]> The Tcpdump Group git mirrors - libpcap/commitdiff
Add support for capturing from Linux iptables NFLOG log groups.
authorJakub Zawadzki <[email protected]>
Fri, 17 Jun 2011 03:06:51 +0000 (20:06 -0700)
committerGuy Harris <[email protected]>
Fri, 17 Jun 2011 03:06:51 +0000 (20:06 -0700)
Reviewed-By: Guy Harris <[email protected]>
Makefile.in
config.h.in
configure
configure.in
pcap-linux.c
pcap-netfilter-linux.c [new file with mode: 0644]

index e6cd36965b7481f74cdf867d6bbb153bbb1bb15b..b59a96e1046671102a61cae845f7a001644b6c1a 100644 (file)
@@ -82,7 +82,7 @@ YACC = @V_YACC@
        @rm -f $@
        $(CC) $(FULL_CFLAGS) -c $(srcdir)/$*.c
 
-PSRC = pcap-@[email protected] @USB_SRC@ @BT_SRC@ @CAN_SRC@
+PSRC = pcap-@[email protected] @USB_SRC@ @BT_SRC@ @CAN_SRC@ @NETFILTER_SRC@
 FSRC =  fad-@[email protected]
 SSRC =  @SSRC@
 CSRC = pcap.c inet.c gencode.c optimize.c nametoaddr.c etherent.c \
@@ -299,6 +299,7 @@ EXTRA_DIST = \
        pcap-libdlpi.c \
        pcap-linux.c \
        pcap-namedb.h \
+       pcap-netfilter-linux.c \
        pcap-nit.c \
        pcap-null.c \
        pcap-pf.c \
index c9627b2bfa079902090679c980d2ad22c7ef3ba7..fb5ae09629273efd8d10b95bab5e2392830abbe4 100644 (file)
 /* target host supports CAN sniffing */
 #undef PCAP_SUPPORT_CAN
 
+/* target host supports netfilter sniffing */
+#undef PCAP_SUPPORT_NETFILTER
+
 /* target host supports USB sniffing */
 #undef PCAP_SUPPORT_USB
 
index 4a649a5a7f4858a2580ad4192e3ebc5308c972a0..8af3b65c6142088f9f93f538e267792a0208405d 100755 (executable)
--- a/configure
+++ b/configure
@@ -697,6 +697,8 @@ MAN_FILE_FORMATS
 MAN_MISC_INFO
 PCAP_SUPPORT_USB
 USB_SRC
+PCAP_SUPPORT_NETFILTER
+NETFILTER_SRC
 PCAP_SUPPORT_BT
 BT_SRC
 PCAP_SUPPORT_CAN
@@ -11187,6 +11189,27 @@ esac
 
 
 
+{ echo "$as_me:$LINENO: checking for netfilter sniffing support" >&5
+echo $ECHO_N "checking for netfilter sniffing support... $ECHO_C" >&6; }
+case "$host_os" in
+linux*)
+
+cat >>confdefs.h <<\_ACEOF
+#define PCAP_SUPPORT_NETFILTER 1
+_ACEOF
+
+       NETFILTER_SRC=pcap-netfilter-linux.c
+       { echo "$as_me:$LINENO: result: yes" >&5
+echo "${ECHO_T}yes" >&6; }
+       ;;
+*)
+       { echo "$as_me:$LINENO: result: no" >&5
+echo "${ECHO_T}no" >&6; }
+       ;;
+esac
+
+
+
 # Check whether --enable-bluetooth was given.
 if test "${enable_bluetooth+set}" = set; then
   enableval=$enable_bluetooth;
@@ -12379,6 +12402,8 @@ MAN_FILE_FORMATS!$MAN_FILE_FORMATS$ac_delim
 MAN_MISC_INFO!$MAN_MISC_INFO$ac_delim
 PCAP_SUPPORT_USB!$PCAP_SUPPORT_USB$ac_delim
 USB_SRC!$USB_SRC$ac_delim
+PCAP_SUPPORT_NETFILTER!$PCAP_SUPPORT_NETFILTER$ac_delim
+NETFILTER_SRC!$NETFILTER_SRC$ac_delim
 PCAP_SUPPORT_BT!$PCAP_SUPPORT_BT$ac_delim
 BT_SRC!$BT_SRC$ac_delim
 PCAP_SUPPORT_CAN!$PCAP_SUPPORT_CAN$ac_delim
@@ -12389,7 +12414,7 @@ INSTALL_DATA!$INSTALL_DATA$ac_delim
 LTLIBOBJS!$LTLIBOBJS$ac_delim
 _ACEOF
 
