bgpd/bgp_aspath.c        |   40 +++++++++
 bgpd/bgp_aspath.h        |    2 
 bgpd/bgp_attr.c          |   11 ++
 bgpd/bgp_fsm.c           |   52 ++++++++----
 bgpd/bgp_network.c       |  190 ++++++++++++++++++++++++-----------------------
 bgpd/bgp_network.h       |    3 
 bgpd/bgp_vty.c           |    5 -
 bgpd/bgpd.c              |   78 +++++++++++--------
 lib/memtypes.c           |    1 
 ospf6d/ospf6_zebra.c     |    5 -
 ospfd/ospf_zebra.c       |    7 -
 ripd/rip_interface.c     |   11 +-
 ripd/ripd.c              |    3 
 ripngd/ripng_interface.c |   11 +-
 zebra/kernel_null.c      |    1 
 zebra/main.c             |    3 
 zebra/zserv.c            |   18 ++--
 zebra/zserv.h            |    1 
 18 files changed, 279 insertions(+), 163 deletions(-)

New commits:
commit 2158ad2359a370dd349f4543a2d5d535bb5a8770
Author: Paul Jakma <[email protected]>
Date:   Tue Jul 28 18:10:55 2009 +0100

    bgpd/trivial: Fix indentation in previous

commit fc4dc59d08589a887ac33a70d946eb7905541f5f
Author: Steve Hill <[email protected]>
Date:   Tue Jul 28 17:54:35 2009 +0100

    bgpd: Allow inbound connections to non-default view
    
    * bgpd.c: (peer_lookup) Search through all BGP instances for matches, not 
just
      the default instance, if no specific instance is given.
      (peer_lookup_with_open) same.

commit 3117b5c47df044f8b2197fff60641075394fdce5
Author: Steve Hill <[email protected]>
Date:   Tue Jul 28 17:50:00 2009 +0100

    bgpd: fd leak in bgpd
    
    * bgp_fsm.c: I have found an fd leak in bgpd that is caused by the 'new'
      Clearing state.  I've been seeing it from hold timer failures, but it can
      also be triggered by other things.
    
      When Hold_Timer_expired fires in Established state, a notify is sent and
      BGP_Stop event queued.  The fsm then transitions into Clearing state.
      That is the problem; When the BGP_Stop event is serviced, the state table
      says to ignore it while in Clearing.  Thus bgp_stop is not called and the
      fd leaks.  Previously the peer would be in Idle state, which correctly
      handles the BGP_Stop event.
    
      Fix by making bgp_stop safe to call from Clearing state, without losing
      ClearingCompleted events, and then ensuring it is called prior to
      transition from Clearing->Idle.

commit cb9e0ceec988a059dd2269ebc54722c01afb7cca
Author: Paul Jakma <[email protected]>
Date:   Sun Apr 20 00:31:57 2008 +0100

    [zebra] Fix crash in test_zebra
    
    * kernel_null.c: we're pretending to add an address, so
      set IFC_REAL. If this isn't done, calling into the 'got address
      from kernel' half of zebra will implicitly-withdraw the ifc, which
      could cause a crash in test_zebra.c.

commit 736d440845c8c0cfdb56b86254887bab5fc7f3f5
Author: Vasilis Tsiligiannis <[email protected]>
Date:   Mon Jul 20 01:59:10 2009 +0300

    bgpd: Delete AS_CONFED_SEQUENCE when prepending an AS_SEQUENCE type segment
    
    AS_CONFED_SEQUENCE segment should always be deleted when prepending
    (e.g. with a route-map) an AS_SEQUENCE segment to an AS path. Otherwise,
    AS_CONFED_SEQUENCE will not be deleted later when updating EBGP peers (since
    it is not the leftmost segment) and will leak outside the confederation.

commit ca87e1d37b3c30648e9bacb476a3c89729512f6d
Author: Vasilis Tsiligiannis <[email protected]>
Date:   Mon Jul 20 01:28:35 2009 +0300

    bgpd: Implement BGP confederation error handling (RFC5065, Par. 5)
    
    This patch implements BGP confederation error handling in Quagga as 
described
    in RFC5065, paragraph 5.
    
