Section: C Library Functions (3)
Updated: 21 November 2003
#include <pcap.h>
char errbuf[PCAP_ERRBUF_SIZE];
pcap_t *pcap_open_live(const char *device, int snaplen,
int promisc, int to_ms, char *errbuf)
pcap_t *pcap_open_dead(int linktype, int snaplen)
pcap_t *pcap_open_offline(const char *fname, char *errbuf)
pcap_dumper_t *pcap_dump_open(pcap_t *p, const char *fname)
int pcap_setnonblock(pcap_t *p, int nonblock, char *errbuf);
int pcap_getnonblock(pcap_t *p, char *errbuf);
int pcap_findalldevs(pcap_if_t **alldevsp, char *errbuf)
void pcap_freealldevs(pcap_if_t *alldevs)
char *pcap_lookupdev(char *errbuf)
int pcap_lookupnet(const char *device, bpf_u_int32 *netp,
bpf_u_int32 *maskp, char *errbuf)
int pcap_dispatch(pcap_t *p, int cnt,
pcap_handler callback, u_char *user)
int pcap_loop(pcap_t *p, int cnt,
pcap_handler callback, u_char *user)
void pcap_dump(u_char *user, struct pcap_pkthdr *h,
u_char *sp)
int pcap_compile(pcap_t *p, struct bpf_program *fp,
char *str, int optimize, bpf_u_int32 netmask)
int pcap_setfilter(pcap_t *p, struct bpf_program *fp)
void pcap_freecode(struct bpf_program *);
const u_char *pcap_next(pcap_t *p, struct pcap_pkthdr *h)
int pcap_next_ex(pcap_t *p, struct pcap_pkthdr **pkt_header,
const u_char **pkt_data)
void pcap_breakloop(pcap_t *)
int pcap_datalink(pcap_t *p)
int pcap_list_datalinks(pcap_t *p, int **dlt_buf);
int pcap_set_datalink(pcap_t *p, int dlt);
int pcap_datalink_name_to_val(const char *name);
const char *pcap_datalink_val_to_name(int dlt);
const char *pcap_datalink_val_to_description(int dlt);
int pcap_snapshot(pcap_t *p)
int pcap_is_swapped(pcap_t *p)
int pcap_major_version(pcap_t *p)
int pcap_minor_version(pcap_t *p)
int pcap_stats(pcap_t *p, struct pcap_stat *ps)
FILE *pcap_file(pcap_t *p)
int pcap_fileno(pcap_t *p)
void pcap_perror(pcap_t *p, char *prefix)
char *pcap_geterr(pcap_t *p)
char *pcap_strerror(int error)
const char *pcap_lib_version(void)
void pcap_close(pcap_t *p)
int pcap_dump_flush(pcap_dumper_t *p)
FILE *pcap_dump_file(pcap_dumper_t *p)
void pcap_dump_close(pcap_dumper_t *p)
The Packet Capture library provides a high level interface to packet capture systems. All packets on the network, even those destined for other hosts, are accessible through this mechanism.
NOTE: errbuf in pcap_open_live(), pcap_open_dead(), pcap_open_offline(), pcap_setnonblock(), pcap_getnonblock(), pcap_findalldevs(), pcap_lookupdev(), and pcap_lookupnet() is assumed to be able to hold at least PCAP_ERRBUF_SIZE chars.
pcap_open_live() is used to obtain a packet capture descriptor to look at packets on the network. device is a string that specifies the network device to open; on Linux systems with 2.2 or later kernels, a device argument of "any" or NULL can be used to capture packets from all interfaces. snaplen specifies the maximum number of bytes to capture. If this value is less than the size of a packet that is captured, only the first snaplen bytes of that packet will be captured and provided as packet data. A value of 65535 should be sufficient, on most if not all networks, to capture all the data available from the packet. promisc specifies if the interface is to be put into promiscuous mode. (Note that even if this parameter is false, the interface could well be in promiscuous mode for some other reason.) For now, this doesn't work on the "any" device; if an argument of "any" or NULL is supplied, the promisc flag is ignored. to_ms specifies the read timeout in milliseconds. The read timeout is used to arrange that the read not necessarily return immediately when a packet is seen, but that it wait for some amount of time to allow more packets to arrive and to read multiple packets from the OS kernel in one operation. Not all platforms support a read timeout; on platforms that don't, the read timeout is ignored. A zero value for to_ms, on platforms that support a read timeout, will cause a read to wait forever to allow enough packets to arrive, with no timeout. errbuf is used to return error or warning text. It will be set to error text when pcap_open_live() fails and returns NULL. errbuf may also be set to warning text when pcap_open_live() succeds; to detect this case the caller should store a zero-length string in errbuf before calling pcap_open_live() and display the warning to the user if errbuf is no longer a zero-length string.
