On Tue, Nov 30, 2010 at 10:12 PM, Michael Shigorin <mike@osdn.org.ua>= wrote:

On Tue, Nov 30, 2010 at 11:07:08PM +0100, Bal?zs M?t?ffy = wrote:
> Version of iftop is 0.17 on FreeBSD 8.1-RELEASE-p1 with mpd v5.5.

Maybe it's worth trying just released 1.0-pre1 first?

http://www.ex-parrot.com/~pdw/iftop/download/iftop-1.0= pre1.tar.gz

fyi, 1.0-pre1 doesn't bui= ld on FreeBSD 8.1 (or at least on my box):

make =A0all-recursive

Making all in config

if gcc -DHAVE_CONFIG_H -I. -I. -I. =A0 =A0 =A0-g -O2 -MT addr_hash.= o -MD -MP -MF ".deps/addr_hash.Tpo" -c -o addr_hash.o addr_hash.c= ; =A0then mv -f ".deps/addr_hash.Tpo" ".deps/addr_hash.Po&qu= ot;; else rm -f ".deps/addr_hash.Tpo"; exit 1; fi

addr_hash.c: In function 'hash':

addr_hash.c:46: err= or: 'struct in6_addr' has no member named 's6_addr32'

addr_hash.c:54: error: 'struct in6_addr' has no member named &= #39;s6_addr32'

*** Error code 1

full output atta= ched.

=A0

--
=A0---- WBR, Michael Shigorin <mike@= altlinux.ru>
=A0------ Linux.Kiev http://www.linux.kiev.ua/

_______________________________________________
iftop-users mailing list
iftop-users@lists.beasts.or= g
http://lists.beasts.org/mailman/listinfo/iftop-users

