From patchwork Thu Mar 26 14:14:05 2009 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Evgeniy Polyakov X-Patchwork-Id: 25156 X-Patchwork-Delegate: davem@davemloft.net Return-Path: X-Original-To: patchwork-incoming@ozlabs.org Delivered-To: patchwork-incoming@ozlabs.org Received: from vger.kernel.org (vger.kernel.org [209.132.176.167]) by ozlabs.org (Postfix) with ESMTP id 65922DDDA1 for ; Fri, 27 Mar 2009 01:14:46 +1100 (EST) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756035AbZCZOOP (ORCPT ); Thu, 26 Mar 2009 10:14:15 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1754617AbZCZOOO (ORCPT ); Thu, 26 Mar 2009 10:14:14 -0400 Received: from broadrack.ru ([195.178.208.66]:48921 "EHLO tservice.net.ru" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750996AbZCZOOL (ORCPT ); Thu, 26 Mar 2009 10:14:11 -0400 Received: by tservice.net.ru (Postfix, from userid 1000) id C2E97FE4A; Thu, 26 Mar 2009 17:14:05 +0300 (MSK) Date: Thu, 26 Mar 2009 17:14:05 +0300 From: Evgeniy Polyakov To: Patrick McHardy Cc: netdev@vger.kernel.org, David Miller , "Paul E. McKenney" , Netfilter Development Mailinglist , Jan Engelhardt Subject: Passive OS fingerprint xtables match. Message-ID: <20090326141405.GA25404@ioremap.net> MIME-Version: 1.0 Content-Disposition: inline User-Agent: Mutt/1.5.13 (2006-08-11) Sender: netdev-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: netdev@vger.kernel.org Hi. Passive OS fingerprinting netfilter module allows to passively detect remote OS and perform various netfilter actions based on that knowledge. This module compares some data (WS, MSS, options and it's order, ttl, df and others) from packets with SYN bit set with dynamically loaded OS fingerprints. Fingerprint matching rules can be downloaded from OpenBSD source tree and loaded via netfilter netlink subsystem into the kernel via special util found in archive. Archive contains library file (also attached), which was shipped with iptables extensions some time ago (at least when ipt_osf existed in patch-o-matic). This release moves library build to the 1.4.3 features and adds nfnetlink support for the module configuration. Also I incorporated all comments which were rised during the previous sends. Fingerprints can be downloaded from http://www.openbsd.org/cgi-bin/cvsweb/src/etc/pf.os Example usage: # modrpobe xt_osf # ./nfnl_osf -f ./pf.os -d switch removes fingerprints # iptables -I INPUT -j ACCEPT -p tcp -m osf --genre Linux --log 0 --ttl 2 You will find something like this in the syslog: Windows [2000:SP3:Windows XP Pro SP1, 2000 SP3]: 11.22.33.55:4024 -> 11.22.33.44:139 hops=4 Linux [2.5:] : 1.2.3.4:44448 -> 11.22.33.44:22 hops=4 Please consider for inclusion. Thank you. Passive OS fingerprint homepage (archives, examples): http://www.ioremap.net/projects/osf And a small foot note: netfilter guys, you have really, Really, REALLY weird feel of the reality since porting whatever else to the netfilter netlink code requires non-trivial amount of