<html>
  <head>
    <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
  </head>
  <body text="#000000" bgcolor="#FFFFFF">
    <div class="moz-cite-prefix">Mark,<br>
      <br>
      go ahead, I'll follow. I don't have any bad reports from edge or
      eap.<br>
      <br>
      Regards,<br>
      Michael<br>
      <br>
      <br>
      Am 25.09.19 um 03:47 schrieb Mark Webb-Johnson:<br>
    </div>
    <blockquote type="cite"
      cite="mid:D27416CD-73D4-48CA-BFC7-0D531E424AF5@webb-johnson.net">
      <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
      This is what I have (<a href="http://api.openvehicles.com"
        class="" moz-do-not-send="true">api.openvehicles.com</a>):
      <div class=""><br class="">
      </div>
      <blockquote style="margin: 0 0 0 40px; border: none; padding:
        0px;" class="">
        <div class="">
          <div class=""><font class="" face="Andale Mono"><span
                style="font-style: normal; font-size: 14px;" class="">==>
                eap/ovms3.ver <==</span></font></div>
          <div class=""><font class="" face="Andale Mono"><span
                style="font-style: normal; font-size: 14px;" class="">3.2.005</span></font></div>
          <div class=""><font class="" face="Andale Mono"><span
                style="font-style: normal; font-size: 14px;" class="">Tue
                Sep 19 08:00:00 UTC 2019 OTA release</span></font></div>
          <div class=""><font class="" face="Andale Mono"><span
                style="font-style: normal; font-size: 14px;" class=""><br
                  class="">
              </span></font></div>
          <div class=""><font class="" face="Andale Mono"><span
                style="font-style: normal; font-size: 14px;" class="">==>
                edge/ovms3.ver <==</span></font></div>
          <div class=""><font class="" face="Andale Mono"><span
                style="font-style: normal; font-size: 14px;" class="">3.2.005-1-g7f86e9c</span></font></div>
          <div class=""><font class="" face="Andale Mono"><span
                style="font-style: normal; font-size: 14px;" class="">Thu
                Sep 19 16:01:18 UTC 2019 Automated build (markhk8)</span></font></div>
          <div class=""><font class="" face="Andale Mono"><span
                style="font-style: normal; font-size: 14px;" class=""><br
                  class="">
              </span></font></div>
          <div class=""><font class="" face="Andale Mono"><span
                style="font-style: normal; font-size: 14px;" class="">==>
                main/ovms3.ver <==</span></font></div>
          <div class=""><font class="" face="Andale Mono"><span
                style="font-style: normal; font-size: 14px;" class="">3.2.002</span></font></div>
          <div class=""><font class="" face="Andale Mono"><span
                style="font-style: normal; font-size: 14px;" class="">Sun
                May 12 08:00:00 UTC 2019 OTA release</span></font></div>
        </div>
      </blockquote>
      <div class="">
        <div><br class="">
        </div>
        <div>The 3.2.005 seems stable, so I think it can now go
          eap->main.</div>
        <div><br class="">
        </div>
        <div>@Michael: Should we co-ordinate and do this later today, or
          have you already released 3.2.005 to main?</div>
        <div><br class="">
        </div>
        <div>Regards, Mark.</div>
        <div><br class="">
          <blockquote type="cite" class="">
            <div class="">On 25 Sep 2019, at 2:21 AM, Bernd Geistert
              <<a href="mailto:b_ghosti@gmx.de" class=""
                moz-do-not-send="true">b_ghosti@gmx.de</a>> wrote:</div>
            <br class="Apple-interchange-newline">
            <div class="">
              <meta http-equiv="Content-Type" content="text/html;
                charset=UTF-8" class="">
              <div text="#000000" bgcolor="#FFFFFF" class="">
                <div class="moz-cite-prefix">IMHO, currently:</div>
                <div class="moz-cite-prefix">- in MAIN is 3.2.002</div>
                <div class="moz-cite-prefix">- in EAP is  3.2.003</div>
                <div class="moz-cite-prefix">- in EDGE is
                  3.2.005-1-g7f86e9c</div>
                <div class="moz-cite-prefix"><br class="">
                </div>
                <div class="moz-cite-prefix"><br class="">
                </div>
                <div class="moz-cite-prefix">Am 19.09.2019 um 10:22
                  schrieb Mark Webb-Johnson:<br class="">
                </div>
                <blockquote type="cite"
                  cite="mid:E0247A1D-DF0C-4056-A394-3D8A9C850617@webb-johnson.net"
                  class="">
                  <meta http-equiv="Content-Type" content="text/html;
                    charset=UTF-8" class="">
                  OK, I’ve built:
                  <div class=""><br class="">
                  </div>
                  <blockquote style="margin: 0 0 0 40px; border: none;
                    padding: 0px;" class="">
                    <div class="">
                      <div class=""><font class="" face="Andale Mono"><span
                            style="font-style: normal; font-size: 14px;"
                            class="">2019-09-19 MWJ  3.2.005  OTA
                            release</span></font></div>
                      <div class=""><font class="" face="Andale Mono"><span
                            style="font-style: normal; font-size: 14px;"
                            class="">- Default module/debug.tasks to
                            FALSE</span></font></div>
                      <div class=""><font class="" face="Andale Mono"><span
                            style="font-style: normal; font-size: 14px;"
                            class="">  Users that volunteer to submit
                            tasks debug historical data to the Open
                            Vehicles</span></font></div>
                      <div class=""><font class="" face="Andale Mono"><span
                            style="font-style: normal; font-size: 14px;"
                            class="">  project, should (with
                            appreciation) set:</span></font></div>
                      <div class=""><font class="" face="Andale Mono"><span
                            style="font-style: normal; font-size: 14px;"
                            class="">    config set module debug.tasks
                            yes</span></font></div>
                      <div class=""><font class="" face="Andale Mono"><span
                            style="font-style: normal; font-size: 14px;"
                            class="">  This will be transmit
                            approximately 700MB of data a month (over
                            cellular/wifi).</span></font></div>
                      <div class=""><font class="" face="Andale Mono"><span
                            style="font-style: normal; font-size: 14px;"
                            class=""><br class="">
                          </span></font></div>
                      <div class=""><font class="" face="Andale Mono"><span
                            style="font-style: normal; font-size: 14px;"
                            class="">2019-09-19 MWJ  3.2.004  OTA
                            release</span></font></div>
                      <div class=""><font class="" face="Andale Mono"><span
                            style="font-style: normal; font-size: 14px;"
                            class="">- Skipped for Chinese superstitous
                            reasons</span></font></div>
                    </div>