--0016368336f87efa1a0496738c65-- --0016368336f87efa240496738c67 Content-Type: application/octet-stream; name="fbsd81.out" Content-Disposition: attachment; filename="fbsd81.out" Content-Transfer-Encoding: base64 X-Attachment-Id: f_gh84ae7u0 JCB1bmFtZSAtYQpGcmVlQlNEIGJsYWguYWxvb2Yub3JnIDguMS1SRUxFQVNFIEZyZWVCU0QgOC4x LVJFTEVBU0UgIzA6IE1vbiBKdWwgMTkgMDI6MzY6NDkgVVRDIDIwMTAgICAgIHJvb3RAbWFzb24u Y3NlLmJ1ZmZhbG8uZWR1Oi91c3Ivb2JqL3Vzci9zcmMvc3lzL0dFTkVSSUMgIGFtZDY0CgokIC4v Y29uZmlndXJlICYmIG1ha2UKY2hlY2tpbmcgYnVpbGQgc3lzdGVtIHR5cGUuLi4gYW1kNjQtdW5r bm93bi1mcmVlYnNkOC4xCmNoZWNraW5nIGhvc3Qgc3lzdGVtIHR5cGUuLi4gYW1kNjQtdW5rbm93 bi1mcmVlYnNkOC4xCmNoZWNraW5nIHRhcmdldCBzeXN0ZW0gdHlwZS4uLiBhbWQ2NC11bmtub3du LWZyZWVic2Q4LjEKY2hlY2tpbmcgZm9yIGEgQlNELWNvbXBhdGlibGUgaW5zdGFsbC4uLiAvdXNy L2Jpbi9pbnN0YWxsIC1jCmNoZWNraW5nIHdoZXRoZXIgYnVpbGQgZW52aXJvbm1lbnQgaXMgc2Fu ZS4uLiB5ZXMKY2hlY2tpbmcgZm9yIGdhd2suLi4gbm8KY2hlY2tpbmcgZm9yIG1hd2suLi4gbm8K Y2hlY2tpbmcgZm9yIG5hd2suLi4gbmF3awpjaGVja2luZyB3aGV0aGVyIG1ha2Ugc2V0cyAkKE1B S0UpLi4uIHllcwpjaGVja2luZyBmb3IgZ2NjLi4uIGdjYwpjaGVja2luZyB3aGV0aGVyIHRoZSBD IGNvbXBpbGVyIHdvcmtzLi4uIHllcwpjaGVja2luZyBmb3IgQyBjb21waWxlciBkZWZhdWx0IG91 dHB1dCBmaWxlIG5hbWUuLi4gYS5vdXQKY2hlY2tpbmcgZm9yIHN1ZmZpeCBvZiBleGVjdXRhYmxl cy4uLiAKY2hlY2tpbmcgd2hldGhlciB3ZSBhcmUgY3Jvc3MgY29tcGlsaW5nLi4uIG5vCmNoZWNr aW5nIGZvciBzdWZmaXggb2Ygb2JqZWN0IGZpbGVzLi4uIG8KY2hlY2tpbmcgd2hldGhlciB3ZSBh cmUgdXNpbmcgdGhlIEdOVSBDIGNvbXBpbGVyLi4uIHllcwpjaGVja2luZyB3aGV0aGVyIGdjYyBh Y2NlcHRzIC1nLi4uIHllcwpjaGVja2luZyBmb3IgZ2NjIG9wdGlvbiB0byBhY2NlcHQgSVNPIEM4 OS4uLiBub25lIG5lZWRlZApjaGVja2luZyBmb3Igc3R5bGUgb2YgaW5jbHVkZSB1c2VkIGJ5IG1h a2UuLi4gR05VCmNoZWNraW5nIGRlcGVuZGVuY3kgc3R5bGUgb2YgZ2NjLi4uIGdjYzMKY2hlY2tp bmcgaG93IHRvIHJ1biB0aGUgQyBwcmVwcm9jZXNzb3IuLi4gZ2NjIC1FCmNoZWNraW5nIGZvciBn cmVwIHRoYXQgaGFuZGxlcyBsb25nIGxpbmVzIGFuZCAtZS4uLiAvdXNyL2Jpbi9ncmVwCmNoZWNr aW5nIGZvciBlZ3JlcC4uLiAvdXNyL2Jpbi9ncmVwIC1FCmNoZWNraW5nIGZvciBBTlNJIEMgaGVh ZGVyIGZpbGVzLi4uIHllcwpjaGVja2luZyBmb3Igc3lzL3R5cGVzLmguLi4geWVzCmNoZWNraW5n IGZvciBzeXMvc3RhdC5oLi4uIHllcwpjaGVja2luZyBmb3Igc3RkbGliLmguLi4geWVzCmNoZWNr aW5nIGZvciBzdHJpbmcuaC4uLiB5ZXMKY2hlY2tpbmcgZm9yIG1lbW9yeS5oLi4uIHllcwpjaGVj a2luZyBmb3Igc3RyaW5ncy5oLi4uIHllcwpjaGVja2luZyBmb3IgaW50dHlwZXMuaC4uLiB5ZXMK Y2hlY2tpbmcgZm9yIHN0ZGludC5oLi4uIHllcwpjaGVja2luZyBmb3IgdW5pc3RkLmguLi4geWVz CmNoZWNraW5nIHN5cy9pb2N0bC5oIHVzYWJpbGl0eS4uLiB5ZXMKY2hlY2tpbmcgc3lzL2lvY3Rs LmggcHJlc2VuY2UuLi4geWVzCmNoZWNraW5nIGZvciBzeXMvaW9jdGwuaC4uLiB5ZXMKY2hlY2tp bmcgc3lzL3RpbWUuaCB1c2FiaWxpdHkuLi4geWVzCmNoZWNraW5nIHN5cy90aW1lLmggcHJlc2Vu Y2UuLi4geWVzCmNoZWNraW5nIGZvciBzeXMvdGltZS5oLi4uIHllcwpjaGVja2luZyBzeXMvc29j a2lvLmggdXNhYmlsaXR5Li4uIHllcwpjaGVja2luZyBzeXMvc29ja2lvLmggcHJlc2VuY2UuLi4g eWVzCmNoZWNraW5nIGZvciBzeXMvc29ja2lvLmguLi4geWVzCmNoZWNraW5nIGZvciB1bmlzdGQu aC4uLiAoY2FjaGVkKSB5ZXMKY2hlY2tpbmcgZm9yIGFuIEFOU0kgQy1jb25mb3JtaW5nIGNvbnN0 Li4uIHllcwpjaGVja2luZyBmb3Igc2l6ZV90Li4uIHllcwpjaGVja2luZyB3aGV0aGVyIHRpbWUu aCBhbmQgc3lzL3RpbWUuaCBtYXkgYm90aCBiZSBpbmNsdWRlZC4uLiB5ZXMKY2hlY2tpbmcgc3lz L2RscGkuaCB1c2FiaWxpdHkuLi4gbm8KY2hlY2tpbmcgc3lzL2RscGkuaCBwcmVzZW5jZS4uLiBu bwpjaGVja2luZyBmb3Igc3lzL2RscGkuaC4uLiBubwpjaGVja2luZyBmb3IgcmVnY29tcC4uLiB5 ZXMKY2hlY2tpbmcgZm9yIHNlbGVjdC4uLiB5ZXMKY2hlY2tpbmcgZm9yIHN0cmR1cC4uLiB5ZXMK Y2hlY2tpbmcgZm9yIHN0cmVycm9yLi4uIHllcwpjaGVja2luZyBmb3Igc3Ryc3BuLi4uIHllcwpj aGVja2luZyBmb3IgbGlicmFyeSBjb250YWluaW5nIHNvY2tldC4uLiBub25lIHJlcXVpcmVkCmNo ZWNraW5nIGZvciBsaWJyYXJ5IGNvbnRhaW5pbmcgbG9nLi4uIC1sbQpjaGVja2luZyBmb3IgZ2V0 aG9zdGJ5bmFtZS4uLiB5ZXMKY2hlY2tpbmcgZm9yIGxpYnJhcnkgY29udGFpbmluZyBpbmV0X2F0 b24uLi4gbm9uZSByZXF1aXJlZApjaGVja2luZyBmb3IgbGlicmFyeSBjb250YWluaW5nIGluZXRf