summaryrefslogtreecommitdiff
path: root/usr.sbin/bgpd/rde.c
AgeCommit message (Collapse)Author
2004-07-05fix a few KNF falloutsHenning Brauer
2004-07-042 more file descriptors for each RDE and SE inherited from the parentHenning Brauer
we should close
2004-07-04when getting rid of the listen_addr TAILQ after forking actually closeHenning Brauer
the file descriptors in RDE and parent process, not needed or used there
2004-07-03Switch mrt dumping to fd passing. This gives some speed up when extensiveClaudio Jeker
dumping is done. Acctually mrt dumps were broken because of the fd passing. The nice side effect is a much cleaner code, especially in the parent process. OK henning@
2004-06-24First step at multiprotocol support, only partially done.Claudio Jeker
OK henning@
2004-06-23kill that stupid AS path loop whining, that is normalHenning Brauer
2004-06-22Cleanup. jajaja henning@Claudio Jeker
2004-06-22Make the RDE IPv6 ready missing is the message handling. The internalClaudio Jeker
prefix tree changed form a hash table to a per AF RB tree. OK henning@ some ideas are from Brent Graveland.
2004-06-20at least somewhat consistently name the TAILQ_ENTRYs... this confused meHenning Brauer
more than once
2004-06-06rework bgpd's handling of listening sockets. instead of one for eachHenning Brauer
supported address familiy, keep a tailq of an arbitary number of them. the new struct listen_addr contains the sockaddr and the fd. this fixes quite some nasty behaviour which was a consequence of the previous model. looks right deraadt@, and discussed with claudio
2004-05-21RFC 2796 bgp route reflector support. This is very useful in conjunctionClaudio Jeker
with templates. looks good, go for it henning@
2004-05-21Grrr. Forgot to cleanup dynamic announcements on shutdown.Claudio Jeker
2004-05-21Add support for dynamic announcements. Usefule to annouce temporaryClaudio Jeker
blackhole routes or to make network announcements dependent on a external state (e.g. for carp setups) OK henning@
2004-05-08KNFHenning Brauer
2004-05-07add a filter option to dump prefixes learned in UPDATEs into a PF table,Damien Miller
intended for building realtime BGP blacklists (e.g. with spamd); ok claudio & henning
2004-04-30spellingTheo de Raadt
2004-04-29sock -> fd; ok henningTheo de Raadt
2004-04-28Unbreak the stuff that I commited first that was totaly broken. OK henning@Claudio Jeker
2004-04-28make this at least compileHenning Brauer
2004-04-28Enable route refresh in the RDE. Now peer can request route refreshes.Claudio Jeker
OK henning@
2004-04-27crud stripping; henning okTheo de Raadt
2004-04-25Remove the no longer needed configure stuff in RDE. The peer list needs noClaudio Jeker
longer to be synced between parent, SE and RDE. OK henning@
2004-04-25add "neighbor cloning", allowing you to specify a prefix and prefixlengthHenning Brauer
instead of the neighbor's IP address. WHen a connection comes in matching that mask we clone the neighbor spec. IPv6 match code by itojun, rde feeding by claudio, ok claudio
2004-04-25&d and %d have different meanings in format strings, yeahHenning Brauer
2004-03-20comment spelling fixes; ok henning@David Krause
2004-03-15properly check chdir success after chroot() and tweak err msgs a little,Henning Brauer
claudio ok
2004-03-12tweak err reportingHenning Brauer
2004-03-12not debugging any moreHenning Brauer
2004-03-11KNFHenning Brauer
2004-03-11Shutdown the RDE cleanly on exit. Plug some memleaks. OK henning@Claudio Jeker
2004-03-10when we exit try to write out what is left in the msg buffers for the imsgHenning Brauer
pipes and clear buffers afterwards
2004-03-05Plug some memory leaks in rde. Based on a patch by Patrick Latifi.Claudio Jeker
Added attr_move() so that we can copy the attribute before calling the filter. path_update() will now use the passed attribute so it can't be simply reused. OK henning@
2004-03-02Framework for rib lookups by prefix. OK henning@Claudio Jeker
2004-03-01Make it possible to diable the decision process. This is a feature only usefulClaudio Jeker
for route-collectors. OK henning@
2004-02-27remove unneded peer pointer in struct prefix and change a in_addr_t toClaudio Jeker
struct in_addr. OK henning@
2004-02-27Cleanup no functional changes. OK henning@Claudio Jeker
2004-02-26show rib infrastructure. At least full dumps and per as dumps. Per prefixClaudio Jeker
dump need some more work. OK henning@
2004-02-26Implement "enforce neighbor-as yes|no" which is by default on for ebgpClaudio Jeker
neighbors. While doing that check also that the nexthop is valid (not class D or E and not in 127/8 range). Kill some TODO and XXX and rename the british neighbour to neighbor as used everywhere else. OK henning@
2004-02-26Add per netwok definition filter sets. So you can now useClaudio Jeker
network 10.0.0.0/8 set localpref 100 OK henning@
2004-02-25Rewrite some parts of the mrt dump handling. It is no longer possible toClaudio Jeker
dump the filtered updates but therefore it is now possible to dump per neighbor and also to dump the outgoing messages. OK henning@
2004-02-24Enhance filters. prefixlen knows now 8-24 and 8><24. It is possible to useClaudio Jeker
prefix 10.0.0.0/8 prefixlen >= 8 and set localpref 100 can be set on a per neighbor basis. OK henning@
2004-02-23I was hunting this bug for quite some time. Don't use a value you need laterClaudio Jeker
as counter. This fixes the bad nlri prefix errors I got. OK henning@
2004-02-19Add support for basic filters. Nothing optimized and it has some issues butClaudio Jeker
this is a huge step forward. OK henning@
2004-02-19Make the code more portable. Add some missing header files and make the useClaudio Jeker
of the queue(3) makros more portable. OK henning@ some time ago.
2004-02-18Correctly handle parse errors in aspath and prefixes. Also do input aspathClaudio Jeker
loop detection. OK henning@
2004-02-18Bring the path attribute parsing on big step closer to the RFC.Claudio Jeker
Check that evry attribut is only allowed once and ensure that the mandatory attributes are present. no objections henning@
2004-02-17Somehow size needs some time to settle. The direct inline call of attr_errorClaudio Jeker
caused troubles on my test machines which manifested in garbage size values. OK henning@
2004-02-16Extend imsg API so that messages can be composed on the fly. In the RDE itClaudio Jeker
happens often, that imsg consists of multiple objects. OK henning@
2004-02-16Make the path attribute handling more RFC conformant. Also move theClaudio Jeker
parser to rde_attr.c where it belongs. Still missing: better aspath loop detection (should be done afterwards) and some basic error checking for optional attributes. OK henning@
2004-02-09replace a bunch of u_long by u_int32_tHenning Brauer