pcap_open_dead() is used for creating a pcap_t structure to use when calling the other functions in libpcap. It is typically used when just using libpcap for compiling BPF code.
pcap_open_offline() is called to open a ``savefile'' for reading. fname specifies the name of the file to open. The file has the same format as those used by tcpdump(1) and tcpslice(1). The name "-" in a synonym for stdin. errbuf is used to return error text and is only set when pcap_open_offline() fails and returns NULL.
pcap_dump_open() is called to open a ``savefile'' for writing. The name "-" in a synonym for stdout. NULL is returned on failure. p is a pcap struct as returned by pcap_open_offline() or pcap_open_live(). fname specifies the name of the file to open. If NULL is returned, pcap_geterr() can be used to get the error text.
pcap_setnonblock() puts a capture descriptor, opened with pcap_open_live(), into ``non-blocking'' mode, or takes it out of ``non-blocking'' mode, depending on whether the nonblock argument is non-zero or zero. It has no effect on ``savefiles''. If there is an error, -1 is returned and errbuf is filled in with an appropriate error message; otherwise, 0 is returned. In ``non-blocking'' mode, an attempt to read from the capture descriptor with pcap_dispatch() will, if no packets are currently available to be read, return 0 immediately rather than blocking waiting for packets to arrive. pcap_loop() and pcap_next() will not work in ``non-blocking'' mode.
pcap_getnonblock() returns the current ``non-blocking'' state of the capture descriptor; it always returns 0 on ``savefiles''. If there is an error, -1 is returned and errbuf is filled in with an appropriate error message.
pcap_findalldevs() constructs a list of network devices that can be opened with pcap_open_live(). (Note that there may be network devices that cannot be opened with pcap_open_live() by the process calling pcap_findalldevs(), because, for example, that process might not have sufficient privileges to open them for capturing; if so, those devices will not appear on the list.) alldevsp is set to point to the first element of the list; each element of the list is of type pcap_if_t, and has the following members:
Each element of the list of addresses is of type pcap_addr_t, and has the following members:
-1 is returned on failure, in which case errbuf is filled in with an appropriate error message; 0 is returned on success.
pcap_freealldevs() is used to free a list allocated by pcap_findalldevs().
pcap_lookupdev() returns a pointer to a network device suitable for use with pcap_open_live() and pcap_lookupnet(). If there is an error, NULL is returned and errbuf is filled in with an appropriate error message.
pcap_lookupnet() is used to determine the network number and mask associated with the network device device. Both netp and maskp are bpf_u_int32 pointers. A return of -1 indicates an error in which case errbuf is filled in with an appropriate error message.
pcap_dispatch() is used to collect and process packets. cnt specifies the maximum number of packets to process before returning. This is not a minimum number; when reading a live capture, only one bufferful of packets is read at a time, so fewer than cnt packets may be processed. A cnt of -1 processes all the packets received in one buffer when reading a live capture, or all the packets in the file when reading a ``savefile''. callback specifies a routine to be called with three arguments: a u_char pointer which is passed in from pcap_dispatch(), a const struct pcap_pkthdr pointer to a structure with the following members:
and a const u_char pointer to the first caplen (as given in the struct pcap_pkthdr a pointer to which is passed to the callback routine) bytes of data from the packet (which won't necessarily be the entire packet; to capture the entire packet, you will have to provide a value for snaplen in your call to pcap_open_live() that is sufficiently large to get all of the packet's data - a value of 65535 should be sufficient on most if not all networks).
The number of packets read is returned. 0 is returned if no packets were read from a live capture (if, for example, they were discarded because they didn't pass the packet filter, or if, on platforms that support a read timeout that starts before any packets arrive, the timeout expires before any packets arrive, or if the file descriptor for the capture device is in non-blocking mode and no packets were available to be read) or if no more packets are available in a ``savefile.'' A return of -1 indicates an error in which case pcap_perror() or pcap_geterr() may be used to display the error text. A return of -2 indicates that the loop terminated due to a call to pcap_breakloop() before any packets were processed. If your application uses pcap_breakloop(), make sure that you explicitly check for -1 and -2, rather than just checking for a return value < 0.
