Return to TODO CVS log

Up to [ELWIX - Embedded LightWeight unIX -] / embedaddon / mtr

Version 0.95

    1: Hi everyone,
    2: 
    3: This is the "todo" file for mtr. I just realized that some people
    4: might think that this is all in MY queue to implement. That is not
    5: true: This is the "for everybody" todo list. Feel free to pick a
    6: "project" and implement something off this list.
    7: 
    8: Students: Feel free to take up one of these as a programming exercise
    9: for one of your courses.
   10: 
   11: Everybody: If you want to start on something, contact me first, so
   12: that the effort isn't wasted by someone who finishes just a tad
   13: earlier. I'll happily provide "coaching" to anyone who wants to
   14: implement something on this list. That way we get the design of
   15: these things the way I like them. This should result in a better
   16: maintainable mtr.
   17: 
   18: Oh, Feel free to provide suggestions for this list.
   19: 
   20: 
   21: -- REW
   22: 
   23: ----------------------------------------------------------------------
   24: 
   25: - cleanup
   26:   - cleanup warnings that the newer GCC produces.
   27: 
   28: - Stuff to implement:
   29: 
   30:   - Allow mtr to log the return packets, for later analysis.
   31:     Done: 0.25 . Todo: allow the user interface(s) to work while
   32:     still logging to a file. Write a "logfile displaying" mode to
   33:     mtr.
   34: 
   35:   - Request timestamping at the remote site.
   36:        Andreas Fasbender has an algorithm that will allow us to
   37:        convert these measurements into one-way measurements, not just
   38:        round-trip.
   39: 
   40:   - allow "keyboard navigation" in the GTK version.
   41: 
   42:   - Keep all packets and make the "best" and "worst" columns show the
   43:     xx-th percentile....
   44: 
   45:   - Being able to expand the "column width" of the hosts listed would
   46:     be nice, too.
   47: 
   48:   - Display per host stats when multiple servers respond at a particular
   49:     hop count.
   50: 
   51: 
   52: - Bugs to fix?
   53: 
   54:   - Do something useful if host couldn't be resolved.
   55:        -- Done.
   56: 
   57:   - Revert to curses mode even if DISPLAY is set, but a problem
   58:     prevents us from running in X11 mode.
   59:        -->  The problem is that gtk_init simply calls exit for us if
   60:             it finds a problem. Tricky! Suggestions welcome.
   61:        --> Call "gtk_check_init" when available. (i.e. new enough
   62:            (1.2?) GTK version).
   63: 
   64: - Nice to have:
   65: 
   66:   - stop sending packets when a new host is getting entered.
   67: 
   68:   - Show state ("looking up host") while doing the DNS lookup for a new
   69:     host.
   70: 
   71:   - to have a choice of icmp, tcp, and udp pings. -- Matt Martini
   72: 
   73:   - Autoconf 2.13 has a neat function that can be used to find the
   74:     res_init function:
   75: 
   76: 	   AC_SEARCH_LIBS(res_init, bind resolv, ,
   77: 	      AC_MSG_ERROR(No resolver library found))
   78: 
   79:     At the moment (march 1999) autoconf 2.13 is still too new to require
   80:     everyone to upgrade. About a year from now we can put this in....
   81: 
   82:   - Implement rfc2317 mechanism to do reverse lookups for networks that
   83:     have DNS delegations on non-octet boundaries. -- Daniel Bergstrom
   84:     (noa@melody.se)
   85: 
   86:   - The longer MTR runs, the less meaningful the packet loss
   87:     statistic. Or more meaningful, depending on your point of view.
   88:     Perhaps MTR should use a circular buffer of some configurable
   89:     number of results, and calculate the loss against that. -- Jacob Elder
   90: 
   91:   - It would be nice if the window size wasn't fixed. If I'm only 5
   92:     hops from the host I'm monitoring, MTR wastes a lot of screen real
   93:     estate. -- Jacob Elder
   94: 
   95:   - Colors in the curses version. -- Amix
   96: 
   97:   - If we run a mtr to monitor a connection it would be nice if the time at
   98:     which mtr was started is print somewhere. -- Sebastian Ganschow
   99: 
  100: 
  101: 
  102: ------------------------------------------------------------------------
  103: 
  104: Things that shouldn't be on the TODO list because they're done. ;-)
  105: 
  106:   - Allow a toggle between hostname/IP number display. (for example a
  107:     click on the hostname could revert to ip number display in gtk version.
  108:     curses: "n" key toggles hostnames/ipnumbers?)
  109: 
  110:   - Allow mtr to also send larger packets.
  111:        This will enable us to get a feel for the speed of the links
  112:        we're traversing. (Van Jacobson was working on this His tool
  113:        was slow, mtr will rock with this feature.... :-)
  114:        (Anybody have the statistics experience to tell me how
  115:        to do the data analysis?)
  116: 	-- DONE. Thanks to Olav Kvittem ...
  117: 
  118:   - The "don't probe all hosts at once" strategy can be improved a bit.
  119:     It should not probe more than 10 unknown hosts, but the counter need
  120:     not be reset at the start of the "round". This way if you probe
  121:     slowly (relative to the RTT time to the end host), it can probe
  122:     all hosts in the first "round".
  123:        -- DONE.
  124: 
  125:   - Read environment variable "MTR_DEFAULTS" as a commandline before
  126:     parsing the commandline.  -- DONE. (ok it's MTR_OPTIONS.)