cHRvbi4uLiBub25lIHJlcXVpcmVkCmNoZWNraW5nIGZvciBpbmV0X2F0b24uLi4geWVzCmNoZWNr aW5nIGZvciBpbmV0X3B0b24uLi4geWVzCmNoZWNraW5nIHNpemUgb2YgdV9pbnQ4X3QuLi4gdW5r bm93biB0eXBlCmNoZWNraW5nIHNpemUgb2YgdV9pbnQxNl90Li4uIHVua25vd24gdHlwZQpjaGVj a2luZyBzaXplIG9mIHVfaW50MzJfdC4uLiB1bmtub3duIHR5cGUKY2hlY2tpbmcgZm9yIHN0ZGlu dC5oLi4uIChjYWNoZWQpIHllcwpjaGVja2luZyBmb3IgbGlicmFyeSBjb250YWluaW5nIGdldG5h bWVpbmZvLi4uIG5vbmUgcmVxdWlyZWQKY2hlY2tpbmcgZm9yIGxpYnJhcnkgY29udGFpbmluZyBn ZXRob3N0YnlhZGRyX3IuLi4gbm9uZSByZXF1aXJlZApjaGVja2luZyBob3cgdG8gY2FsbCBnZXRo b3N0YnlhZGRyX3IuLi4gZG9uJ3Qga25vdyBob3cKY2hlY2tpbmcgZm9yIGxpYnJhcnkgY29udGFp bmluZyBnZXRob3N0YnlhZGRyLi4uIG5vbmUgcmVxdWlyZWQKY29uZmlndXJlOiBXQVJOSU5HOiB1 c2luZyBzaW5nbGUtdGhyZWFkZWQgcmVzb2x2ZXIgd2l0aCBnZXRob3N0YnlhZGRyCiAgQ29uc2lk ZXIgb2J0YWluaW5nIEFSRVMgb3IgYSBtYWNoaW5lIHdpdGggYSB3b3JraW5nIGdldGhvc3RieWFk ZHJfci4KY2hlY2tpbmcgd2hlcmUgdG8gZmluZCBwY2FwLmguLi4gL2luY2x1ZGUKY2hlY2tpbmcg Zm9yIHBjYXBfb3Blbl9saXZlIGluIC1scGNhcC4uLiB5ZXMKY2hlY2tpbmcgcGNhcC5oIHVzYWJp bGl0eS4uLiB5ZXMKY2hlY2tpbmcgcGNhcC5oIHByZXNlbmNlLi4uIHllcwpjaGVja2luZyBmb3Ig cGNhcC5oLi4uIHllcwpjaGVja2luZyBmb3IgYSBjdXJzZXMgbGlicmFyeSBjb250YWluaW5nIG12 Y2hnYXQuLi4gLWxjdXJzZXMKY2hlY2tpbmcgUE9TSVggdGhyZWFkcyBjb21waWxhdGlvbi4uLiBD RkxBR1M9IGFuZCBMSUJTPS1scHRocmVhZApjaGVja2luZyBQT1NJWCB0aHJlYWRzIHVzYWJpbGl0 eS4uLiB5ZXMKY2hlY2tpbmcgaWYgd2UgbmVlZCB0byBlbmFibGUgcHJvbWlzY3VvdXMgbW9kZSBi eSBkZWZhdWx0Li4uIG5vCmNvbmZpZ3VyZTogY3JlYXRpbmcgLi9jb25maWcuc3RhdHVzCmNvbmZp Zy5zdGF0dXM6IGNyZWF0aW5nIE1ha2VmaWxlCmNvbmZpZy5zdGF0dXM6IGNyZWF0aW5nIGNvbmZp Zy9NYWtlZmlsZQpjb25maWcuc3RhdHVzOiBjcmVhdGluZyBjb25maWcuaApjb25maWcuc3RhdHVz OiBleGVjdXRpbmcgZGVwZmlsZXMgY29tbWFuZHMKMS4wcHJlMQpjb25maWd1cmU6IFdBUk5JTkc6 CioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKioq KioqKioqKioqKioqKioqKioqKioqKgoKVGhpcyBpcyBhIHByZS1yZWxlYXNlIHZlcnNpb24uICBQ cmUtcmVsZWFzZXMgYXJlIHN1YmplY3QgdG8gbGltaXRlZAphbm5vdW5jZW1lbnRzLCBhbmQgdGhl cmVmb3JlIGxpbWl0ZWQgY2lyY3VsYXRpb24sIGFzIGEgbWVhbnMgb2YgdGVzdGluZwp0aGUgbW9y ZSB3aWRlbHkgY2lyY3VsYXRlZCBmaW5hbCByZWxlYXNlcy4KClBsZWFzZSBkbyBub3QgYmUgc3Vy cHJpc2VkIGlmIHRoaXMgcmVsZWFzZSBpcyBicm9rZW4sIGFuZCBpZiBpdCBpcyBicm9rZW4sIGRv Cm5vdCBhc3N1bWUgdGhhdCBzb21lb25lIGVsc2UgaGFzIHNwb3R0ZWQgaXQuICBJbnN0ZWFkLCBw bGVhc2UgZHJvcCBhIG5vdGUgb24KdGhlIG1haWxpbmcgbGlzdCwgb3IgYSBicmllZiBlbWFpbCB0 byBtZSBvbiBwZHdAZXgtcGFycm90LmNvbQoKVGhhbmsgeW91IGZvciB0YWtpbmcgdGhlIHRpbWUg dG8gYmUgdGhlIHRlc3RpbmcgcGhhc2Ugb2YgdGhpcyBkZXZlbG9wbWVudApwcm9jZXNzLgoKUGF1 bCBXYXJyZW4KCioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKioq KioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKgoKbWFrZSAgYWxsLXJlY3Vyc2l2ZQpNYWtp bmcgYWxsIGluIGNvbmZpZwppZiBnY2MgLURIQVZFX0NPTkZJR19IIC1JLiAtSS4gLUkuICAgICAg LWcgLU8yIC1NVCBhZGRyX2hhc2gubyAtTUQgLU1QIC1NRiAiLmRlcHMvYWRkcl9oYXNoLlRwbyIg LWMgLW8gYWRkcl9oYXNoLm8gYWRkcl9oYXNoLmM7ICB0aGVuIG12IC1mICIuZGVwcy9hZGRyX2hh c2guVHBvIiAiLmRlcHMvYWRkcl9oYXNoLlBvIjsgZWxzZSBybSAtZiAiLmRlcHMvYWRkcl9oYXNo LlRwbyI7IGV4aXQgMTsgZmkKYWRkcl9oYXNoLmM6IEluIGZ1bmN0aW9uICdoYXNoJzoKYWRkcl9o YXNoLmM6NDY6IGVycm9yOiAnc3RydWN0IGluNl9hZGRyJyBoYXMgbm8gbWVtYmVyIG5hbWVkICdz Nl9hZGRyMzInCmFkZHJfaGFzaC5jOjU0OiBlcnJvcjogJ3N0cnVjdCBpbjZfYWRkcicgaGFzIG5v IG1lbWJlciBuYW1lZCAnczZfYWRkcjMyJwoqKiogRXJyb3IgY29kZSAxCgpTdG9wIGluIC9ob21l L2N2YS90L2lmdG9wLTEuMHByZTEuCioqKiBFcnJvciBjb2RlIDEKClN0b3AgaW4gL2hvbWUvY3Zh L3QvaWZ0b3AtMS4wcHJlMS4KKioqIEVycm9yIGNvZGUgMQoKU3RvcCBpbiAvaG9tZS9jdmEvdC9p ZnRvcC0xLjBwcmUxLgoK --0016368336f87efa240496738c67-- From frederic.perrin at resel.fr Mon Dec 6 09:16:50 2010 From: frederic.perrin at resel.fr (=?utf-8?Q?Fr=C3=A9d=C3=A9ric_Perrin?