're reading the last line of the message right,
> there. To me, based on other sendmail logs here, it looks more like
> "whatever my nameserver is, it can't lookup smtp.freeserve.net", not that
> it's using smtp.f.n *as* a nameserver.

Actually, I think you're right.
-- 
Raj Rijhwani        (umtsb5/16) |  This is the voice of the Mysterons...
raj@rijhwani.org                |  ... We know that you can hear us Earthmen
http://www.rijhwani.org/raj/    |  "Lieutenant Green:  Launch all Angels!"