    * bgp_aspath.c: (aspath_confed_check, aspath_left_confed_check) new 
functions
    * bgp_attr.c: (bgp_attr_aspath_check) apply previous and NOTIFY if there's
      a problem.

commit 30d205909c5a81cc306b161c16abff0fd7f566a7
Author: Stephen Hemminger <[email protected]>
Date:   Tue Jul 28 11:58:51 2009 +0100

    [global] interface flags format warnings
    
    */*: ifp->flags is 64 bit unsigned which can not be handled by %l on 32
         bit architectures - requires %ll and the appropriate cast.

commit d023aec49f70156d2ed894a8fba65bcfa221ff02
Author: Stephen Hemminger <[email protected]>
Date:   Tue Jul 21 16:27:21 2009 -0700

    bgpd: start listener on first instance
    
    Start BGP listener only after first instance is started.  This helps the
    security if BGP is not used but daemon is started.  It also addresses some
    issues like MD5 not working on listener unless IPV6 configured (because
    listener was not in list); as well as compiler warnings.
    
    * bgp_network.c: (bgp_listener) listen socket creation consolidated here
      (bgp_socket) Use bgp_listener
    * bgpd.c: (bgp_get) call bgp_socket on creation of first struct bgp.
      (bgp_init) remove bgp_socket call.
    * memtypes.c: Add MTYPE_BGP_LISTENER

commit 03621955d8ab0ea30024bd8fc83afda8e48db712
Author: Stephen Hemminger <[email protected]>
Date:   Tue Jul 21 16:27:20 2009 -0700

    bgpd: peer action table static/const
    
    bgpd.c: (various tables) only used in one place and are immutable.

commit 78b31d5cf511c400d7f09158a3d8863f1bf63c2a
Author: Stephen Hemminger <[email protected]>
Date:   Tue Jul 21 16:27:26 2009 -0700

    ripd: set IP TOS for control socket
    
    * ripd.c: (rip_create_socket) RIP packets should go out with Type Of Service
      (DSCP) set to Internet control (like OSPF and BGP).

commit 9a1a331d63ae9db1912a1edb48c5c176454562f8
Author: Paul Jakma <[email protected]>
Date:   Mon Jul 27 12:27:55 2009 +0100

    [bgpd] Restore ability of 'neighbor ... update-source' to take interface 
name
    
    * bgp_vty.c: (BGP_UPDATE_SOURCE_STR) Restore ability to accept arbitrary
      interface names for the 'neighbor ... update-source' command - shouldn't
      have been deleted.
      (BGP_UPDATE_SOURCE_STR) Add help for same.

commit 97be79f9d38b07214d5ad9b437e1a852c594a985
Author: Denis Ovsienko <[email protected]>
Date:   Fri Jul 24 20:45:31 2009 +0400

    [zebra] do not touch socket before pidfile locking
    
    Move zserv socket creation code into zebra_zserv_socket_init() and
    call it only after pidfile lock has been acquired exclusively. This
    keeps subsequent zebra daemons from deleting the working socket of
    an already running process (bug #403).

http://suva.vyatta.com/git/?p=vyatta-quagga.git;a=commitdiff;h=2158ad2359a370dd349f4543a2d5d535bb5a8770
http://suva.vyatta.com/git/?p=vyatta-quagga.git;a=commitdiff;h=fc4dc59d08589a887ac33a70d946eb7905541f5f
http://suva.vyatta.com/git/?p=vyatta-quagga.git;a=commitdiff;h=3117b5c47df044f8b2197fff60641075394fdce5
http://suva.vyatta.com/git/?p=vyatta-quagga.git;a=commitdiff;h=cb9e0ceec988a059dd2269ebc54722c01afb7cca
http://suva.vyatta.com/git/?p=vyatta-quagga.git;a=commitdiff;h=736d440845c8c0cfdb56b86254887bab5fc7f3f5
http://suva.vyatta.com/git/?p=vyatta-quagga.git;a=commitdiff;h=ca87e1d37b3c30648e9bacb476a3c89729512f6d
http://suva.vyatta.com/git/?p=vyatta-quagga.git;a=commitdiff;h=30d205909c5a81cc306b161c16abff0fd7f566a7
http://suva.vyatta.com/git/?p=vyatta-quagga.git;a=commitdiff;h=d023aec49f70156d2ed894a8fba65bcfa221ff02
http://suva.vyatta.com/git/?p=vyatta-quagga.git;a=commitdiff;h=03621955d8ab0ea30024bd8fc83afda8e48db712
http://suva.vyatta.com/git/?p=vyatta-quagga.git;a=commitdiff;h=78b31d5cf511c400d7f09158a3d8863f1bf63c2a
http://suva.vyatta.com/git/?p=vyatta-quagga.git;a=commitdiff;h=9a1a331d63ae9db1912a1edb48c5c176454562f8
http://suva.vyatta.com/git/?p=vyatta-quagga.git;a=commitdiff;h=97be79f9d38b07214d5ad9b437e1a852c594a985
_______________________________________________
svn mailing list
[email protected]
http://mailman.vyatta.com/mailman/listinfo/svn

Reply via email to