=) Date: Mon, 06 Dec 2010 10:16:50 +0100 Subject: [Iftop-users] Porting iftop 1.0 to FreeBSD Message-ID: <87tyirqn0d.fsf@resel.fr> --=-=-= Content-Type: multipart/signed; boundary="==-=-="; micalg=pgp-sha1; protocol="application/pgp-signature" --==-=-= Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Hello, Today I compiled iftop-1.0pre1 if a FreeBSD system. I had to make the following changes: =2D config/int_ghba_r.c uses AF_INET, which is defined in , without #include'ing it ; =2D add =C2=AB #define define s6_addr32 __u6_addr.__u6_addr32 =C2=BB in con= fig.h (this is used in addr_hash and ns_hash.c). FreeBSD has a note in saying it is not standard ; =2D in resolver.c, only the getnameinfo() method tries to resolve ip6 addresses. I ported the AF guessing to the gethostbyaddr{,_r} methods, as I am not familiar with the libresolv or the ares librairies. This enables reverse ip6 resolution. The patch is attached. Could you please review it, and apply it if it DTRT ? =2D-=20 Fred --==-=-= Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.11 (GNU/Linux) iEYEARECAAYFAkz8qgIACgkQlSqR5GqTBEOE7gCeLtLltULeaFsZjyUAl2SnaFVC 3WoAn0dvcU6X7aaW+LPC0XVAZyWAI1ei =H/xF -----END PGP SIGNATURE----- --==-=-=-- --=-=-= Content-Type: text/x-patch Content-Disposition: inline; filename=iftop.fBSD.patch Content-Description: FreeBSD port effort diff -ur iftop-1.0pre1.orig/config/int_ghba_r.c iftop-1.0pre1/config/int_ghba_r.c --- iftop-1.0pre1.orig/config/int_ghba_r.c 2002-11-04 13:27:35.000000000 +0100 +++ iftop-1.0pre1/config/int_ghba_r.c 2010-12-04 12:43:29.000000000 +0100 @@ -6,6 +6,7 @@ static const char rcsid[] = "$Id: int_ghba_r.c,v 1.1 2002/11/04 12:27:35 chris Exp $"; +#include #include #include diff -ur iftop-1.0pre1.orig/config.h.in iftop-1.0pre1/config.h.in --- iftop-1.0pre1.orig/config.h.in 2010-11-27 12:41:51.000000000 +0100 +++ iftop-1.0pre1/config.h.in 2010-12-03 17:11:59.000000000 +0100 @@ -158,3 +158,9 @@ /* Define to `unsigned int' if does not define. */ #undef size_t + +/* fBSD doesn't define this in user-land code, with a note saying this +/* is not standard */ +#ifndef s6_addr32 +# define s6_addr32 __u6_addr.__u6_addr32 +#endif diff -ur iftop-1.0pre1.orig/resolver.c iftop-1.0pre1/resolver.c --- iftop-1.0pre1.orig/resolver.c 2010-11-27 11:56:44.000000000 +0100 +++ iftop-1.0pre1/resolver.c 2010-12-05 14:48:00.000000000 +0100 @@ -37,6 +37,22 @@ extern options_t options; +int guess_af(struct in6_addr *addr) +{ + /* If the upper three (network byte order) uint32-parts + * are null, then there ought to be an IPv4 address here. + * Any such IPv6 would have to be 'xxxx::'. Neglectable? */ + uint32_t* probe; + int af; + + probe = (uint32_t *) addr; + return (probe[1] || probe[2] || probe[3]) ? AF_INET6 : AF_INET; +} + +socklen_t aflength(int af) +{ + return af == AF_INET6 ? sizeof(struct in6_addr) : sizeof(struct in_addr); +} /* * We have a choice of resolver methods. Real computers have getnameinfo or @@ -60,18 +76,11 @@ struct sockaddr_in6 sin6; char buf[NI_MAXHOST]; /* 1025 */ int res, af; - uint32_t* probe; memset(&sin, '\0', sizeof(sin)); memset(&sin6, '\0', sizeof(sin6)); - /* If the upper three (network byte order) uint32-parts - * are null, then there ought to be an IPv4 address here. - * Any such IPv6 would have to be 'xxxx::'. Neglectable? */ - probe = (uint32_t *) addr; - af = (probe[1] || probe[2] || probe[3]) ? AF_INET6 : AF_INET; - - switch (af) { + switch (guess_af(addr)) { case AF_INET: sin.sin_family = af; sin.sin_port = 0; @@ -106,26 +115,29 @@ * Some implementations of libc choose to implement gethostbyaddr_r