-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi,

On 02/12/13 15:11, Simon Green wrote:

> SMTP error from remote mail server after initial connection: host
> mx.bt.lon5.cpcloud.co.uk [65.20.0.49]: 421 
> smtpin41.bt.ext.cpcloud.co.uk Service not available - no PTR record
> for 134.0.22.242: retry timeout exceeded

We hit a similar problem with Critical Path [cpcloud.co.uk] - it was
found that their DNS lookup would be in uppercase and a bug in a Cisco
ASA firewall in the path was dropping the request.

Not sure if this is a similar issue.


> $ dig +short -x 134.0.22.242
> 
> shared4.wirehive.net.

Confirmed:

marcus@marcus-laptop ~ $ host 134.0.22.242
242.22.0.134.in-addr.arpa domain name pointer shared4.wirehive.net.

> Perhaps BT have recently changed something that they check against
> and it's a badly trapped error?!

Snip from discussions with Critical Path:


Your email was forwarded to us to research and help resolve.
We (Critical Path) run the mail servers that handle the inbound traffic to
btinternet.com / btopenworld.com / talk21.com
We have also recently taken over handling mail to
[email protected]
- - so if you have problems in the future you should find that email to that
address gets to us.

When we do a reverse DNS check to find the PTR we look up like this
4.234.66.195.IN-ADDR.ARPA this is a valid and common way of checking -
with
the domain name in uppercase.

A while ago CISCO released an update to their IOS firmware that
introduced a
bug. As a consequence, it no longer accepted PTR lookups that used
uppercase
alphabetic characters.
They have subsequently fixed this and have issued updates (e.g. in ASA
9.0(3) in July 2013) - but the users of their equipment might not have
implemented it yet.

So, if you (or your partners) are using CISCO equipment then there is
a good
chance that this is the cause of the problem.

The CISCO bug reference for the ASA 9.x dropping case sensitive DNS PTR
requests is CSCud64817:
http://tools.cisco.com/Support/BugToolKit/search/getBugDetails.do?method=fet
chBugDetails&bugId=CSCud64817&Submit=Search
(but will need a CISCO account to access that)

Cisco list it as fixed in versions:
9.0(3)
9.0(2.10)
100.9(0.19)M
100.8(50.12)M
100.8(34.1)M
100.8(27.30)M
100.8(38.4)M
100.9(4.1)M
100.9(9.1)M
9.0(2.100)


You can see the problem by using a DNS lookup tool - e.g. using "dig" in
Linux shows:
dig 4.234.66.195.in-addr.arpa PTR +short
gold.linx.net.
but
dig 4.234.66.195.IN-ADDR.ARPA PTR
connection timed out; no servers could be reached


Paul Webster
Critical Path
http://www.cp.net

Hope this helps....

- -- 
Marcus Taylor (Database Application Developer)
London Internet Exchange Ltd. 2nd Floor Trinity Court, Trinity Street,
PE1 1DA
Registered England and Wales number 3137929
DDI 01733 207724
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
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=r0j/
-----END PGP SIGNATURE-----

Reply via email to