9 The Software Porting And Archive Centre for HP-UX, at
11 http://hpux.csc.liv.ac.uk/
13 (and with mirrors in various countries, listed on the Centre's home
14 page) has ported versions, in both source and binary form, for Ethereal,
15 as well as for the "libpcap", GLib, GTK+, "zlib", and CMU SNMP libraries
18 The changes they've made appear largely to be compile option changes; if
19 you've downloaded the source to the latest version of Ethereal (the
20 version on the Centre's site may not necessarily be the latest version),
21 it should be able to compile, perhaps with those changes.
23 They appear to have used HP-UX's "cc" compiler, with the options "-Ae
24 -O"; there's a comment "Add -Dhpux_9 if building under 9.X". It may
29 nettl is used on HP-UX to trace various streams based subsystems. Ethereal
30 can read nettl files containing IP frames (NS_LS_IP subsystem) and LAPB
31 frames (SX25L2 subsystem).
32 It has been tested with files generated on HP-UX 9.04 and 10.20.
34 Use the following commands to generate a trace (cf. nettl(1M)):
36 # IP capture. 0x30000000 means PDU in and PDU out :
37 nettl -tn 0x30000000 -e NS_LS_IP -f tracefile
38 # X25 capture. You must specify an interface :
39 nettl -tn 0x30000000 -e SX25l2 -d /dev/x25_0 -f tracefile
40 # stop capture. subsystem is NS_LS_IP or SX25L2 :
41 nettl -tf -e subsystem
43 One may be able to specify "-tn pduin pduout" rather than
44 "-tn 0x30000000"; the nettl man page for HP-UX 10.30 implies that it
47 3 - "libpcap" on HP-UX
49 If you want to use Ethereal to capture packets, you will have to install
50 "libpcap"; the INSTALL file for "libpcap" has several comments about
51 HP-UX, which you should read if you're going to install and use
54 Another note, from a mail message to the "ethereal-users" list:
56 Date: Wed, 22 Dec 1999 09:05:47 -0600 (EST)
57 From: Gerald Combs <gerald@zing.org>
58 To: Lothar Seitter <lothar.seitter@arcormail.de>
59 cc: ethereal-users@zing.org
60 Subject: Re: [ethereal-users] permission problem with capturing
62 On Wed, 22 Dec 1999, Lothar Seitter wrote:
64 > running 'ethereal' under HP-UX 11 with root permission and
65 > /dev/lan0 set to 777, I always get the message:
66 > "There are no network interfaces that can be opened.
67 > Please to make sure you have sufficient permission to
70 > I start ethereal with 'etheral -i lan0' and lan0 is definitely
73 > What am I missing???
75 You may need to reference the card's DLPI device directly. We were having
76 trouble getting Ethereal to capture on an HP-UX 10.20 machine here. I
77 found an article on Deja News that says:
79 "To access a particular interface, you would say "tcpdump -i /dev/dlpiN"
80 where N is the PPA of the interface you wish to use. You get the PPA by
81 looking at the output of lanscan. On 10.20, it is the same value as the
82 NMID. On 11.X, it is the Card Instance number."
84 This didn't help in our case, but it might in yours. The full article is
85 at http://x34.deja.com/[ST_rn=ps]/getdoc.xp?AN=549366486 .
87 Another article by the same author mentions that experimental versions of
88 libpcap and tcpdump are available at
89 ftp://ftp.cup.hp.com/dist/networking/tools/ . The article itself is at
90 http://x34.deja.com/[ST_rn=ps]/getdoc.xp?AN=558665378 .
92 The first of those articles also says:
94 BTW, before you have to make a follow-up post, you will find that
95 unless you have the latest lan common/DLPI/driver patches installed,
96 you will _not_ see the system's own outbound traffic.
98 It appears that a consequence of the fact that HP-UX's DLPI doesn't work
99 like Solaris's, in that, on Solaris, to get at the device "hme0", say,
100 "libpcap" has to open "/dev/hme" and then tell it to use the 0th
101 interface, whilst on HP-UX you have to go through "/dev/dlpi", you won't
102 get a list of interfaces in the dialog box for "Capture:Start" - you'll
103 have to do through the aforementioned song and dance to find the PPA of
104 the interface you want to use, and supply the "dlpiN" name by hand (I
105 think you can omit the "/dev/" in both tcpdump and Ethereal).
