Ethical Hacking Learn to find vulnerabilities before the bad guys do! Gain real world hands on hacking experience in our state of the art hacking lab. Course designed and taught by expert instructors with years of penetration testing experience. 12 student maximum in every class. Certification attempt included in every package. | Computer Forensics Training at InfoSec Institute Gain the in-demand skills of a certified computer examiner, learn to recover trace data left behind by fraud, theft, and cybercrime perpetrators. Discover the source of computer crime and abuse at your organization so that it never happens again. All of our class sizes are guaranteed to be 12 students or less to facilitate one-on-one interaction with one of our expert instructors. |

| Subject: | Re: [Snort-sigs] Sig 1147 |
|---|---|
| Date: | Wed, 26 Apr 2006 21:56:15 -0400 |
BassPlayer wrote:
Sorry my mod_security killed my message. Resending.... My point exactly but instead of not using the rule, which seems to be the default response when I ask a bout a rule, why not make it better so it traps only on the actual exploit.
The exploit referenced is but one of a class of command injections. It really serves to provide information that someone is poking around playing with command injection. Hopefully it gives you a chance to respond before they find a real command injection.
With my modded guardian script I can ignore idividual sids but then I still see the alerts in my BASE console and they still take up log and DB space.
Any updates made to the rule would eliminate many other potential exploitation vectors. This is not the only cgi to have ever been vulnerable nor is maintaining a list of every cgi a practical thing to do. The %20 is even a compromise as some scripts are invoked by shells that still allow the use of IFS in insecure ways. Other scripts will replace specific chars with a space in a misguided attempt to make them safe so the exploit string becomes something like ;cat#/path/to/file The rule is about as close as you can get to actual detection of the class of vulnerabilities in a generic sense without being overly loose in it's application. For these classes of vulnerabilities there is little that can be done to ensure perfect coverage short of writing a rule for every exploit opportunity. There are other rules that are very similar in applicability such as ones that look for id, rm... but you would have to look for every potential command that can be executed in every potential way to detect the class of vulnerabilities all the time every time. If you are confident that every cgi on your server is free of command injection then turn it off. If you are not then leave it on and use pass rules to handle _your_ _known_ false positive cases. Under rare circumstance would it ever make sense to automatically block on a rule such as this. Regardless of what you do any change to the rule will have an affect on it's effectiveness in one direction or the other favoring false positives or negatives.
I guess I should of been more concise and asked if there was a way to get the rule updated. What is the process for doing that?
There are several ways to ask for a change. One such way is exactly what you have done, open a discussion on list. We monitor the lists and are always looking for feedback and suggestions and it is great to see new people getting involved and asking questions; often with great suggestions resulting from the discussion. Other ways to ask for a change are: - Send an email to snort-team@sourcefire.com - Submit a new rule as a suggested replacement to the appropriate rules category @ http://www.snort.org/reg-bin/rulesubmit.cgi - Catch us on freenode in #snort