as * a non thread-safe wrapper to gethostbyaddr. An interesting choice... */ -char* do_resolve(struct in_addr * addr) { +char* do_resolve(struct in6_addr * addr) { struct hostent hostbuf, *hp; size_t hstbuflen = 1024; char *tmphstbuf; int res; int herr; + int af; char * ret = NULL; /* Allocate buffer, remember to free it to avoid memory leakage. */ tmphstbuf = xmalloc (hstbuflen); + af = guess_af(addr); + /* Some machines have gethostbyaddr_r returning an integer error code; on * others, it returns a struct hostent*. */ #ifdef GETHOSTBYADDR_R_RETURNS_INT - while ((res = gethostbyaddr_r((char*)addr, sizeof(struct in_addr), AF_INET, + while ((res = gethostbyaddr_r((char*)addr, aflength(af), af, &hostbuf, tmphstbuf, hstbuflen, &hp, &herr)) == ERANGE) #else /* ... also assume one fewer argument.... */ - while ((hp = gethostbyaddr_r((char*)addr, sizeof(struct in_addr), AF_INET, + while ((hp = gethostbyaddr_r((char*)addr, aflength(af), af, &hostbuf, tmphstbuf, hstbuflen, &herr)) == NULL && errno == ERANGE) #endif @@ -152,22 +164,24 @@ #elif defined(USE_GETHOSTBYADDR) /** - * Implementation using gethostbyname. Since this is nonreentrant, we have to + * Implementation using gethostbyaddr. Since this is nonreentrant, we have to * wrap it in a mutex, losing all benefit of multithreaded resolution. */ char *do_resolve(struct in_addr *addr) { static pthread_mutex_t ghba_mtx = PTHREAD_MUTEX_INITIALIZER; char *s = NULL; struct hostent *he; + int af; + af = guess_af(*addr); + pthread_mutex_lock(&ghba_mtx); - he = gethostbyaddr((char*)addr, sizeof *addr, AF_INET); + he = gethostbyaddr((char*)addr, aflength(af), af); if (he) s = xstrdup(he->h_name); pthread_mutex_unlock(&ghba_mtx); return s; } - #elif defined(USE_LIBRESOLV) #include @@ -299,7 +313,7 @@ #elif defined(USE_FORKING_RESOLVER) /** - * Resolver which forks a process, then uses gethostbyname. + * Resolver which forks a process, then uses gethostbyaddr. */ #include --=-=-=-- From cva at pobox.com Mon Dec 6 09:37:04 2010 From: cva at pobox.com (Chris Anderson) Date: Mon, 6 Dec 2010 01:37:04 -0800 Subject: [Iftop-users] Porting iftop 1.0 to FreeBSD In-Reply-To: <87tyirqn0d.fsf@resel.fr> References: <87tyirqn0d.fsf@resel.fr> Message-ID: --001485f7cc4682fba80496baa12f Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable I just applied the patch and now it fails to build here: if gcc -DHAVE_CONFIG_H -I. -I. -I. -g -O2 -MT counter_hash.o -MD -MP -MF ".deps/counter_hash.Tpo" -c -o counter_hash.o counter_hash.c; then mv -f ".deps/counter_hash.Tpo" ".deps/counter_hash.Po"; else rm -f ".deps/counter_hash.Tpo"; exit 1; fi counter_hash.c:6:26: error: counter_hash.h: No such file or directory *** Error code 1 2010/12/6 Fr=E9d=E9ric Perrin > Hello, > > Today I compiled iftop-1.0pre1 if a FreeBSD system. I had to make the > following changes: > > - config/int_ghba_r.c uses AF_INET, which is defined in , > without #include'ing it ; > > - add =AB #define define s6_addr32 __u6_addr.__u6_addr32 =BB in config.h > (this is used in addr_hash and ns_hash.c). FreeBSD has a note in > saying it is not standard ; > > - in resolver.c, only the getnameinfo() method tries to resolve ip6 > addresses. I ported the AF guessing to the gethostbyaddr{,_r} methods, > as I am not familiar with the libresolv or the ares librairies. This > enables reverse ip6 resolution. > > The patch is attached. Could you please review it, and apply it if it > DTRT ? > > -- > Fred > > --001485f7cc4682fba80496baa12f Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable I just applied the patch and now it fails to build here:

if gcc -DHAVE_CONFIG_H -I. -I. -I. =A0 =A0 =A0-g -O2 -MT counter_hash= .o -MD -MP -MF ".deps/counter_hash.Tpo" -c -o counter_hash.o coun= ter_hash.c; =A0then mv -f ".deps/counter_hash.Tpo" ".deps/co= unter_hash.Po"; else rm -f ".deps/counter_hash.Tpo"; exit 1;= fi

counter_hash.c:6:26: error: counter_hash.h: No such file or directory<= /div>

*** Error code 1

2010/12/6 Fr=E9d=E9ric Perrin <frederic.perrin@resel.fr>

Hello,

Today I compiled iftop-1.0pre1 if a FreeBSD system. I had to make the
following changes:

- config/int_ghba_r.c uses AF_INET, which is defined in <sys/socket.h>= ;,
=A0without #include'ing it ;

- add =AB #define define s6_addr32 __u6_addr.__u6_addr32 =BB in config.h =A0(this is used in addr_hash and ns_hash.c). FreeBSD has a note in
=A0<netinet6/in6.h> saying it is not standard ;

- in resolver.c, only the getnameinfo() method tries to resolve ip6
=A0addresses. I ported the AF guessing to the gethostbyaddr{,_r} methods,<= br> =A0as I am not familiar with the libresolv or the ares librairies. This =A0enables reverse ip6 resolution.