107 Here is a patch to "pcap-dlpi.c" in libpcap that, at least on HP-UX
108 11.X, allows the name of the network interface, rather than the "dlpiN"
109 name, to be specified to tcpdump and Ethereal. It has not been tried on
110 HP-UX 10.20; it fixes one bug that could have caused the code in vanilla
111 libpcap not to correctly find the PPA for an interface on HP-UX 10.20,
112 but HP-UX 10.20's DLPI doesn't supply, in the data returned by a
113 DL_HP_PPA_REQ request, network interface names, so, on systems without
114 the network interface names in that data, the code continues to check
115 the major device number.
117 On HP-UX 11.00, the patch allows a network interface to be specified by
118 name as an argument to tcpdump, rather than requiring that you specify a
119 "dlpiN" name (and it should work equally well with Ethereal).
121 If you try this code on HP-UX 10.20, and it doesn't let you specify the
122 interface by name, please send mail to ethereal-dev@zing.org, so that we
123 know that it didn't work - we'll probably send you debugging patches in
124 the hopes of being able to make it work on 10.20 as well.
126 Here's the patch (to vanilla libpcap 0.4; it patches "configure",
127 "aclocal.m4", and "configure.in", to make the configure script check
128 whether your version of HP-UX supplies the interface names in the reply
129 to a DL_HP_PPA_REQ request, and "pcap-dlpi.c"):
131 *** ../libpcap-0.4/configure Sat Jul 25 12:41:51 1998
132 --- configure Fri Jan 14 00:40:04 2000
138 echo $ac_n "checking if unaligned accesses fail""... $ac_c" 1>&6
139 ! echo "configure:1707: checking if unaligned accesses fail" >&5
140 if eval "test \"`echo '$''{'ac_cv_lbl_unaligned_fail'+set}'`\" = set"; then
141 echo $ac_n "(cached) $ac_c" 1>&6
147 + echo $ac_n "checking if dl_hp_ppa_info_t struct has dl_module_id_1 member""... $ac_c" 1>&6
148 + echo "configure:1707: checking if dl_hp_ppa_info_t struct has dl_module_id_1 member" >&5
149 + if eval "test \"`echo '$''{'ac_cv_lbl_dl_hp_ppa_info_t_has_dl_module_id_1'+set}'`\" = set"; then
150 + echo $ac_n "(cached) $ac_c" 1>&6
152 + cat > conftest.$ac_ext <<EOF
153 + #line 1712 "configure"
154 + #include "confdefs.h"
156 + # include <sys/types.h>
157 + # include <sys/dlpi.h>
158 + # include <sys/dlpi_ext.h>
160 + u_int i = sizeof(((dl_hp_ppa_info_t *)0)->dl_module_id_1)
163 + if { (eval echo configure:1722: \"$ac_compile\") 1>&5; (eval $ac_compile) 2>&5; }; then
165 + ac_cv_lbl_dl_hp_ppa_info_t_has_dl_module_id_1=yes
167 + echo "configure: failed program was:" >&5
168 + cat conftest.$ac_ext >&5
170 + ac_cv_lbl_dl_hp_ppa_info_t_has_dl_module_id_1=no
175 + echo "$ac_t""$ac_cv_lbl_dl_hp_ppa_info_t_has_dl_module_id_1" 1>&6
176 + if test $ac_cv_lbl_dl_hp_ppa_info_t_has_dl_module_id_1 = yes ; then
177 + cat >> confdefs.h <<\EOF
178 + #define HAVE_HP_PPA_INFO_T_DL_MODULE_ID_1 1
183 echo $ac_n "checking if unaligned accesses fail""... $ac_c" 1>&6
184 ! echo "configure:1743: checking if unaligned accesses fail" >&5
185 if eval "test \"`echo '$''{'ac_cv_lbl_unaligned_fail'+set}'`\" = set"; then
186 echo $ac_n "(cached) $ac_c" 1>&6
190 # SVR4 /usr/ucb/install, which tries to use the nonexistent group "staff"
191 # ./install, which can be erroneously created by make from ./install.sh.
