cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
747
Views
9
Helpful
15
Replies

New sendmail exploit signatures

brok3n
Level 1
Level 1

Anyone have a signature for the new remote sendmail exploit?

15 Replies 15

mcerha
Level 3
Level 3

We are currently building the S42 signature update which will contain signature 3115 to deal with this worm. It will hopefully be done this evening. I will post here again when the package is ready.

The S41 package is available for early use at:

ftp://ftpeng.cisco.com/csids-sig-updates/S41/IDS-sig-3.1-3-S41.bin

Signature 3115 covers the Sendmail vulnerability. Mgmt. packages will follow shortly.

Is the string available anywhere else? I am having problems with this link.

Try this instead...

First, goto the following URL:

ftp://ftpeng.cisco.com/csids-sig-updates/S41/

Once there, select either the .bin or .zip file, right-click it and select "Save as..." and download it as you see fit.

I suspect that, like myself, the file is opening in your browser as ASCII text if you follow the link to IDS-sig-3.1-3-S41.bin directly...

Hope this helps out!

This sig is false alarming on valid exchange server email traffic, anyone else experiencing this?

Thx,

brkn!

Why is this triggering the alarm? (names changed to protect the innocent)

SENSOR ID: 100

TIME UTC: 2003/03/05 21:16:13 Local: 2003/03/05 15:16:13

SIGID: 3115 "Sendmail Data Header Overflow" LEVEL: 5

SRC IP:PORT: 10.0.0.1:2926

DST IP:PORT: 10.0.2.10:25

MSG: From: user@host...aaaa...

CONTEXT S->D: 600 (CST)..Received: from unknown(XXX.XXX.XXX.XXX) by megarad.foo.com via csmap (2.0)...id srcAAAaeaioX; Wed, 5 Mar 03 15:15:58 -0600..

Message-Id: <200303052116.h25LG3V25247@megarad.foo.com>..From: "...

¢...." <fruity@lame.com.cn>.

Yes..we are seeing false positives also at SBC DataComm.

pheuch
Level 1
Level 1

The overflow is triggerd by a large number of <> bracket pairs. I think following custom signature will do the trick:

Current Signature: Engine STRING.TCP SIGID 20020

SigName: Sendmail exploit

___________________________________________________________________________

0 - Edit ALL Parameters

1 - AlarmInterval =

2 - AlarmThrottle = FireOnce

3 - ChokeThreshold =

4 - Direction = ToService

5 - FlipAddr =

6 - LimitSummary =

7 - MaxInspectLength =

8 - MinHits = 1

9 - MinMatchLength =

10 - MultipleHits =

11 * RegexString = [\<\>]{100}

12 - ResetAfterIdle = 15

13 - ServicePorts = 25

14 - SigComment =

15 - SigName = Sendmail exploit

16 - SigStringInfo =

17 - StripTelnetOptions =

18 - ThrottleInterval = 15

19 - WantFrag =

d - Delete a value

u - UNDO and continue

x - SAVE and continue

___________________________________________________________________________

Note it could also be triggerd by multiple single brackets <<< or >>>, but packetd was not working with (\<\>){100}

Working with a customer, I have identified at least one false positive in signature 3115. A colon needs to be added to the regex after the To, From, and CC patterns. This will be fixed in sig update S42 for 3.1 sensors. If any one is willing to send traffic samples, you can send them to mcerha@cisco.com, and I'll look at them for you.

Any ETA for S42 -- we need this sig functioning.

thx,

-brkn!

Same here.

And we'd also like to know what is the exact detecting strings for the SigID 3115

on S41 since we are experiencing lots of alerts which looks like benigned.

If you look into the SigWizMenu, it doesn't look like there's a regex string or so.

Thank you.

S42 should be posted to CCO this afternoon.

Are you talking about http://cve.mitre.org/cgi-bin/cvename.cgi?name=CAN-2002-1337?

Anyway last sensor version available in CCO is S40 even if I received Cisco IDS Active Update Bulletin #51 announcing S41. Please help.

Yes, this is the vulnerability. Also, S42 has been on CCO for over a day now.

Getting Started

Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: