Differences between revisions 2 and 4 (spanning 2 versions)
← Revision 2 as of 2016-08-23 23:52:14 →
Size: 2259
Editor: RussellSenior
Comment: added some more notes about NodeKlickitat
← Revision 4 as of 2016-08-24 09:18:53 →
Size: 2432
Editor: JasonBergstrom
Comment:
Deletions are marked like this. Additions are marked like this.
Line 21: Line 21:
   * currently using the BATMAN IV protocol, but might try out BATMAN V. The IV protocol uses ETX as a metric (inverse of packet loss), the V protocol includes link rate from wireless and wired interfaces as a metric.    * currently using the BATMAN IV protocol, but might try out BATMAN V. The IV protocol uses ETX as a metric (ETX is the expected number of transmissions, the inverse of packet success rate), the V protocol includes link rate from wireless and wired interfaces as a metric.
Line 26: Line 26:
  * No posts to http://www.portlandfiber.com/ for a few years. Do we have anything we (Russell) would like to say?

Location: NodeLutz
Date and Time: Wednesday, August 24, 2016, 6:30pm-8:00pm
Scribe: You
Roll Call: Them

Agenda

  • Pick location for next weekly
  • Node notes:

    • NodeFlorio -- reported a splash page outage Tuesday morning, apache on knod was not responding (or responding sluggishly), restarted, api recovered

    • NodeBuffaloGap -- nocatauth turned off for the Fantasy Football, until September something, upon request.

    • NodeLutz -- checking out node at the meeting

    • NodeKlickitat -- was updated over the weekend:

      • rooftop flash went smoothly
      • the wgt634u's first-boot sequence after flash wiped some nvram values, needed to recover from serial, distributed replacements
      • one wgt discovered with a bad radio, replaced an accton mr3201a with a wgt
      • some of the wgt634u's wlan LED doesn't seem to light, LED seems okay, firmware seems to be almost identical (except for ipaddr, ssid and hostname), yet some LEDs light and others don't... baffling.
      • a couple limitations of modern batman-adv, particularly the user space tools:
        • batadv-vis (modern replacement for "batctl vis") does not utilize /etc/bat-hosts, a file which maps MAC addresses to names, making interpretation of mesh link metric status more problematic
        • alfred (a user-space information distribution system) exits if the ipv6 link-local address is not derived from the layer 2 address of the bridge, in two cases, they do not match and so alfred initialization fails ... this can be corrected manually, by changing the bridge mac address to match the link-local address.
      • currently using the BATMAN IV protocol, but might try out BATMAN V. The IV protocol uses ETX as a metric (ETX is the expected number of transmissions, the inverse of packet success rate), the V protocol includes link rate from wireless and wired interfaces as a metric.
  • Mailbag:
    • Got a request from a person, low income, looking for internet connection help, Russell referred them to the public library.
  • Corporate renewal ($50) has been filed online as of Monday (oops, slightly late).

  • Google Fiber opportunities, discussion with DavidSchor, if he can make it.

  • Your Item Here

Notes

  • None Yet

[CategoryMeetingNotes]

WeeklyMeeting20160824 (last edited 2016-08-24 20:52:20 by JasonBergstrom)