192 echo $ac_n "checking for a BSD compatible install""... $ac_c" 1>&6
193 ! echo "configure:1803: checking for a BSD compatible install" >&5
194 if test -z "$INSTALL"; then
195 if eval "test \"`echo '$''{'ac_cv_path_install'+set}'`\" = set"; then
196 echo $ac_n "(cached) $ac_c" 1>&6
198 # SVR4 /usr/ucb/install, which tries to use the nonexistent group "staff"
199 # ./install, which can be erroneously created by make from ./install.sh.
200 echo $ac_n "checking for a BSD compatible install""... $ac_c" 1>&6
201 ! echo "configure:1839: checking for a BSD compatible install" >&5
202 if test -z "$INSTALL"; then
203 if eval "test \"`echo '$''{'ac_cv_path_install'+set}'`\" = set"; then
204 echo $ac_n "(cached) $ac_c" 1>&6
207 *** ../libpcap-0.4/aclocal.m4 Fri Jun 12 03:45:15 1998
208 --- aclocal.m4 Tue Jan 4 21:02:13 2000
215 + dnl Checks to see if the dl_hp_ppa_info_t struct has the HP-UX 11.00
216 + dnl dl_module_id_1 member
220 + dnl AC_LBL_HP_PPA_INFO_T_DL_MODULE_ID_1
224 + dnl HAVE_HP_PPA_INFO_T_DL_MODULE_ID_1 (defined)
226 + dnl NOTE: any compile failure means we conclude that it doesn't have
227 + dnl that member, so if we don't have DLPI, don't have a <sys/dlpi_ext.h>
228 + dnl header, or have one that doesn't declare a dl_hp_ppa_info_t type,
229 + dnl we conclude it doesn't have that member (which is OK, as either we
230 + dnl won't be using code that would use that member, or we wouldn't
231 + dnl compile in any case).
233 + AC_DEFUN(AC_LBL_HP_PPA_INFO_T_DL_MODULE_ID_1,
234 + [AC_MSG_CHECKING(if dl_hp_ppa_info_t struct has dl_module_id_1 member)
235 + AC_CACHE_VAL(ac_cv_lbl_dl_hp_ppa_info_t_has_dl_module_id_1,
237 + # include <sys/types.h>
238 + # include <sys/dlpi.h>
239 + # include <sys/dlpi_ext.h>],
240 + [u_int i = sizeof(((dl_hp_ppa_info_t *)0)->dl_module_id_1)],
241 + ac_cv_lbl_dl_hp_ppa_info_t_has_dl_module_id_1=yes,
242 + ac_cv_lbl_dl_hp_ppa_info_t_has_dl_module_id_1=no))
243 + AC_MSG_RESULT($ac_cv_lbl_dl_hp_ppa_info_t_has_dl_module_id_1)
244 + if test $ac_cv_lbl_dl_hp_ppa_info_t_has_dl_module_id_1 = yes ; then
245 + AC_DEFINE(HAVE_HP_PPA_INFO_T_DL_MODULE_ID_1)
249 dnl Checks to see if -R is used
254 *** ../libpcap-0.4/configure.in Sun Jul 27 22:16:22 1997
255 --- configure.in Tue Jan 4 21:02:13 2000
260 AC_LBL_SOCKADDR_SA_LEN
262 + AC_LBL_HP_PPA_INFO_T_DL_MODULE_ID_1
264 AC_LBL_UNALIGNED_ACCESS
266 if test "${srcdir}" = "." ; then
268 *** ../libpcap-0.4/pcap-dlpi.c Wed Oct 15 21:59:34 1997
269 --- pcap-dlpi.c Tue Jan 4 21:02:13 2000
273 memset(p, 0, sizeof(*p));
276 ! ** Determine device and ppa
278 ! cp = strpbrk(device, "0123456789");
280 sprintf(ebuf, "%s missing unit number", device);
284 memset(p, 0, sizeof(*p));
286 + #ifdef HAVE_DEV_DLPI
288 ! ** Remove any "/dev/" on the front of the device.
290 ! cp = strrchr(device, '/');
298 ! * Split the name into a device type and a unit number.