god-blessed substances processed by the brain of the poor human. Signed-off-by: Evgeniy Polyakov diff --git a/include/linux/netfilter/nfnetlink.h b/include/linux/netfilter/nfnetlink.h index 7d8e045..71babbd 100644 --- a/include/linux/netfilter/nfnetlink.h +++ b/include/linux/netfilter/nfnetlink.h @@ -46,7 +46,8 @@ struct nfgenmsg { #define NFNL_SUBSYS_CTNETLINK_EXP 2 #define NFNL_SUBSYS_QUEUE 3 #define NFNL_SUBSYS_ULOG 4 -#define NFNL_SUBSYS_COUNT 5 +#define NFNL_SUBSYS_OSF 5 +#define NFNL_SUBSYS_COUNT 6 #ifdef __KERNEL__ diff --git a/include/linux/netfilter/xt_osf.h b/include/linux/netfilter/xt_osf.h new file mode 100644 index 0000000..11903a9 --- /dev/null +++ b/include/linux/netfilter/xt_osf.h @@ -0,0 +1,119 @@ +/* + * Copyright (c) 2003+ Evgeniy Polyakov + * + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + */ + +#ifndef _XT_OSF_H +#define _XT_OSF_H + +#define MAXGENRELEN 32 +#define MAXDETLEN 64 + +#define XT_OSF_GENRE (1<<0) +#define XT_OSF_TTL (1<<1) +#define XT_OSF_LOG (1<<2) +#define XT_OSF_UNUSED (1<<3) +#define XT_OSF_CONNECTOR (1<<4) +#define XT_OSF_INVERT (1<<5) + +#define XT_OSF_LOGLEVEL_ALL 0 +#define XT_OSF_LOGLEVEL_FIRST 1 +#define XT_OSF_LOGLEVEL_ALL_KNOWN 2 + +#define XT_OSF_TTL_TRUE 0 /* True ip and fingerprint TTL comparison */ +#define XT_OSF_TTL_LESS 1 /* Check if ip TTL is less than fingerprint one */ +#define XT_OSF_TTL_NOCHECK 2 /* Do not compare ip and fingerprint TTL at all */ + +struct xt_osf_info { + char genre[MAXGENRELEN]; + __u32 len; + __u32 flags; + __u32 loglevel; + __u32 ttl; +}; + +/* + * Wildcard MSS (kind of). + * It is used to implement a state machine for the different wildcard values + * of the MSS and window sizes. + */ +struct xt_osf_wc { + __u32 wc; + __u32 val; +}; + +/* + * This struct represents IANA options + * http://www.iana.org/assignments/tcp-parameters + */ +struct xt_osf_opt { + __u16 kind, length; + struct xt_osf_wc wc; +}; + +struct xt_osf_user_finger { + struct xt_osf_wc wss; + + __u8 ttl, df; + __u16 ss, mss; + __u16 opt_num; + + char genre[MAXGENRELEN]; + char version[MAXGENRELEN]; + char subtype[MAXGENRELEN]; + + /* MAX_IPOPTLEN is maximum if all options are NOPs or EOLs */ + struct xt_osf_opt opt[MAX_IPOPTLEN]; +}; + +struct xt_osf_nlmsg { + struct xt_osf_user_finger f; + struct iphdr ip; + struct tcphdr tcp; +}; + +/* Defines for IANA option kinds */ + +enum iana_options { + OSFOPT_EOL = 0, /* End of options */ + OSFOPT_NOP, /* NOP */ + OSFOPT_MSS, /* Maximum segment size */ + OSFOPT_WSO, /* Window scale option */ + OSFOPT_SACKP, /* SACK permitted */ + OSFOPT_SACK, /* SACK */ + OSFOPT_ECHO, + OSFOPT_ECHOREPLY, + OSFOPT_TS, /* Timestamp option */ + OSFOPT_POCP, /* Partial Order Connection Permitted */ + OSFOPT_POSP, /* Partial Order Service Profile */ + + /* Others are not used in the current OSF */ + OSFOPT_EMPTY = 255, +}; + +enum xt_osf_msg_types { + OSF_MSG_SETUP, + OSF_MSG_MAX, +}; + +enum xt_osf_attr_type { + OSF_ATTR_UNSPEC, + OSF_ATTR_FINGER, + OSF_ATTR_MAX, +}; + +#endif /* _XT_OSF_H */ diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index c2bac9c..1e74c97 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -853,6 +853,19 @@ config NETFILTER_XT_MATCH_U32 Details and examples are in the kernel module source. +config NETFILTER_XT_MATCH_OSF + tristate '"osf" Passive OS fingerprint match' + depends on NETFILTER_ADVANCED + help + This option selects the Passive OS Fingerprinting match module + that allows to passively match the remote operating system by + analyzing incoming TCP SYN packets. + + Rules and loading software can be downloaded from + http://www.ioremap.net/projects/osf + + To compile it as a module, choose M here. If unsure, say N. + endif # NETFILTER_XTABLES endmenu diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index da3d909..96b7754 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -73,6 +73,7 @@ obj-$(CONFIG_NETFILTER_XT_MATCH_LIMIT) += xt_limit.o obj-$(CONFIG_NETFILTER_XT_MATCH_MAC) += xt_mac.o obj-$(CONFIG_NETFILTER_XT_MATCH_MARK) += xt_mark.o obj-$(CONFIG_NETFILTER_XT_MATCH_MULTIPORT) += xt_multiport.o +obj-$(CONFIG_NETFILTER_XT_MATCH_OSF) += xt_osf.o obj-$(CONFIG_NETFILTER_XT_MATCH_OWNER) += xt_owner.o obj-$(CONFIG_NETFILTER_XT_MATCH_PHYSDEV) += xt_physdev.o obj-$(CONFIG_NETFILTER_XT_MATCH_PKTTYPE) += xt_pkttype.o diff --git a/net/netfilter/xt_osf.c b/net/netfilter/xt_osf.c new file mode 100644 index 0000000..9c15cb2 --- /dev/null +++ b/net/netfilter/xt_osf.c @@ -0,0 +1,503 @@ +/* + * Copyright (c) 2003+ Evgeniy Polyakov + * + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + */ + +#include +#include + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include + +#include +#include +#include + +struct xt_osf_finger { + struct rcu_head rcu_head; + struct list_head finger_entry; + struct xt_osf_user_finger finger; +}; + +enum osf_fmatch_states { + /* Packet does not match the fingerprint */ + FMATCH_WRONG = 0, + /* Packet matches the fingerprint */ + FMATCH_OK, + /* Options do not match the fingerprint, but header does */ + FMATCH_OPT_WRONG, +}; + +struct xt_osf_finger_storage +{ + struct list_head finger_list; + spinlock_t finger_lock; +}; + +/* + * Indexed by dont-fragment bit. + * It is the only constant value in the fingerprint. + */ +struct xt_osf_finger_storage xt_osf_fingers[2]; + +struct xt_osf_message { + struct cn_msg cmsg; + struct xt_osf_nlmsg nlmsg; +}; + +static const struct nla_policy xt_osf_policy[OSF_ATTR_MAX + 1] = { + [OSF_ATTR_FINGER] = { .len = sizeof(struct xt_osf_user_finger) }, +}; + +static void xt_osf_finger_free_rcu(struct rcu_head *rcu_head) +{ + struct xt_osf_finger *f = container_of(rcu_head, struct xt_osf_finger, rcu_head); + + kfree(f); +} + +static int xt_osf_setup_callback(struct sock *ctnl, struct sk_buff *skb, + struct nlmsghdr *nlh, struct nlattr *osf_attrs[]) +{ + struct xt_osf_user_finger *f; + struct nfgenmsg *nfmsg = NLMSG_DATA(nlh); + u16 delete = ntohs(nfmsg->res_id); + struct xt_osf_finger *kf = NULL, *sf; + struct