-  if test `sed -n "s/.*$ac_delim\$/X/p" conf$$subs.sed | grep -c X` = 91; then
+  if test `sed -n "s/.*$ac_delim\$/X/p" conf$$subs.sed | grep -c X` = 93; then
     break
   elif $ac_last_try; then
     { { echo "$as_me:$LINENO: error: could not make $CONFIG_STATUS" >&5
index 7a9e446a1b0589874049cdf211c1aee245196547..f5f374eed60899f642b0335b633b807278656871 100644 (file)
@@ -1311,6 +1311,21 @@ esac
 AC_SUBST(PCAP_SUPPORT_USB)
 AC_SUBST(USB_SRC)
 
+dnl check for netfilter sniffing support
+AC_MSG_CHECKING(for netfilter sniffing support)
+case "$host_os" in
+linux*)
+       AC_DEFINE(PCAP_SUPPORT_NETFILTER, 1, [target host supports netfilter sniffing])
+       NETFILTER_SRC=pcap-netfilter-linux.c
+       AC_MSG_RESULT(yes)
+       ;;
+*)
+       AC_MSG_RESULT(no)
+       ;;
+esac
+AC_SUBST(PCAP_SUPPORT_NETFILTER)
+AC_SUBST(NETFILTER_SRC)
+
 AC_ARG_ENABLE([bluetooth],
 [AC_HELP_STRING([--enable-bluetooth],[enable Bluetooth support @<:@default=yes, if support available@:>@])],
 ,enable_bluetooth=yes)
index fa1d028dba9b4a29ac0c1fdd4f10fe4eb2471a1a..a70d585f6c965203e4ab95a9bcb74702891e0826 100644 (file)
@@ -199,6 +199,11 @@ static const char rcsid[] _U_ =
 #include "pcap-can-linux.h"
 #endif
 
+#ifdef PCAP_SUPPORT_NETFILTER
+int netfilter_platform_finddevs(pcap_if_t **alldevsp, char *err_str);
+pcap_t *nflog_create(const char *device, char *ebuf);
+#endif
+
 /*
  * If PF_PACKET is defined, we can use {SOCK_RAW,SOCK_DGRAM}/PF_PACKET
  * sockets rather than SOCK_PACKET sockets.
@@ -419,6 +424,12 @@ pcap_create(const char *device, char *ebuf)
        }
 #endif
 
+#ifdef PCAP_SUPPORT_NETFILTER
+       if (strncmp(device, "nflog", strlen("nflog")) == 0) {
+               return nflog_create(device, ebuf);
+       }
+#endif
+
        handle = pcap_create_common(device, ebuf);
        if (handle == NULL)
                return NULL;
@@ -2199,6 +2210,14 @@ pcap_platform_finddevs(pcap_if_t **alldevsp, char *errbuf)
                return (-1);
 #endif
 
+#ifdef PCAP_SUPPORT_NETFILTER
+       /*
+        * Add netfilter devices.
+        */
+       if (netfilter_platform_finddevs(alldevsp, errbuf) < 0)
+               return (-1);
+#endif
+
        return (0);
 }
 