NOTE: when reading a live capture, pcap_dispatch() will not necessarily return when the read times out; on some platforms, the read timeout isn't supported, and, on other platforms, the timer doesn't start until at least one packet arrives. This means that the read timeout should NOT be used in, for example, an interactive application, to allow the packet capture loop to ``poll'' for user input periodically, as there's no guarantee that pcap_dispatch() will return after the timeout expires.
pcap_loop() is similar to pcap_dispatch() except it keeps reading packets until cnt packets are processed or an error occurs. It does not return when live read timeouts occur. Rather, specifying a non-zero read timeout to pcap_open_live() and then calling pcap_dispatch() allows the reception and processing of any packets that arrive when the timeout occurs. A negative cnt causes pcap_loop() to loop forever (or at least until an error occurs). -1 is returned on an error; 0 is returned if cnt is exhausted; -2 is returned if the loop terminated due to a call to pcap_breakloop() before any packets were processed. If your application uses pcap_breakloop(), make sure that you explicitly check for -1 and -2, rather than just checking for a return value < 0.
pcap_next() reads the next packet (by calling pcap_dispatch() with a cnt of 1) and returns a u_char pointer to the data in that packet. (The pcap_pkthdr struct for that packet is not supplied.) NULL is returned if an error occured, or if no packets were read from a live capture (if, for example, they were discarded because they didn't pass the packet filter, or if, on platforms that support a read timeout that starts before any packets arrive, the timeout expires before any packets arrive, or if the file descriptor for the capture device is in non-blocking mode and no packets were available to be read), or if no more packets are available in a ``savefile.'' Unfortunately, there is no way to determine whether an error occured or not.
pcap_next_ex() reads the next packet and returns a success/failure indication:
If the packet was read without problems, the pointer pointed to by the pkt_header argument is set to point to the pcap_pkthdr struct for the packet, and the pointer pointed to by the pkt_data argument is set to point to the data in the packet.
pcap_breakloop() sets a flag that will force pcap_dispatch() or pcap_loop() to return rather than looping; they will return the number of packets that have been processed so far, or -2 if no packets have been processed so far.
This routine is safe to use inside a signal handler on UNIX or a console control handler on Windows, as it merely sets a flag that is checked within the loop.
The flag is checked in loops reading packets from the OS - a signal by itself will not necessarily terminate those loops - as well as in loops processing a set of packets returned by the OS. Note that if you are catching signals on UNIX systems that support restarting system calls after a signal, and calling pcap_breakloop() in the signal handler, you must specify, when catching those signals, that system calls should NOT be restarted by that signal. Otherwise, if the signal interrupted a call reading packets in a live capture, when your signal handler returns after calling pcap_breakloop(), the call will be restarted, and the loop will not terminate until more packets arrive and the call completes.
Note that pcap_next() will, on some platforms, loop reading packets from the OS; that loop will not necessarily be terminated by a signal, so pcap_breakloop() should be used to terminate packet processing even if pcap_next() is being used.
pcap_breakloop() does not guarantee that no further packets will be processed by pcap_dispatch() or pcap_loop() after it is called; at most one more packet might be processed.
If -2 is returned from pcap_dispatch() or pcap_loop(), the flag is cleared, so a subsequent call will resume reading packets. If a positive number is returned, the flag is not cleared, so a subsequent call will return -2 and clear the flag.
pcap_dump() outputs a packet to the ``savefile'' opened with pcap_dump_open(). Note that its calling arguments are suitable for use with pcap_dispatch() or pcap_loop(). If called directly, the user parameter is of type pcap_dumper_t as returned by pcap_dump_open().
pcap_compile() is used to compile the string str into a filter program. program is a pointer to a bpf_program struct and is filled in by pcap_compile(). optimize controls whether optimization on the resulting code is performed. netmask specifies the IPv4 netmask of the network on which packets are being captured; it is used only when checking for IPv4 broadcast addresses in the filter program. If the netmask of the network on which packets are being captured isn't known to the program, or if packets are being captured on the Linux "any" pseudo-interface that can capture on more than one network, a value of 0 can be supplied; tests for IPv4 broadcast addreses won't be done correctly, but all other tests in the filter program will be OK. A return of -1 indicates an error in which case pcap_geterr() may be used to display the error text.
pcap_compile_nopcap() is similar to pcap_compile() except that instead of passing a pcap structure, one passes the snaplen and linktype explicitly. It is intended to be used for compiling filters for direct BPF usage, without necessarily having called pcap_open(). A return of -1 indicates an error; the error text is unavailable. (pcap_compile_nopcap() is a wrapper around pcap_open_dead(), pcap_compile(), and pcap_close(); the latter three routines can be used directly in order to get the error text for a compilation error.)
pcap_setfilter() is used to specify a filter program. fp is a pointer to a bpf_program struct, usually the result of a call to pcap_compile(). -1 is returned on failure, in which case pcap_geterr() may be used to display the error text; 0 is returned on success.
pcap_freecode() is used to free up allocated memory pointed to by a bpf_program struct generated by pcap_compile() when that BPF program is no longer needed, for example after it has been made the filter program for a pcap structure by a call to pcap_setfilter().