                  </blockquote>
                  <div class="">
                    <div class=""><br class="">
                    </div>
                    <div class="">In EAP now, and I will announce.</div>
                    <div class=""><br class="">
                    </div>
                    <div class="">Regards, Mark.</div>
                    <div class=""><br class="">
                      <blockquote type="cite" class="">
                        <div class="">On 19 Sep 2019, at 3:34 PM,
                          Michael Balzer <<a
                            href="mailto:dexter@expeedo.de" class=""
                            moz-do-not-send="true">dexter@expeedo.de</a>>
                          wrote:</div>
                        <br class="Apple-interchange-newline">
                        <div class="">
                          <meta http-equiv="Content-Type"
                            content="text/html; charset=UTF-8" class="">
                          <div text="#000000" bgcolor="#FFFFFF" class="">
                            <div class="moz-cite-prefix">Correct.<br
                                class="">
                              <br class="">
                              Regards,<br class="">
                              Michael<br class="">
                              <br class="">
                              <br class="">
                              Am 19.09.19 um 09:29 schrieb Mark
                              Webb-Johnson:<br class="">
                            </div>
                            <blockquote type="cite"
                              cite="mid:BF141BD4-3322-46A4-86D1-34C6CD613291@webb-johnson.net"
                              class="">
                              <meta http-equiv="Content-Type"
                                content="text/html; charset=UTF-8"
                                class="">
                              I’m just worried about the users who don’t
                              know about this new feature. When they
                              deploy this version, they suddenly start
                              sending 6MB of data a month up to us.
                              <div class=""><br class="">
                              </div>
                              <div class="">I think the ‘fix’ is just to
                                change ovms_module.c:</div>
                              <div class=""><br class="">
                              </div>
                              <blockquote style="margin: 0 0 0 40px;
                                border: none; padding: 0px;" class="">
                                <div class="">MyConfig.GetParamValueBool("module",
                                  "debug.tasks", true)</div>
                                <div class=""><br class="">
                                </div>
                                <div class="">to</div>
                                <div class=""><br class="">
                                </div>
                                <div class="">MyConfig.GetParamValueBool("module",
                                  "debug.tasks", false)</div>
                              </blockquote>
                              <div class="">
                                <div class=""><br class="">
                                </div>
                                <div class="">That would then only
                                  submit these logs for those that
                                  explicitly turn it on?</div>
                                <div class=""><br class="">
                                </div>
                                <div class="">Regards, Mark.</div>
                                <div class=""><br class="">
                                </div>
                                <div class="">
                                  <blockquote type="cite" class="">
                                    <div class="">On 19 Sep 2019, at
                                      3:23 PM, Michael Balzer <<a
                                        href="mailto:dexter@expeedo.de"
                                        class="" moz-do-not-send="true">dexter@expeedo.de</a>>
                                      wrote:</div>
                                    <br
                                      class="Apple-interchange-newline">
                                    <div class="">
                                      <meta http-equiv="Content-Type"
                                        content="text/html;
                                        charset=UTF-8" class="">
                                      <div text="#000000"
                                        bgcolor="#FFFFFF" class="">
                                        <div class="moz-cite-prefix">Sorry,
                                          I didn't think about this
                                          being an issue elsewhere --
                                          german data plans typically
                                          start at minimum 100 MB/month
                                          flat (that's my current plan
                                          at 3 € / month).<br class="">
                                          <br class="">
                                          No need for a new release, it
                                          can be turned off OTA by
                                          issueing<br class="">
                                          <br class="">
                                          <tt class="">config set module
                                            debug.tasks no</tt><br
                                            class="">
                                          <br class="">
                                          Regards,<br class="">
                                          Michael<br class="">
                                          <br class="">
                                          <br class="">
                                          Am 19.09.19 um 09:08 schrieb
                                          Mark Webb-Johnson:<br class="">
                                        </div>
                                        <blockquote type="cite"
                                          cite="mid:BFED773E-0CFE-4EBB-A3B4-9B44FE4A783D@webb-johnson.net"
                                          class="">
                                          <meta
                                            http-equiv="Content-Type"
                                            content="text/html;
                                            charset=UTF-8" class="">
                                          Yep:
                                          <div class=""><br class="">
                                          </div>
                                          <blockquote style="margin: 0 0
                                            0 40px; border: none;
                                            padding: 0px;" class="">
                                            <div class="">758 bytes *
                                              (86400 / 300) * 30 =
                                              6.5MB/month</div>
                                          </blockquote>
                                          <div class="">
                                            <div class=""><br class="">
                                            </div>
                                            <div class="">That is going
                                              over data (not SD).