The patch is attached. Could you please review it, and apply it if it
DTRT ?

--
Fred

--001485f7cc4682fba80496baa12f-- From cva at pobox.com Mon Dec 6 09:37:55 2010 From: cva at pobox.com (Chris Anderson) Date: Mon, 6 Dec 2010 01:37:55 -0800 Subject: [Iftop-users] Porting iftop 1.0 to FreeBSD In-Reply-To: References: <87tyirqn0d.fsf@resel.fr> Message-ID: --0016364d27978dae550496baa4b1 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable sorry, forgot to mention this was FreeBSD 8.1-RELEASE. 2010/12/6 Chris Anderson > I just applied the patch and now it fails to build here: > > if gcc -DHAVE_CONFIG_H -I. -I. -I. -g -O2 -MT counter_hash.o -MD -MP > -MF ".deps/counter_hash.Tpo" -c -o counter_hash.o counter_hash.c; then m= v > -f ".deps/counter_hash.Tpo" ".deps/counter_hash.Po"; else rm -f > ".deps/counter_hash.Tpo"; exit 1; fi > counter_hash.c:6:26: error: counter_hash.h: No such file or directory > *** Error code 1 > > > 2010/12/6 Fr=E9d=E9ric Perrin > > Hello, >> >> Today I compiled iftop-1.0pre1 if a FreeBSD system. I had to make the >> following changes: >> >> - config/int_ghba_r.c uses AF_INET, which is defined in , >> without #include'ing it ; >> >> - add =AB #define define s6_addr32 __u6_addr.__u6_addr32 =BB in config.h >> (this is used in addr_hash and ns_hash.c). FreeBSD has a note in >> saying it is not standard ; >> >> - in resolver.c, only the getnameinfo() method tries to resolve ip6 >> addresses. I ported the AF guessing to the gethostbyaddr{,_r} methods, >> as I am not familiar with the libresolv or the ares librairies. This >> enables reverse ip6 resolution. >> >> The patch is attached. Could you please review it, and apply it if it >> DTRT ? >> >> -- >> Fred >> >> > --0016364d27978dae550496baa4b1 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable sorry, forgot to mention this was=A0FreeBSD 8.1-RELEASE.

2010/12/6 Chris Anderson <cva@pobox.com>

I just applied the patch and now it fails to build here:

if gcc -DHAVE_CONFIG_H -I. -I. -I. =A0 =A0 =A0-g -O2 -MT counter_hash= .o -MD -MP -MF ".deps/counter_hash.Tpo" -c -o counter_hash.o coun= ter_hash.c; =A0then mv -f ".deps/counter_hash.Tpo" ".deps/co= unter_hash.Po"; else rm -f ".deps/counter_hash.Tpo"; exit 1;= fi

counter_hash.c:6:26: error: counter_hash.h: No such file or directory<= /div>
*** Error code 1

2010/12/6 Fr=E9d=E9ric Perrin <frederic.perrin@resel.fr><= /span>

Hello,

Today I compiled iftop-1.0pre1 if a FreeBSD system. I had to make the
following changes:

- config/int_ghba_r.c uses AF_INET, which is defined in <sys/socket.h>= ;,
=A0without #include'ing it ;

- add =AB #define define s6_addr32 __u6_addr.__u6_addr32 =BB in config.h =A0(this is used in addr_hash and ns_hash.c). FreeBSD has a note in
=A0<netinet6/in6.h> saying it is not standard ;

- in resolver.c, only the getnameinfo() method tries to resolve ip6
=A0addresses. I ported the AF guessing to the gethostbyaddr{,_r} methods,<= br> =A0as I am not familiar with the libresolv or the ares librairies. This =A0enables reverse ip6 resolution.

The patch is attached. Could you please review it, and apply it if it
DTRT ?