pcap_datalink() returns the link layer type; link layer types it can return include:
a 1-byte flag, which is 0 for packets received by the machine and 1 for packets sent by the machine;
a 1-byte field, the upper 4 bits of which indicate the type of packet, as per RFC 1144:
for UNCOMPRESSED_TCP, the rest of the modified IP header, and for COMPRESSED_TCP, the compressed TCP/IP datagram header;
for a total of 16 bytes; the uncompressed IP datagram follows the header.
a 2-byte "packet type", in network byte order, which is one of:
a 2-byte field, in network byte order, containing a Linux ARPHRD_ value for the link layer device type;
a 2-byte field, in network byte order, containing the length of the link layer address of the sender of the packet (which could be 0);
an 8-byte field containing that number of bytes of the link layer header (if there are more than 8 bytes, only the first 8 are present);
a 2-byte field containing an Ethernet protocol type, in network byte order, or containing 1 for Novell 802.3 frames without an 802.2 LLC header or 4 for frames beginning with an 802.2 LLC header.
a 4-byte PF_ value, in network byte order;
a 16-character interface name;
a 2-byte rule number, in network byte order;
a 2-byte reason code, in network byte order, which is one of:
a 2-byte action code, in network byte order, which is one of:
a 2-byte direction, in network byte order, which is one of:
a 1-byte flag field, containing a direction flag in the uppermost bit, which is set for packets transmitted by the machine and clear for packets received by the machine, and a 4-byte traffic type in the low-order 4 bits, which is one of:
a 1-byte VPI value;
a 2-byte VCI field, in network byte order.
pcap_list_datalinks() is used to get a list of the supported data link types of the interface associated with the pcap descriptor. pcap_list_datalinks() allocates an array to hold the list and sets *dlt_buf. The caller is responsible for freeing the array. -1 is returned on failure; otherwise, the number of data link types in the array is returned.
pcap_set_datalink() is used to set the current data link type of the pcap descriptor to the type specified by dlt. -1 is returned on failure.
pcap_datalink_name_to_val() translates a data link type name, which is a DLT_ name with the DLT_ removed, to the corresponding data link type value. The translation is case-insensitive. -1 is returned on failure.
pcap_datalink_val_to_name() translates a data link type value to the corresponding data link type name. NULL is returned on failure.
pcap_datalink_val_to_description() translates a data link type value to a short description of that data link type. NULL is returned on failure.
pcap_snapshot() returns the snapshot length specified when pcap_open_live() was called.
pcap_is_swapped() returns true if the current ``savefile'' uses a different byte order than the current system.
pcap_major_version() returns the major number of the file format of the savefile; pcap_minor_version() returns the minor number of the file format of the savefile. The version number is stored in the header of the savefile.
pcap_file() returns the standard I/O stream of the ``savefile,'' if a ``savefile'' was opened with pcap_open_offline(), or NULL, if a network device was opened with pcap_open_live().
pcap_stats() returns 0 and fills in a pcap_stat struct. The values represent packet statistics from the start of the run to the time of the call. If there is an error or the underlying packet capture doesn't support packet statistics, -1 is returned and the error text can be obtained with pcap_perror() or pcap_geterr(). pcap_stats() is supported only on live captures, not on ``savefiles''; no statistics are stored in ``savefiles'', so no statistics are available when reading from a ``savefile''.
pcap_fileno() returns the file descriptor number from which captured packets are read, if a network device was opened with pcap_open_live(), or -1, if a ``savefile'' was opened with pcap_open_offline().
pcap_perror() prints the text of the last pcap library error on stderr, prefixed by prefix.
pcap_geterr() returns the error text pertaining to the last pcap library error. NOTE: the pointer it returns will no longer point to a valid error message string after the pcap_t passed to it is closed; you must use or copy the string before closing the pcap_t.
pcap_strerror() is provided in case strerror(1) isn't available.
pcap_lib_version() returns a pointer to a string giving information about the version of the libpcap library being used; note that it contains more information than just a version number.
pcap_close() closes the files associated with p and deallocates resources.
pcap_dump_file() returns the standard I/O stream of the ``savefile'' opened by pcap_dump_open().
pcap_dump_flush() flushes the output buffer to the ``savefile,'' so that any packets written with pcap_dump() but not yet written to the ``savefile'' will be written. -1 is returned on error, 0 on success.
pcap_dump_close() closes the ``savefile.''
The original authors are:
Van Jacobson, Craig Leres and Steven McCanne, all of the Lawrence Berkeley National Laboratory, University of California, Berkeley, CA.
The current version is available from "The Tcpdump Group"'s Web site at