diff --git a/pcap-netfilter-linux.c b/pcap-netfilter-linux.c
new file mode 100644 (file)
index 0000000..d55b5c3
--- /dev/null
@@ -0,0 +1,443 @@
+/*
+ * Copyright (c) 2011 Jakub Zawadzki
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * 1. Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution.
+ * 3. The name of the author may not be used to endorse or promote 
+ * products derived from this software without specific prior written 
+ * permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include "pcap-int.h"
+
+#ifdef NEED_STRERROR_H
+#include "strerror.h"
+#endif
+
+#include <errno.h>
+#include <stdlib.h>
+#include <unistd.h>
+#include <string.h>
+#include <sys/socket.h>
+#include <arpa/inet.h>
+
+#include <time.h>
+#include <sys/time.h>
+#include <netinet/in.h>
+#include <linux/types.h>
+
+#include <linux/netlink.h>
+#include <linux/netfilter/nfnetlink.h>
+#include <linux/netfilter/nfnetlink_log.h>
+
+#define HDR_LENGTH (NLMSG_LENGTH(NLMSG_ALIGN(sizeof(struct nfgenmsg))))
+
+#define NFLOG_IFACE "nflog"
+
+static int
+nflog_read_linux(pcap_t *handle, int max_packets, pcap_handler callback, u_char *user)
+{
+       const unsigned char *buf;
+       int count = 0;
+       int len;
+
+       /* ignore interrupt system call error */
+       do {
+               len = recv(handle->fd, handle->buffer, handle->bufsize, 0);
+               if (handle->break_loop) {
+                       handle->break_loop = 0;
+                       return -2;
+               }
+       } while ((len == -1) && (errno == EINTR));
+
+       if (len < 0) {
+               snprintf(handle->errbuf, PCAP_ERRBUF_SIZE, "Can't receive packet %d:%s", errno, pcap_strerror(errno));
+               return -1;
+       }
+
+       buf = handle->buffer;
+       while (len >= NLMSG_SPACE(0)) {
+               const struct nlmsghdr *nlh = (const struct nlmsghdr *) buf;
+               u_int32_t msg_len;
+
+               if (nlh->nlmsg_len < sizeof(struct nlmsghdr) || len < nlh->nlmsg_len) {
+                       snprintf(handle->errbuf, PCAP_ERRBUF_SIZE, "Message truncated: (got: %d) (nlmsg_len: %u)", len, nlh->nlmsg_len);
+                       return -1;
+               }
+
+               if (NFNL_SUBSYS_ID(nlh->nlmsg_type) == NFNL_SUBSYS_ULOG && 
+                       NFNL_MSG_TYPE(nlh->nlmsg_type) == NFULNL_MSG_PACKET) 
+               {
+                       const struct nfattr *payload_attr = NULL;
+
+                       if (nlh->nlmsg_len < HDR_LENGTH) {
+                               snprintf(handle->errbuf, PCAP_ERRBUF_SIZE, "Malformed message: (nlmsg_len: %u)", nlh->nlmsg_len);
+                               return -1;
+                       }
+
+                       if (nlh->nlmsg_len > HDR_LENGTH) {
+                               struct nfattr *attr = NFM_NFA(NLMSG_DATA(nlh));
+                               int attr_len = nlh->nlmsg_len - NLMSG_ALIGN(HDR_LENGTH);
+
+                               while (NFA_OK(attr, attr_len)) {
+                                       switch (NFA_TYPE(attr)) {
+                                               case NFULA_PAYLOAD:
+                                                       payload_attr = attr;
+                                                       break;
+                                       }
+                                       attr = NFA_NEXT(attr, attr_len);
+                               }
+                       }
+
+                       if (payload_attr) {
+                               struct pcap_pkthdr pkth;
+
+                               const unsigned char *payload = NFA_DATA(payload_attr);
+                               int payload_len = NFA_PAYLOAD(payload_attr);
+
+                               pkth.len = pkth.caplen = payload_len;
+                               /* pkth.caplen = min (payload_len, handle->snapshot); */