300 ! cp = strpbrk(dname, "0123456789");
302 sprintf(ebuf, "%s missing unit number", device);
306 sprintf(ebuf, "%s bad unit number", device);
310 ! if (*device == '/')
311 ! strcpy(dname, device);
313 ! sprintf(dname, "%s/%s", PCAP_DEV_PREFIX, device);
314 ! #ifdef HAVE_DEV_DLPI
315 ! /* Map network device to /dev/dlpi unit */
317 if ((p->fd = open(cp, O_RDWR)) < 0) {
318 sprintf(ebuf, "%s: %s", cp, pcap_strerror(errno));
321 ! /* Map network interface to /dev/dlpi unit */
322 ppa = get_dlpi_ppa(p->fd, dname, ppa, ebuf);
326 /* Try device without unit number */
327 strcpy(dname2, dname);
328 cp = strchr(dname, *cp);
330 sprintf(ebuf, "%s bad unit number", device);
336 ! * Use "/dev/dlpi" as the device.
338 ! * XXX - HP's DLPI Programmer's Guide for HP-UX 11.00 says that
339 ! * the "dl_mjr_num" field is for the "major number of interface
340 ! * driver"; that's the major of "/dev/dlpi" on the system on
341 ! * which I tried this, but there may be DLPI devices that
342 ! * use a different driver, in which case we may need to
343 ! * search "/dev" for the appropriate device with that major
344 ! * device number, rather than hardwiring "/dev/dlpi".
346 ! * I'm assuming that the code that was used for HP-UX 10.x
347 ! * is valid, and therefore that, in 10.x, DLPIable devices have
348 ! * "/dev" entries corresponding to them and that their
349 ! * major device number is what appears in the "dl_hp_ppa_info_t"
350 ! * structure for that device. We can search by name in 11.x
351 ! * (and we may have to, as it's not clear that DLPIable devices
352 ! * will have those sorts of "/dev" entries), but we can't in
353 ! * 10.x, as 10.x doesn't have the "dl_module_id_1" and
354 ! * "dl_module_id_2" members in the "dl_hp_ppa_info_t" structure.
357 if ((p->fd = open(cp, O_RDWR)) < 0) {
358 sprintf(ebuf, "%s: %s", cp, pcap_strerror(errno));
363 ! * Get a table of all PPAs for that device, and search that
364 ! * table for the specified device type name and unit number.
366 ppa = get_dlpi_ppa(p->fd, dname, ppa, ebuf);
371 + ** Determine device and ppa
373 + cp = strpbrk(device, "0123456789");
375 + sprintf(ebuf, "%s missing unit number", device);
378 + ppa = strtol(cp, &eos, 10);
379 + if (*eos != '\0') {
380 + sprintf(ebuf, "%s bad unit number", device);
384 + if (*device == '/')
385 + strcpy(dname, device);
387 + sprintf(dname, "%s/%s", PCAP_DEV_PREFIX, device);
389 /* Try device without unit number */
390 strcpy(dname2, dname);
391 cp = strchr(dname, *cp);
397 ! sprintf(ebuf, "unknown mac type 0x%lu", infop->dl_mac_type);
405 ! sprintf(ebuf, "unknown mac type 0x%lu",
406 ! (unsigned long)infop->dl_mac_type);
413 #ifdef DL_HP_PPA_ACK_OBS
415 ! * Under HP-UX 10, we can ask for the ppa
421 #ifdef DL_HP_PPA_ACK_OBS
423 ! * Under HP-UX 10 and HP-UX 11, we can ask for the ppa
431 register dl_hp_ppa_ack_t *ap;
432 ! register dl_hp_ppa_info_t *ip;
434 register u_long majdev;
435 - dl_hp_ppa_req_t req;
437 bpf_u_int32 buf[MAXDLBUF];
439 ! if (stat(device, &statbuf) < 0) {
440 ! sprintf(ebuf, "stat: %s: %s", device, pcap_strerror(errno));
443 majdev = major(statbuf.st_rdev);
445 memset((char *)&req, 0, sizeof(req));
446 req.dl_primitive = DL_HP_PPA_REQ;
450 register dl_hp_ppa_ack_t *ap;
451 ! register dl_hp_ppa_info_t *ipstart, *ip;
453 + #ifndef HAVE_HP_PPA_INFO_T_DL_MODULE_ID_1
455 register u_long majdev;
458 + dl_hp_ppa_req_t req;
459 bpf_u_int32 buf[MAXDLBUF];
461 ! #ifndef HAVE_HP_PPA_INFO_T_DL_MODULE_ID_1
463 ! * HP-UX 10 doesn't put the interface name in the
464 ! * "dl_hp_ppa_info_t" structure, so we have to
465 ! * try to get the major device number for the device
466 ! * corresponding to the device and unit name provided to
467 ! * us, and search for the entry with that major device number.