xt_osf_finger_storage *st; + int err; + + if (!osf_attrs[OSF_ATTR_FINGER]) + return -EINVAL; + + f = nla_data(osf_attrs[OSF_ATTR_FINGER]); + st = &xt_osf_fingers[!!f->df]; + + /* + * If 'delete' is set to 0 then we add attached fingerprint, + * otherwise remove, and in this case we do not need to allocate data. + */ + if (!delete) { + kf = kmalloc(sizeof(struct xt_osf_finger), GFP_KERNEL); + if (!kf) + return -ENOMEM; + + memcpy(&kf->finger, f, sizeof(struct xt_osf_user_finger)); + } + + err = -ENOENT; + + rcu_read_lock(); + list_for_each_entry_rcu(sf, &st->finger_list, finger_entry) { + if (memcmp(&sf->finger, f, sizeof(struct xt_osf_user_finger))) + continue; + + if (delete) { + spin_lock_bh(&st->finger_lock); + list_del_rcu(&sf->finger_entry); + spin_unlock_bh(&st->finger_lock); + call_rcu(&sf->rcu_head, xt_osf_finger_free_rcu); + } else { + kfree(kf); + kf = NULL; + } + + err = 0; + break; + } + + if (kf) { + spin_lock_bh(&st->finger_lock); + list_add_tail_rcu(&kf->finger_entry, &st->finger_list); + spin_unlock_bh(&st->finger_lock); +#if 0 + printk(KERN_INFO "Added rule for %s:%s:%s.\n", + kf->finger.genre, kf->finger.version, kf->finger.subtype); +#endif + } + rcu_read_unlock(); + + return 0; +} + +static const struct nfnl_callback xt_osf_nfnetlink_callbacks[OSF_MSG_MAX] = { + [OSF_MSG_SETUP] = { + .call = xt_osf_setup_callback, + .attr_count = OSF_ATTR_MAX, + .policy = xt_osf_policy, + }, +}; + +static const struct nfnetlink_subsystem xt_osf_nfnetlink = { + .name = "osf", + .subsys_id = NFNL_SUBSYS_OSF, + .cb_count = OSF_MSG_MAX, + .cb = xt_osf_nfnetlink_callbacks, +}; + +static void xt_osf_send_connector(const struct xt_osf_user_finger *f, + const struct sk_buff *skb) +{ +#if 1 + return; +#else + struct xt_osf_message *msg = &per_cpu(xt_osf_mbuf, smp_processor_id()); + struct xt_osf_nlmsg *data = &msg->nlmsg; + struct tcphdr *tcp; + + + memcpy(&data->f, f, sizeof(struct xt_osf_user_finger)); + memcpy(&data->ip, ip_hdr(skb), sizeof(struct iphdr)); + tcp = skb_header_pointer(skb, ip_hdrlen(skb), sizeof(struct tcphdr), &data->tcp); + if (tcp) + memcpy(&data->tcp, tcp, sizeof(struct tcphdr)); + + cn_netlink_send(&msg->cmsg, CN_IDX_OSF, GFP_ATOMIC); +#endif +} + +static inline int xt_osf_ttl(const struct sk_buff *skb, const struct xt_osf_info *info, + unsigned char f_ttl) +{ + const struct iphdr *ip = ip_hdr(skb); + + if (info->flags & XT_OSF_TTL) { + if (info->ttl == XT_OSF_TTL_TRUE) + return ip->ttl == f_ttl; + if (info->ttl == XT_OSF_TTL_NOCHECK) + return 1; + else if (ip->ttl <= f_ttl) + return 1; + else { + struct in_device *in_dev = in_dev_get(skb->dev); + int ret = 0; + + for_ifa(in_dev) { + if (inet_ifa_match(ip->saddr, ifa)) { + ret = (ip->ttl == f_ttl); + break; + } + } + endfor_ifa(in_dev); + + in_dev_put(in_dev); + return ret; + } + } + + return ip->ttl == f_ttl; +} + +static bool xt_osf_match_packet(const struct sk_buff *skb, + const struct xt_match_param *p) +{ + const struct xt_osf_info *info = p->matchinfo; + const