+
+                               gettimeofday(&pkth.ts, NULL);
+                               if (handle->fcode.bf_insns == NULL ||
+                                               bpf_filter(handle->fcode.bf_insns, payload, pkth.len, pkth.caplen)) 
+                               {
+                                       handle->md.packets_read++;
+                                       callback(user, &pkth, payload);
+                                       count++;
+                               }
+                       }
+               }
+
+               msg_len = NLMSG_ALIGN(nlh->nlmsg_len);
+               if (msg_len > len)
+                       msg_len = len;
+
+               len -= msg_len;
+               buf += msg_len;
+       }
+       return count;
+}
+
+static int
+netfilter_stats_linux(pcap_t *handle, struct pcap_stat *stats)
+{
+       stats->ps_recv = handle->md.packets_read;
+       stats->ps_drop = 0;
+       stats->ps_ifdrop = 0;
+       return 0;
+}
+
+static int
+netfilter_inject_linux(pcap_t *handle, const void *buf, size_t size)
+{
+       snprintf(handle->errbuf, PCAP_ERRBUF_SIZE, "inject not supported on netfilter devices");
+       return (-1);
+}                           
+
+struct my_nfattr {
+       u_int16_t nfa_len;
+       u_int16_t nfa_type;
+       void *data;
+};
+
+static int
+nflog_send_config_msg(const pcap_t *handle, u_int8_t family, u_int16_t res_id, const struct my_nfattr *mynfa)
+{
+       char buf[1024] __attribute__ ((aligned));
+
+       struct nlmsghdr *nlh = (struct nlmsghdr *) buf;
+       struct nfgenmsg *nfg = (struct nfgenmsg *) (buf + sizeof(struct nlmsghdr));
+
+       struct sockaddr_nl snl;
+       static unsigned int seq_id;
+       
+       if (!seq_id)
+               seq_id = time(NULL);
+       ++seq_id;
+
+       nlh->nlmsg_len = NLMSG_LENGTH(sizeof(struct nfgenmsg));
+       nlh->nlmsg_type = (NFNL_SUBSYS_ULOG << 8) | NFULNL_MSG_CONFIG;
+       nlh->nlmsg_flags = NLM_F_REQUEST | NLM_F_ACK;
+       nlh->nlmsg_pid = 0;     /* to kernel */
+       nlh->nlmsg_seq = seq_id;
+
+       nfg->nfgen_family = family;
+       nfg->version = NFNETLINK_V0;
+       nfg->res_id = htons(res_id);
+
+       if (mynfa) {
+               struct nfattr *nfa = (struct nfattr *) (buf + NLMSG_ALIGN(nlh->nlmsg_len));
+
+               nfa->nfa_type = mynfa->nfa_type;
+               nfa->nfa_len = NFA_LENGTH(mynfa->nfa_len);
+               memcpy(NFA_DATA(nfa), mynfa->data, mynfa->nfa_len);
+               nlh->nlmsg_len = NLMSG_ALIGN(nlh->nlmsg_len) + NFA_ALIGN(nfa->nfa_len);
+       }
+
+       memset(&snl, 0, sizeof(snl));
+       snl.nl_family = AF_NETLINK;
+
+       if (sendto(handle->fd, nlh, nlh->nlmsg_len, 0, (struct sockaddr *) &snl, sizeof(snl)) == -1)
+               return -1;
+
+       /* waiting for reply loop */
+       do {
+               socklen_t addrlen = sizeof(snl);
+               int len;
+
+               /* ignore interrupt system call error */
+               do {
+                       len = recvfrom(handle->fd, buf, sizeof(buf), 0, (struct sockaddr *) &snl, &addrlen);
+               } while ((len == -1) && (errno == EINTR));
+
+               if (len <= 0)
+                       return len;
+
+               if (addrlen != sizeof(snl) || snl.nl_family != AF_NETLINK) {
+                       errno = EINVAL;
+                       return -1;
+               }
+
+               nlh = (struct nlmsghdr *) buf;
+               if (snl.nl_pid != 0 || seq_id != nlh->nlmsg_seq)        /* if not from kernel or wrong sequence skip */
+                       continue;
+
+               while (len >= NLMSG_SPACE(0) && NLMSG_OK(nlh, len)) {
+                       if (nlh->nlmsg_type == NLMSG_ERROR || (nlh->nlmsg_type == NLMSG_DONE && nlh->nlmsg_flags & NLM_F_MULTI)) {
