W dniu 12.05.2023 o 18:37, Ondrej Filip pisze:
Dear BIRD Users,
we are releasing another alpha version of BIRD 3. We've found some
annoying bugs and fixed some usability issues as well as several
issues with BIRD crashing. We also added a simple work balancing
algorithm between threads. MRT is still switched off, BMP is not
merged; they will get into some future alpha version.
Measured performance in BGP is approx. 6-10x better (with 20 threads)
than version 2.13, depending on the actual configuration. Kernel
synchronization is slow with a large number of routes; we will look
into it soon.
If you have some time for trying this version out, we would love to
hear your feedback.
I thank my colleagues and namely Maria for this release!
Cheers
Ondrej
Thank you for giving us the opportunity to test. So far I spotted only
few flaws:
1. Routes with unknown attributes which are abusively
extending/violating BGP protocol are still rejected.
rtr bird[98404]: bgp_r_v4: Malformed bgp_unknown_0xff attribute -
conflicting flags (e0, expected 80)
rtr bird[98404]: bgp_v_v4: Malformed bgp_unknown_0x24 attribute -
conflicting flags (e0, expected 80)
BIRD 2.13 accepts these routes and shows attributes this way:
BGP.ff [t]: 00 00 07 db 00 00 00 01 00 01 00 0a ff 08 00 00 00 00 ea 44
42 79
BGP.24 [t]: 01 00 00 11 5c 00 01 46
2. MTR protocol won't compile with clang.
CC -o obj/proto/mrt/mrt.o -c proto/mrt/mrt.c
proto/mrt/mrt.c:364:3: error: expected expression
WALK_LIST(P, proto_list)
^
./lib/lists.h:52:33: note: expanded from macro 'WALK_LIST'
#define WALK_LIST(n,list) for(n=HEAD(list); NODE_VALID(n); n=NODE_NEXT(n))
^
./lib/lists.h:48:36: note: expanded from macro 'HEAD'
#define HEAD(list) ((void *)((list).head))
^
proto/mrt/mrt.c:558:43: error: too few arguments to function call,
expected 3, have 2
pool *pool = rp_new(pp, "MRT Table Dump");
~~~~~~ ^
./lib/resource.h:72:7: note: 'rp_new' declared here
pool *rp_new(pool *, struct domain_generic *, const char *);
/* Create a new pool */
^
2 errors generated.
gmake[2]: *** [Makefile:113: obj/proto/mrt/mrt.o] Error 1
3. OSPFv3 regression or verbosity increased:
v6_ospf: Bad DBDES packet from nbr 172.xx.y.254 on vlan2 - too late for
DD exchange (7)
Memory footprint with Netlink on FreeBSD is almost similar to known from
BIRD 2.13, maybe only 5-10% higher, memory leak known from alpha 1 was
fixed.
Cheers
--
Marek Zarychta