469 ! sprintf(dname, "/dev/%s%d", device, unit);
470 ! if (stat(dname, &statbuf) < 0) {
471 ! sprintf(ebuf, "stat: %s: %s", dname, pcap_strerror(errno));
474 majdev = major(statbuf.st_rdev);
477 memset((char *)&req, 0, sizeof(req));
478 req.dl_primitive = DL_HP_PPA_REQ;
483 ap = (dl_hp_ppa_ack_t *)buf;
484 ! ip = (dl_hp_ppa_info_t *)((u_char *)ap + ap->dl_offset);
486 for(i = 0; i < ap->dl_count; i++) {
487 ! if (ip->dl_mjr_num == majdev && ip->dl_instance_num == unit)
490 ! ip = (dl_hp_ppa_info_t *)((u_char *)ip + ip->dl_next_offset);
492 if (i == ap->dl_count) {
493 ! sprintf(ebuf, "can't find PPA for %s", device);
496 if (ip->dl_hdw_state == HDW_DEAD) {
497 ! sprintf(ebuf, "%s: hardware state: DOWN\n", device);
500 return ((int)ip->dl_ppa);
504 ap = (dl_hp_ppa_ack_t *)buf;
505 ! ipstart = (dl_hp_ppa_info_t *)((u_char *)ap + ap->dl_offset);
508 for(i = 0; i < ap->dl_count; i++) {
509 ! #ifdef HAVE_HP_PPA_INFO_T_DL_MODULE_ID_1
510 ! if ((strcmp(ip->dl_module_id_1, device) == 0 ||
511 ! strcmp(ip->dl_module_id_2, device) == 0) &&
513 ! if (ip->dl_mjr_num == majdev &&
515 ! ip->dl_instance_num == unit)
518 ! ip = (dl_hp_ppa_info_t *)((u_char *)ipstart + ip->dl_next_offset);
520 if (i == ap->dl_count) {
521 ! sprintf(ebuf, "can't find PPA for %s%d", device, unit);
524 if (ip->dl_hdw_state == HDW_DEAD) {
525 ! sprintf(ebuf, "%s%d: hardware state: DOWN\n", device, unit);
528 return ((int)ip->dl_ppa);
534 ! char if_name[sizeof(ifnet.if_name)], tifname[32];
536 cp = strrchr(ifname, '/');
542 ! char if_name[sizeof(ifnet.if_name) + 1];
544 cp = strrchr(ifname, '/');
548 if (dlpi_kread(kd, (off_t)addr,
549 &ifnet, sizeof(ifnet), ebuf) < 0 ||
550 dlpi_kread(kd, (off_t)ifnet.if_name,
551 ! if_name, sizeof(if_name), ebuf) < 0) {
555 ! sprintf(tifname, "%.*s%d",
556 ! (int)sizeof(if_name), if_name, ifnet.if_unit);
557 ! if (strcmp(tifname, ifname) == 0)
558 return (ifnet.if_index);
562 if (dlpi_kread(kd, (off_t)addr,
563 &ifnet, sizeof(ifnet), ebuf) < 0 ||
564 dlpi_kread(kd, (off_t)ifnet.if_name,
565 ! if_name, sizeof(ifnet.if_name), ebuf) < 0) {
569 ! if_name[sizeof(ifnet.if_name)] = '\0';
570 ! if (strcmp(if_name, ifname) == 0 && ifnet.if_unit == unit)
571 return (ifnet.if_index);