+                               if (nlh->nlmsg_len < NLMSG_ALIGN(sizeof(struct nlmsgerr))) {
+                                       errno = EBADMSG;
+                                       return -1;
+                               }
+                               errno = -(*((int *)NLMSG_DATA(nlh)));
+                               return (errno == 0) ? 0 : -1;
+                       }
+                       nlh = NLMSG_NEXT(nlh, len);
+               }
+       } while (1);
+
+       return -1; /* never here */
+}
+
+static int
+nflog_send_config_cmd(const pcap_t *handle, u_int16_t group_id, u_int8_t cmd, u_int8_t family)
+{
+       struct nfulnl_msg_config_cmd msg;
+       struct my_nfattr nfa;
+
+       msg.command = cmd;
+
+       nfa.data = &msg;
+       nfa.nfa_type = NFULA_CFG_CMD;
+       nfa.nfa_len = sizeof(msg);
+
+       return nflog_send_config_msg(handle, family, group_id, &nfa);
+}
+
+static int 
+nflog_send_config_mode(const pcap_t *handle, u_int16_t group_id, u_int8_t copy_mode, u_int32_t copy_range)
+{
+       struct nfulnl_msg_config_mode msg;
+       struct my_nfattr nfa;
+
+       msg.copy_range = htonl(copy_range);
+       msg.copy_mode = copy_mode;
+
+       nfa.data = &msg;
+       nfa.nfa_type = NFULA_CFG_MODE;
+       nfa.nfa_len = sizeof(msg);
+
+       return nflog_send_config_msg(handle, AF_UNSPEC, group_id, &nfa);
+}
+
+static int
+nflog_activate(pcap_t* handle)
+{
+       const char *dev = handle->opt.source;
+       unsigned short groups[32];
+       int group_count = 0;
+       int i;
+
+       if (strncmp(dev, NFLOG_IFACE, strlen(NFLOG_IFACE)) == 0) {
+               dev += strlen(NFLOG_IFACE);
+
+               /* nflog:30,33,42 looks nice, allow it */
+               if (*dev == ':')
+                       dev++;
+
+               while (*dev) {
+                       long int group_id;
+                       char *end_dev;
+
+                       if (group_count == 32) {
+                               snprintf(handle->errbuf, PCAP_ERRBUF_SIZE,
+                                               "Maximum 32 netfilter groups! dev: %s", 
+                                               handle->opt.source);
+                               return PCAP_ERROR;
+                       }
+
+                       group_id = strtol(dev, &end_dev, 0);
+                       if (end_dev != dev) {
+                               if (group_id < 0 || group_id > 65535) {
+                                       snprintf(handle->errbuf, PCAP_ERRBUF_SIZE,
+                                                       "Netfilter group range from 0 to 65535 (got %ld)",
+                                                       group_id);
+                                       return PCAP_ERROR;
+                               }
+
+                               groups[group_count++] = (unsigned short) group_id;
+                               dev = end_dev;
+                       }
+                       if (*dev != ',')
+                               break;
+                       dev++;
+               }
+       }
+
+       if (*dev) {
+               snprintf(handle->errbuf, PCAP_ERRBUF_SIZE,
+                               "Can't get netfilter group(s) index from %s", 
+                               handle->opt.source);
+               return PCAP_ERROR;
+       }
+
+       /* if no groups, add default: 0 */
+       if (!group_count) {
+               groups[0] = 0;
+               group_count = 1;
+       }
+
+       /* Initialize some components of the pcap structure. */
+       handle->bufsize = 128 + handle->snapshot;
+       handle->offset = 0;
+       handle->linktype = DLT_IPV4;
+       handle->read_op = nflog_read_linux;
+       handle->inject_op = netfilter_inject_linux;
+       handle->setfilter_op = install_bpf_program; /* no kernel filtering */
