TODO

   1 Core
   2 ~~~~
   3 - debug: static
   4 - debug: pipe
   5 - debug: krt
   6
   7 - static: check validity of route destination?
   8 - static: allow specifying a per-route filter program for setting route attributes?
   9
  10 - krt: rescan interfaces when route addition fails?
  11
  12 - tagging of external routes?
  13
  14 Commands
  15 ~~~~~~~~
  16 - show status
  17 - showing of routing table as seen by given protocol
  18 - printing of dynamic attributes
  19
  20 Documentation
  21 ~~~~~~~~~~~~~
  22 - write doctool
  23 - write documentation :|
  24
  25 Globals
  26 ~~~~~~~
  27 - right usage of DBG vs. debug
  28 - cleanup debugging calls
  29 - logging and tracing; use appropriate log levels
  30 - check incoming packets and log errors!!
  31 - check log calls for trailing newlines
  32 - check if all protocols set proper packet priorities and TTL's.
  33 - replace all NUM, IPA and expr tokens by constant filter expressions
  34 - try compiling with -Wunused
  35 - does everybody test return value of sk_open?
  36 - doc: references to RFC's we did follow
  37 - protocols: implement CLI hooks and per-procotol CLI commands
  38 - protocols: implement reconfigure hook
  39 - protocols: use locking
  40
  41 Various ideas
  42 ~~~~~~~~~~~~~
  43 - netlink: import Linux route attributes to our rta's, so that they can be filtered?
  44 - config: executable config files
  45 - client: access control
  46 - config: when parsing prefix, check zero bits
  47 - io: use poll if available
  48 - IPv6 router advertisements
  49 - real multipath (doesn't seem to be simple at all :()
  50 - fake multipath (even less simple)
  51 - route recalculation timing and flap dampening [see RFC2439 for algorithms]
  52 - aggregate engine: standard route aggregation and summarization [RFC2519]
  53 - aggregate engine: injection of manually configured pseudo-static routes
  54 - generate default route if any working BGP connection exists (aggregate engine again?)
  55 - generate default route to IGP's (aggregate engine yet another time?)
  56 - look at RFC 2386 (QoS-based routing)
  57
  58
  59 RIP
  60 ~~~
  61
  62 OSPF
  63 ~~~~
  64         - refuse running on non-multicast devices
  65         - importing of device routes for networks where we don't run OSPF
  66         - check incoming packets using neighbor cache
  67         - RFC2328 appendix E: Use a better algorithm
  68         - automatic generation of external route tags (RFC1403)
  69         - RFC1587 NSSA areas
  70         - RFC2370 opaque LSA's
  71         - respect interface MTU and try not to create larger packets unless unavoidable