summaryrefslogtreecommitdiffstats
path: root/proto/rip
AgeCommit message (Collapse)Author
2004-05-31One space more.Martin Mares
2004-05-31The initial sequence number for RIP md5 authentication was always zero.Martin Mares
Bad as when bird e.g. was running for two weeks and then restarted it would take another two weeks until the peers of this router would accept data again from this router, as the sequence number would be too low. Changed to use the the current system time as the starting sequence number which is a more sane start value. (by Andreas, cleaned up by me)
2004-05-31Make RIP nolisten mode actually work. The socket is required forMartin Mares
sending, the received data has to be discarded instead. (patch by Andreas Steinmetz modified by me)
2003-12-06Fix reporting of RIP socket errors. Thanks to Eric Leblond for the patch.Martin Mares
2003-04-06Bug fixes to authentication code by Eric Leblond <eleblond@init-sys.com>.Martin Mares
2002-09-21Password does not need to be null-terminated, do not print garbage inPavel Machek
such case. Thanks to silvio@big.net.
2002-03-10Applied Pavel's fix for broadcast/multicast mode.Martin Mares
2000-06-21RIP bugfixPavel Machek
2000-06-07RIPv6 now works.Pavel Machek
2000-06-07rip ipv6 bugfixPavel Machek
2000-06-07Fixes to the progdoc.Martin Mares
2000-06-07Spelling fixes to progdoc.Martin Mares
2000-06-05Better documentation. There are functions whose description is good whenPavel Machek
reading source but whose documentation does not belong to progdocs.
2000-06-05Docs updates:Pavel Machek
make prog.dvi do not overflow in .dvi include filter documentation in the system
2000-06-05Add safety check.Pavel Machek
2000-06-05Updates to both programmers and users docPavel Machek
2000-06-04Don't set precedence and TTL for the dummy socket.Martin Mares
2000-06-04Removed lots of trailing newlines in log messages.Martin Mares
Please note that the only calls which don't add newlines automatically (i.e., don't print a full line of output) are debug() and DBG().
2000-06-04debug -> DBGMartin Mares
2000-06-01Updated RIP to new neighbor cache semantics. When presented with next hopMartin Mares
equal to a local address, just ignore the route (as it is usually just an external route exported by us echoed by some RIP peer), when given an invalid next hop, moan to the log.
2000-05-31Updated all the Doc files to new format.Martin Mares
2000-05-31IPv6: Absolutize link-scope addresses of incoming packets.Martin Mares
IPv6 socket interface is hopefully right now.
2000-05-30Removed several unused local variables.Martin Mares
2000-05-30Tried to change garbagetime -> garbage time to be more consistent.Pavel Machek
2000-05-19Latest changes by Pavel have removed the error messages printed afterMartin Mares
unsuccessful socket open, but replaced them by segmentatio fault! Grrrrrrrrr.
2000-05-19Less error messages in case sk_open fails.Pavel Machek
2000-05-19Added name of protocol to messages being logged.Pavel Machek
2000-05-19Commented out the `inserting entry which is already there' message sinceMartin Mares
it's pretty normal: during feeding of the protocol, a new route can appear which will be announced normally and then repeated by the feeding process.
2000-05-17Minor cleaningPavel Machek
2000-05-16Turned off LOCAL_DEBUG.Martin Mares
2000-05-16Interface dummy is too strange for me... but psst, that's secret.Pavel Machek
2000-05-16Rip should now reconfigure itself only if needed.Pavel Machek
2000-05-11Several minor RIP changes (Pavel, please check as usually):Martin Mares
o Use FIB_ITERATE_INIT instead of calling the function fit_init() which is explictly marked as private in route.h. o When printing trace messages, don't spit out protocol name twice. o Some messages are a bit more verbose. o Added a bunch of FIXME's. o When running in broadcast mode, don't forget to bind the local end of the socket to the same broadcast address, so that we don't get the broadcasts looped back via different interface. I'm just going to look up the same problem for multicasts.
2000-05-11Fixed nasty segfault in rip.Pavel Machek
2000-05-11Some more paranoia into rip_insert and rip_deletePavel Machek
2000-05-11Unused variable killed.Pavel Machek
2000-05-10use password_same utility functionPavel Machek
2000-05-10Inlined metric and mode into struct rip_interface to make reconfigPavel Machek
work. reconfigure is conservative but should work.
2000-05-10rip_reconfigure done rightPavel Machek
2000-05-10ipa_same does not exist.Pavel Machek
2000-05-10Nicer messages from rip.Pavel Machek
2000-05-10Use ea_get_int instead of ea_find.Pavel Machek
2000-05-10Reconfigure hook copied from bgp.Pavel Machek
2000-05-10Send first udpate sooner.Pavel Machek
2000-05-10Only announce change to main routing table if it is better than current.Pavel Machek
2000-05-10Don't touch used memory.Pavel Machek
2000-05-10Killed unused variable.Pavel Machek
2000-05-10You can now say "mode multicast".Pavel Machek
2000-05-10Fix segfaults by using new ea_get_int.Pavel Machek
2000-05-08Preference is a configurable parameter.Martin Mares