~ [ source navigation ] ~ [ diff markup ] ~ [ identifier search ] ~ [ freetext search ] ~ [ file search ] ~

Linux Cross Reference
cvs/emstar/tos-contrib/sympathy/narrative


  1 Ahas:
  2 1) Knowing number requests vs how many it tried to send vs... helpful!
  3 2) Knowing how many of each type of packet very helpful - often just certain
  4         type of packet is not making it through!
  5 3) Having events is not very helpful so far - only metrics
  6 4) Giving the positives is helpful as well as the negatives - maybe have
  7    a file where i summarize positives? but TOO MUCH DATA - how do iake
  8    it a useful summary - what goes in a useful summary?
  9 5) Setting check period to 2xrequest period.
 10 6) BIST - even for sympathy - by running troubled_node checks!
 11         - just looking at metrics reported about each app
 12 
 13 Solutions:
 14 1) Start/stop topo-data
 15 
 16 Problem:
 17 1) Not getting sympathy responses, even from one-hop neighbors, but getting 
 18 beacons from them. 
 19 Measured: 
 20 - number of requests they are getting (they were getting requests),
 21         they just weren't sending responses.
 22 - then added, number of times tried to send but failed, also very few.
 23 
 24 Things that help:
 25 - nice to have space to put some external variables - during pre-dep
 26   debugging
 27 - should have sympathy periodically send metric info with these
 28 vars - can use this to see if sympathy just not getting tasked or
 29 somethign else going wrong.
 30 - use sympathy request/response as indicator of whether motes can
 31 be reached and be tasked - cuz sympathy uses same routing overlay
 32 
 33 2) Not getting sympathy responses
 34   - see reported number of sympathy requests received - this needs to
 35         be periodically just broadcast without being requested.
 36 
 37 3) Not getting sypathy respnoses
 38   - periodically have metrics sent without request
 39 
 40 4) Not getting sympathy requests:
 41   - answer questions:
 42      - are we getting regular automated metric returns from node?
 43         - if no: goto check_routing
 44         - if yes: is node transmitting requested metrics?
 45            - if yes: is the node transmitting sympathy events?
 46               - if no: is the node receiving sympathy requests?
 47                  - if no:
 48                  - if yes: are they continuing to increase even though number
 49                         events transmitted are not?
 50               - if yes:
 51            - if no:
 52 
 53       - check_routing:
 54 
 55 5) Fault detection: identify nodes that are not receiving sympathy
 56         requests, or not sending as many event updates as they have 
 57         received - is this because their last event hours is not
 58         same as others?
 59 ** Separate number requests msgs rcvd into: number requested
 60     events rcvd vs number requested metrics rcvd, and
 61     specify number requests sent out!
 62 ** for nodes that havent sent many events:
 63      - check how many events were transmitted compared to
 64            how many were received!
 65      - for these nodes, find nodes with same next-hop
 66      - else check how many requests they received
 67      - else see if last-event time is just < time-awake (so maybe
 68          just dont have events to send)
 69 
 70 6) Too many errors (caught that with num errors) on nodes with node-2 as a next-hop.
 71 so added jitter to response to flooded response
 72 
 73 7) Wasn't getting metrics or data from a node, BUT i was getting
 74 other data packets (beacons/neighbor-lists etc) - so added this to
 75 the test.
 76 
 77 8) observe performance degrade (i.e. stop getting data), and then
 78   node would reboot
 79 
 80 Design:
 81 start/stop symp-requests - command-line/status-device
 82 check for reboot! (ensure its not just a roll-over, but a reboot!)
 83 
 84 Why is wsn different:
 85 - energy debugging, performance/network usage assertions, 
 86 functionality debugging all collide -
 87         how do we express those needs/rules -even dynamically, how do you assess
 88 
 89 X graph time awake
 90 X graph its fault-category
 91 Classify nodes:
 92 - we are not getting data from node
 93    - nodes have no data to send
 94    - data is not getting through
 95 - performance: nodes sending data, but we arent getting a lot of it
 96 - nodes that are fine
 97 

~ [ source navigation ] ~ [ diff markup ] ~ [ identifier search ] ~ [ freetext search ] ~ [ file search ] ~

This page was automatically generated by the LXR engine.
Visit the LXR main site for more information.