It would also be helpful if someone could tell me if the rule syntax would acutally work. Thanks BP BassPlayer wrote:Jamie Riden wrote:Hi BP, There are lots of other circumstances in which you don't want a 'cat' command, e.g. with the awstats exploit, people will use cat/echo/id to test if a script is vulnerable. Something like this: GET /cgi-bin/?configdir=%7cecho%20%3becho%20b_exp%3bcat%20%3a%etc%3apasswd%20e_exp%3b%2500 HTTP/1.1" However, the cat%20 rule does tend to generate a lot of false positives - for example at the vet department of the university I used to work at :) I wouldn't recommend blocking using it. cheers, Jamie On 26/04/06, BassPlayer <bassplayer@angmar.com> wrote:After checking the actual exploit here http://www.securityfocus.com/bid/374/exploit Wouln't it be better to do alert tcp $EXTERNAL_NET any -> $HTTP_SERVERS $HTTP_PORTS (msg:"WEB-MISC cat%20 access"; flow:to_server,established; (pcre:"/webdist.cgi.+cat%20/i";) nocase; reference:bugtraq,374; reference:cve,1999-0039; classtype:attempted-recon; sid:1147; rev:7;) Please excuse my n00bness in rules writing. BP BassPlayer wrote:Hi, Can this rule be tightened up a bit? alert tcp $EXTERNAL_NET any -> $HTTP_SERVERS $HTTP_PORTS(msg:"WEB-MISCcat%20 access"; flow:to_server,established; content:"cat%20"; nocase; reference:bugtraq,374; reference:cve,1999-0039;classtype:attempted-recon;sid:1147; rev:7;) It's triggering on the following request which is my wife streamingmusic.She wasn't too happy when my modified guardian script autoblocked her:D.BP Generated by BASE v1.2.2 (cindy) on Tue, 25 Apr 2006 15:18:24 -0700 ------------------------------------------------------------------------------ #(1 - 113755) [2006-04-25 07:35:06] [cve/1999-0039] [icat/1999-0039] [bugtraq/374] [local/1147] [snort/1147] WEB-MISC cat%20 access IPv4: 143.183.121.1 -> 209.237.15.226 hlen=5 TOS=0 dlen=517 ID=43136 flags=0 offset=0 TTL=47chksum=46826TCP: port=56372 -> dport: 80 flags=***AP*** seq=2968053703 ack=1992497194 off=8 res=0 win=5840 urp=0 chksum=57923 Options: #1 - NOP len=0 #2 - NOP len=0 #3 - TS len=8 data=A0754FBD005A583E Payload: GET /private_music_archive/play/index.php?song=2538&uid=usersid=sid&ds=32&name=/The%20Pussycat%20Dolls%20-%20Bite%20the%20Dust.mp3 HTTP/1.0 Accept: */* User-Agent: Windows-Media-Player/10.00.00.3990 Host: www.angmar.com Cookie: amp_longsess=1; POSTNUKESID=mumble Via: 1.0 scfwpr01.sc.intel.com:911 (squid/2.5.STABLE12) X-Forwarded-For: unknown Cache-Control: max-age=259200 Connection: keep-alive ------------------------------------------------------- Using Tomcat but need to do more? Need to support web services,security?Get stuff done quickly with pre-integrated technology to make yourjobeasier Download IBM WebSphere Application Server v.1.0.1 based on ApacheGeronimohttp://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 _______________________________________________ Snort-sigs mailing list Snort-sigs@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/snort-sigs-- Jamie Riden / jamesr@europe.com / jamie.riden@computer.org "Microsoft: Bringing the world to your desktop - and your desktop to the world." -- Peter Gutmann------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 _______________________________________________ Snort-sigs mailing list Snort-sigs@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/snort-sigs !DSPAM:444edc89215666362979185!------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 _______________________________________________ Snort-sigs mailing list Snort-sigs@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/snort-sigs
-- Jason Brvenik - Sourcefire PGP: 89C6 DE77 3B32 FC03 A5AE B5DD 11DF 4C8B 0D8E 3383 Key: http://cerberus.sourcefire.com/~jbrvenik/jason.brvenik.pgp.key ------------------------------------------------------- All the advantages of Linux Managed Hosting--Without the Cost and Risk! Fully trained technicians. The highest number of Red Hat certifications in the hosting industry. Fanatical Support. Click to learn more http://sel.as-us.falkag.net/sel?cmd=lnk&kid=107521&bid=248729&dat=121642 _______________________________________________ Snort-sigs mailing list Snort-sigs@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/snort-sigs
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: [Snort-sigs] correction and new rule about keyloggers, Nigel Houghton |
|---|---|
| Next by Date: | [Snort-sigs] False positive for sid 1478, Brock, Anthony - NET |
| Previous by Thread: | Re: [Snort-sigs] Sig 1147, Jamie Riden |
| Next by Thread: | Re: [Snort-sigs] Sig 1147, Jamie Riden |
| Indexes: | [Date] [Thread] [Top] [All Lists] |