struct iphdr *ip = ip_hdr(skb); + const struct tcphdr *tcp; + struct tcphdr _tcph; + int fmatch = FMATCH_WRONG, fcount = 0; + unsigned int optsize = 0, check_WSS = 0; + u16 window, totlen, mss = 0; + bool df; + const unsigned char *optp = NULL, *_optp = NULL; + unsigned char opts[MAX_IPOPTLEN]; + const struct xt_osf_finger *kf; + const struct xt_osf_user_finger *f; + const struct xt_osf_finger_storage *st; + + if (!info) + return false; + + tcp = skb_header_pointer(skb, ip_hdrlen(skb), sizeof(struct tcphdr), &_tcph); + if (!tcp) + return false; + + if (!tcp->syn) + return false; + + totlen = ntohs(ip->tot_len); + df = ntohs(ip->frag_off) & IP_DF; + window = ntohs(tcp->window); + + if (tcp->doff * 4 > sizeof(struct tcphdr)) { + optsize = tcp->doff * 4 - sizeof(struct tcphdr); + + if (optsize > sizeof(opts)) + optsize = sizeof(opts); + + _optp = optp = skb_header_pointer(skb, ip_hdrlen(skb) + sizeof(struct tcphdr), + optsize, opts); + } + + st = &xt_osf_fingers[df]; + + rcu_read_lock(); + list_for_each_entry_rcu(kf, &st->finger_list, finger_entry) { + f = &kf->finger; + + if (!(info->flags & XT_OSF_LOG) && strcmp(info->genre, f->genre)) + continue; + + optp = _optp; + fmatch = FMATCH_WRONG; + + if (totlen == f->ss && xt_osf_ttl(skb, info, f->ttl)) { + int foptsize, optnum; + + check_WSS = 0; + + switch (f->wss.wc) { + case 0: + check_WSS = 0; + break; + case 'S': + check_WSS = 1; + break; + case 'T': + check_WSS = 2; + break; + case '%': + check_WSS = 3; + break; + default: + check_WSS = 4; + break; + } + if (check_WSS == 4) + continue; + + /* Check options */ + + foptsize = 0; + for (optnum = 0; optnum < f->opt_num; ++optnum) + foptsize += f->opt[optnum].length; + + if (foptsize > MAX_IPOPTLEN || optsize > MAX_IPOPTLEN || optsize != foptsize) + continue; + + for (optnum = 0; optnum < f->opt_num; ++optnum) { + if (f->opt[optnum].kind == (*optp)) { + __u32 len = f->opt[optnum].length; + const __u8 *optend = optp + len; + int loop_cont = 0; + + fmatch = FMATCH_OK; + + switch (*optp) { + case OSFOPT_MSS: + mss = optp[3]; + mss <<= 8; + mss |= optp[2]; + + mss = ntohs(mss); + break; + case OSFOPT_TS: + loop_cont = 1; + break; + } + + optp = optend; + } else + fmatch = FMATCH_OPT_WRONG; + + if (fmatch != FMATCH_OK) + break; + } + + if (fmatch != FMATCH_OPT_WRONG) { + fmatch = FMATCH_WRONG; + + switch (check_WSS) { + case 0: + if (f->wss.val == 0 || window == f->wss.val) + fmatch = FMATCH_OK; + break; + case 1: /* MSS */ +#define SMART_MSS_1 1460 +#define SMART_MSS_2 1448 + if (window == f->wss.val * mss || + window == f->wss.val * SMART_MSS_1 || + window == f->wss.val * SMART_MSS_2) + fmatch = FMATCH_OK; + break; + case 2: /* MTU */ + if (window == f->wss.val * (mss + 40) || + window == f->wss.val * (SMART_MSS_1 + 40) || + window == f->wss.val * (SMART_MSS_2 + 40)) + fmatch = FMATCH_OK; + break; + case 3: /* MOD */ + if ((window % f->wss.val) == 0) + fmatch = FMATCH_OK; + break; + } + } + + if (fmatch != FMATCH_OK) + continue; + + fcount++; + if (info->flags & XT_OSF_LOG) + printk(KERN_INFO "%s [%s:%s] : " + "%pi4:%d -> %pi4:%d hops=%d\n", + f->genre, f->version, f->subtype, + &ip->saddr, ntohs(tcp->source), + &ip->daddr, ntohs(tcp->dest), + f->ttl - ip->ttl); + + if (info->flags & XT_OSF_CONNECTOR) + xt_osf_send_connector(f, skb); + + if ((info->flags & XT_OSF_LOG) && + info->loglevel == XT_OSF_LOGLEVEL_FIRST) + break; + } + } + rcu_read_unlock(); + + if (!fcount && (info->flags & (XT_OSF_LOG | XT_OSF_CONNECTOR))) { + unsigned int i; + struct xt_osf_user_finger fg; + + memset(&fg, 0, sizeof(fg)); +#if 1 + if (info->flags & XT_OSF_LOG) { + if (info->loglevel != XT_OSF_LOGLEVEL_ALL_KNOWN) + printk(KERN_INFO "Unknown: win: %u, mss: %u, " + "totlen: %u, df: %d, ttl: %u : ", + window, mss, totlen, df, ip->ttl); + else + printk(KERN_INFO ""); + if (_optp) { + optp = _optp; + for (i = 0; i < optsize; i++) + printk("%02X ", optp[i]); + } + + printk("%pi4:%u -> %pi4:%u\n", + &ip->saddr, ntohs(tcp->source), + &ip->daddr, ntohs(tcp->dest)); + } +#endif + if (info->flags & XT_OSF_CONNECTOR) { + fg.wss.val = window; + fg.ttl = ip->ttl; + fg.df = df; + fg.ss = totlen; + fg.mss = mss; + strncpy(fg.genre, "Unknown", MAXGENRELEN); + + xt_osf_send_connector(&fg, skb); + } + } + + if (fcount) + fmatch = FMATCH_OK; + + return fmatch == FMATCH_OK; +} + +static struct xt_match xt_osf_match = { + .name = "osf", + .revision = 0, + .family = NFPROTO_IPV4, + .proto = IPPROTO_TCP, + .hooks = (1 << NF_INET_LOCAL_IN) | (1 << NF_INET_PRE_ROUTING) | (1 << NF_INET_FORWARD), + .match = xt_osf_match_packet, + .matchsize = sizeof(struct xt_osf_info), + .me = THIS_MODULE, +}; + +static int __init xt_osf_init(void) +{ + int err = -EINVAL; + int i; + + for (i=0; ifinger_list); + spin_lock_init(&st->finger_lock); + } + + err = nfnetlink_subsys_register(&xt_osf_nfnetlink); + if (err < 0) { + printk(KERN_ERR "Failed (%d) to register OSF nsfnetlink helper.\n", err); + goto err_out_exit; + } + + err = xt_register_match(&xt_osf_match); + if (err) { + printk(KERN_ERR "Failed (%d) to register OS fingerprint " + "matching module.\n", err); + goto err_out_remove; + } + + printk(KERN_INFO "Started passive OS fingerprint matching module.\n"); + + return 0; + +err_out_remove: + nfnetlink_subsys_unregister(&xt_osf_nfnetlink); +err_out_exit: + return err; +} + +static void __exit xt_osf_fini(void) +{ + struct xt_osf_finger *f; + int i; + + nfnetlink_subsys_unregister(&xt_osf_nfnetlink); + xt_unregister_match(&xt_osf_match); + + rcu_read_lock(); + for (i=0; ifinger_list, finger_entry) { + list_del_rcu(&f->finger_entry); + call_rcu(&f->rcu_head, xt_osf_finger_free_rcu); + } + } + rcu_read_unlock(); + + rcu_barrier(); + + printk(KERN_INFO "Passive OS fingerprint matching module finished.\n"); +} + +module_init(xt_osf_init); +module_exit(xt_osf_fini); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Evgeniy Polyakov "); +MODULE_DESCRIPTION("Passive OS fingerprint matching."); +MODULE_ALIAS_NFNL_SUBSYS(NFNL_SUBSYS_OSF);