TODO

   1 Core
   2 ~~~~
   3 - client: Ctrl-R eats one more enter
   4 - client: paging?
   5 - client: data losses on large dumps?
   6
   7 Documentation
   8 ~~~~~~~~~~~~~
   9 - write doctool
  10 - write documentation :|
  11
  12 Globals
  13 ~~~~~~~
  14 - right usage of DBG vs. debug
  15 - logging and tracing; use appropriate log levels
  16 - check incoming packets and log errors!!
  17 - check log calls for trailing newlines and log levels followed by comma
  18 - check if all protocols set proper packet priorities and TTL's.
  19 - try compiling with -Wunused
  20 - does everybody test return value of sk_open?
  21 - doc: references to RFC's we did follow
  22 - protocols: implement CLI hooks and per-procotol CLI commands
  23 - protocols: implement reconfigure hook
  24 - protocols: use locking
  25 - check use of system includes and sprintf()
  26
  27 Various ideas
  28 ~~~~~~~~~~~~~
  29 - static: allow specifying a per-route filter program for setting route attributes?
  30 - bgp: timing of updates?
  31 - netlink: realms
  32 - netlink: import Linux route attributes to our rta's, so that they can be filtered?
  33 - config: executable config files
  34 - filters: user defined attributes?
  35 - client: access control
  36 - io: use poll if available
  37 - IPv6 router advertisements
  38 - real multipath (doesn't seem to be simple at all :()
  39 - fake multipath (even less simple)
  40 - route recalculation timing and flap dampening [see RFC2439 for algorithms]
  41 - aggregate engine: standard route aggregation and summarization [RFC2519]
  42 - aggregate engine: injection of manually configured pseudo-static routes
  43 - generate default route if any working BGP connection exists (aggregate engine again?)
  44 - generate default route to IGP's (aggregate engine yet another time?)
  45 - look at RFC 2386 (QoS-based routing)
  46 - cli: show tables?
  47
  48 OSPF
  49 ~~~~
  50         - refuse running on non-multicast devices
  51         - importing of device routes for networks where we don't run OSPF
  52         - check incoming packets using neighbor cache
  53         - RFC2328 appendix E: Use a better algorithm
  54         - automatic generation of external route tags (RFC1403)
  55         - RFC1587 NSSA areas
  56         - RFC2370 opaque LSA's
  57         - respect interface MTU and try not to create larger packets unless unavoidable
  58
  59 Documentation (sorry, its in czech)
  60 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  61 Filtry:
  62
  63 o  Vysvetlit nesting a zastinovani.
  64
  65 Protocols:
  66
  67 o  passwords: syntaxe data uz, tusim, davno vypada jinak.
  68
  69 Struktura dokumentace:
  70
  71 o  Na konci (nebo ve zvlast sekci pro kazdy protokol?) by mel byt seznam referenci
  72    na vsechny mozne dokumenty, zejmena vsak vsechna RFC, kterymi se ridime nebo
  73    ktera maji neco spolecneho s tim, co delame (napriklad RPSL).
  74
  75 K jazyku:
  76
  77 o  K "BIRD Internet Routing Daemon" by mel patrit urcity clen.
  78 o  Pokud je v zavorce cela veta, patri pred ')' tecka, pokud neni, tak
  79    nepatri.
  80 o  Davej si pozor na rody -- router je vzdycky `it', nikdy `he'.
  81
  82 Jeste by to chtelo trosku podrobneji:
  83
  84 (1) zminit se o atributech, rici, co vsechno o route rikaji a odkazat
  85     na kapitolu o filtrech, kde jsou presne nadefinovany
  86 (2) kdyz mluvis o preferenci, mel bys rici, co to je (ze to je atribut)
  87     a kdo jej vlastne nastavuje (ze existuje globalni default pro typ
  88     protokolu, ktery muzes overridovat v konfiguraci protokolu, coz
  89     muzes pro konkretni routu overridovat filtrem).
  90 (3) popsat spojeni mezi routovacimi tabulkami, filtry a protokoly