--
Fred

--0016364d27978dae550496baa4b1-- From frederic.perrin at resel.fr Mon Dec 6 11:34:43 2010 From: frederic.perrin at resel.fr (=?utf-8?Q?Fr=C3=A9d=C3=A9ric_Perrin?=) Date: Mon, 06 Dec 2010 12:34:43 +0100 Subject: [Iftop-users] Porting iftop 1.0 to FreeBSD In-Reply-To: (Chris Anderson's message of "Mon, 6 Dec 2010 01:37:04 -0800") References: <87tyirqn0d.fsf@resel.fr> Message-ID: <87mxojqgmk.fsf@resel.fr> Le lundi 6 =C3=A0 10:37, Chris Anderson a =C3=A9crit : > I just applied the patch and now it fails to build here: > > if gcc -DHAVE_CONFIG_H -I. -I. -I. =C2=A0 =C2=A0 =C2=A0-g -O2 -MT counter= _hash.o -MD -MP -MF > ".deps/counter_hash.Tpo" -c -o counter_hash.o counter_hash.c; =C2=A0then = mv -f > ".deps/counter_hash.Tpo" ".deps/counter_hash.Po"; else rm -f ".deps/ > counter_hash.Tpo"; exit 1; fi > counter_hash.c:6:26: error: counter_hash.h: No such file or directory > *** Error code 1 I didn't touch counter_hash.{c,h}. Are you using `make all', by chance? From a mail by Paul Warren, you should be using `make iftop' instead. http://lists.beasts.org/pipermail/iftop-users/2010-November/000318.html > Turns out that there's some stuff in the distribution that shouldn't > be there and will cause the build to fail. > > If you run "make iftop" rather than just "make" it should work. --=20 Fred From cva at pobox.com Mon Dec 6 20:13:50 2010 From: cva at pobox.com (Chris Anderson) Date: Mon, 6 Dec 2010 12:13:50 -0800 Subject: [Iftop-users] Porting iftop 1.0 to FreeBSD In-Reply-To: <87mxojqgmk.fsf@resel.fr> References: <87tyirqn0d.fsf@resel.fr> <87mxojqgmk.fsf@resel.fr> Message-ID: --0016364d2797c2c6220496c386ed Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable oops, that was it (somehow I forgot that since the first time I tried to build it). works great (finally v6 traffic!) cva 2010/12/6 Fr=E9d=E9ric Perrin > Le lundi 6 =E0 10:37, Chris Anderson a =E9crit : > > I just applied the patch and now it fails to build here: > > > > if gcc -DHAVE_CONFIG_H -I. -I. -I. -g -O2 -MT counter_hash.o -MD -= MP > -MF > > ".deps/counter_hash.Tpo" -c -o counter_hash.o counter_hash.c; then mv = -f > > ".deps/counter_hash.Tpo" ".deps/counter_hash.Po"; else rm -f ".deps/ > > counter_hash.Tpo"; exit 1; fi > > counter_hash.c:6:26: error: counter_hash.h: No such file or directory > > *** Error code 1 > > I didn't touch counter_hash.{c,h}. Are you using `make all', by > chance? From a mail by Paul Warren, you should be using `make iftop' > instead. > > http://lists.beasts.org/pipermail/iftop-users/2010-November/000318.html > > Turns out that there's some stuff in the distribution that shouldn't > > be there and will cause the build to fail. > > > > If you run "make iftop" rather than just "make" it should work. > > > -- > Fred > > _______________________________________________ > iftop-users mailing list > iftop-users@lists.beasts.org > http://lists.beasts.org/mailman/listinfo/iftop-users > --0016364d2797c2c6220496c386ed Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable oops, that was it (somehow I forgot that since the first time I tried to bu= ild it).

works great (finally v6 traffic!)

cva

2010/12/6 Fr=E9d=E9ric Pe= rrin <fred= eric.perrin@resel.fr>

Le lundi 6 =E0 10:37, Chris Anderson a =E9c= rit :

> I just applied the patch and now it fails to build h= ere:
>
> if gcc -DHAVE_CONFIG_H -I. -I. -I. =A0 =A0 =A0-g -O2 -MT counter_hash.= o -MD -MP -MF
> ".deps/counter_hash.Tpo" -c -o counter_hash.o counter_hash.c= ; =A0then mv -f
> ".deps/counter_hash.Tpo" ".deps/counter_hash.Po"; = else rm -f ".deps/
> counter_hash.Tpo"; exit 1; fi
> counter_hash.c:6:26: error: counter_hash.h: No such file or directory<= br> > *** Error code 1

I didn't touch counter_hash.{c,h}. Are you using `make all', = by
chance? From a mail by Paul Warren, you should be using `make iftop' instead.

http://lists.beasts.org/pipermail/iftop-users/20= 10-November/000318.html
> Turns out that there's some stuff in the distribution that shouldn= 't
> be there and will cause the build to fail.
>
> If you run "make iftop" rather than just "make" it= should work.

--
Fred

_______________________________________________
iftop-users mailing list
iftop-users@lists.beasts.or= g
http://lists.beasts.org/mailman/listinfo/iftop-users

--0016364d2797c2c6220496c386ed--