+       handle->setdirection_op = NULL;
+       handle->set_datalink_op = NULL;
+       handle->getnonblock_op = pcap_getnonblock_fd;
+       handle->setnonblock_op = pcap_setnonblock_fd;
+       handle->stats_op = netfilter_stats_linux;
+
+       /* Create netlink socket */
+       handle->fd = socket(AF_NETLINK, SOCK_RAW, NETLINK_NETFILTER);
+       if (handle->fd < 0) {
+               snprintf(handle->errbuf, PCAP_ERRBUF_SIZE, "Can't create raw socket %d:%s", errno, pcap_strerror(errno));
+               return PCAP_ERROR;
+       }
+
+       handle->buffer = malloc(handle->bufsize);
+       if (!handle->buffer) {
+               snprintf(handle->errbuf, PCAP_ERRBUF_SIZE, "Can't allocate dump buffer: %s", pcap_strerror(errno));
+               goto close_fail;
+       }
+
+       if (nflog_send_config_cmd(handle, 0, NFULNL_CFG_CMD_PF_UNBIND, AF_INET) < 0) {
+               snprintf(handle->errbuf, PCAP_ERRBUF_SIZE, "NFULNL_CFG_CMD_PF_UNBIND: %s", pcap_strerror(errno));
+               goto close_fail;
+       }
+
+       if (nflog_send_config_cmd(handle, 0, NFULNL_CFG_CMD_PF_BIND, AF_INET) < 0) {
+               snprintf(handle->errbuf, PCAP_ERRBUF_SIZE, "NFULNL_CFG_CMD_PF_BIND: %s", pcap_strerror(errno));
+               goto close_fail;
+       }
+
+       /* Bind socket to the nflog groups */
+       for (i = 0; i < group_count; i++) {
+               if (nflog_send_config_cmd(handle, groups[i], NFULNL_CFG_CMD_BIND, AF_UNSPEC) < 0) {
+                       snprintf(handle->errbuf, PCAP_ERRBUF_SIZE, "Can't listen on group group index: %s", pcap_strerror(errno));
+                       goto close_fail;
+               }
+
+               if (nflog_send_config_mode(handle, groups[i], NFULNL_COPY_PACKET, handle->snapshot) < 0) {
+                       snprintf(handle->errbuf, PCAP_ERRBUF_SIZE, "NFULNL_COPY_PACKET: %s", pcap_strerror(errno));
+                       goto close_fail;
+               }
+       }
+
+       if (handle->opt.rfmon) {
+               /*
+                * Monitor mode doesn't apply to netfilter devices.
+                */
+               pcap_cleanup_live_common(handle);
+               return PCAP_ERROR_RFMON_NOTSUP;
+       }
+
+       if (handle->opt.buffer_size != 0) {
+               /*
+                * Set the socket buffer size to the specified value.
+                */
+               if (setsockopt(handle->fd, SOL_SOCKET, SO_RCVBUF, &handle->opt.buffer_size, sizeof(handle->opt.buffer_size)) == -1) {
+                       snprintf(handle->errbuf, PCAP_ERRBUF_SIZE, "SO_RCVBUF: %s", pcap_strerror(errno));
+                       goto close_fail;
+               }
+       }
+
+       handle->selectable_fd = handle->fd;
+       return 0;
+
+close_fail:
+       pcap_cleanup_live_common(handle);
+       return PCAP_ERROR;
+}
+
+pcap_t *
+nflog_create(const char *device, char *ebuf)
+{
+       pcap_t *p;
+
+       p = pcap_create_common(device, ebuf);
+       if (p == NULL)
+               return (NULL);
+
+       p->activate_op = nflog_activate;
+       return (p);
+}
+
+int 
+netfilter_platform_finddevs(pcap_if_t **alldevsp, char *err_str)
+{
+       pcap_if_t *found_dev = *alldevsp;
+       int sock;
+       
+       sock = socket(AF_NETLINK, SOCK_RAW, NETLINK_NETFILTER);
+       if (sock < 0) {
+               /* if netlink is not supported this this is not fatal */
+               if (errno == EAFNOSUPPORT)
+                       return 0;
+               snprintf(err_str, PCAP_ERRBUF_SIZE, "Can't open netlink socket %d:%s",
+                       errno, pcap_strerror(errno));
+               return -1;
+       }
+       close(sock);
+
+       if (pcap_add_if(&found_dev, NFLOG_IFACE, 0, "Linux netfilter log (NFLOG) interface", err_str) < 0)
+               return -1;
+       return 0;
+}
+