                                              Presumably cellular data
                                              for a large portion of the
                                              time.</div>
                                            <div class=""><br class="">
                                            </div>
                                            <div class="">I think we
                                              need to default this to
                                              OFF, and make a 3.2.004 to
                                              avoid this becoming an
                                              issue.</div>
                                            <div class=""><br class="">
                                            </div>
                                            <div class="">Regards, Mark.</div>
                                            <div class=""><br class="">
                                              <blockquote type="cite"
                                                class="">
                                                <div class="">On 19 Sep
                                                  2019, at 2:04 PM,
                                                  Stephen Casner <<a
href="mailto:casner@acm.org" class="" moz-do-not-send="true">casner@acm.org</a>>
                                                  wrote:</div>
                                                <br
                                                  class="Apple-interchange-newline">
                                                <div class="">
                                                  <div class="">That's
                                                    6.55MB/month, unless
                                                    you have unusually
                                                    short months!  :-)<br
                                                      class="">
                                                    <br class="">
                                                    In what space is
                                                    that data stored?  A
                                                    log written to SD?
                                                     That's not<br
                                                      class="">
                                                    likely to fill up
                                                    the SD card too
                                                    fast, but what
                                                    happens if no SD
                                                    card<br class="">
                                                    is installed?<br
                                                      class="">
                                                    <br class="">
                                                       -- Steve<br
                                                      class="">
                                                    <br class="">
                                                    On Thu, 19 Sep 2019,
                                                    Mark Webb-Johnson
                                                    wrote:<br class="">
                                                    <br class="">
                                                    <blockquote
                                                      type="cite"
                                                      class="">
                                                      <blockquote
                                                        type="cite"
                                                        class="">    To
                                                        enable CPU usage
                                                        statistics,
                                                        apply the
                                                        changes to
                                                        sdkconfig<br
                                                          class="">
                                                           included.<br
                                                          class="">
                                                           New history
                                                        record:<br
                                                          class="">
                                                           -
                                                        "*-OVM-DebugTasks"
                                                        v1:
                                                        <taskcnt,totaltime>
                                                        + per task:<br
                                                          class="">
       <tasknum,name,state,stack_now,stack_max,stack_total,<br
                                                          class="">
        heap_total,heap_32bit,heap_spi,runtime><br class="">
                                                             Note: CPU
                                                        core use
                                                        percentage =
                                                        runtime /
                                                        totaltime<br
                                                          class="">
                                                      </blockquote>
                                                      <br class="">
                                                      I’ve just noticed
                                                      that this is
                                                      enabled by default
                                                      now (my production
                                                      build has the
                                                      sdkconfig updated,
                                                      as per defaults).<br
                                                        class="">
                                                      <br class="">
                                                      I am seeing 758
                                                      bytes of history
                                                      record, every 5
                                                      minutes. About
                                                      218KB/day, or
                                                      654KB/month.<br
                                                        class="">
                                                      <br class="">
                                                      Should this be
                                                      opt-in?<br
                                                        class="">
                                                      <br class="">
                                                      Regards, Mark.<br
                                                        class="">
                                                      <br class="">
                                                      <blockquote
                                                        type="cite"
                                                        class="">On 8
                                                        Sep 2019, at
                                                        5:43 PM, Michael
                                                        Balzer <<a
                                                          href="mailto:dexter@expeedo.de"
                                                          class=""
                                                          moz-do-not-send="true">dexter@expeedo.de</a>>
                                                        wrote:<br
                                                          class="">
                                                        <br class="">
                                                        I've pushed some
                                                        modifications
                                                        and improvements
                                                        to (hopefully)
                                                        fix the timer
                                                        issue or at
                                                        least be able to
                                                        debug it.<br
                                                          class="">
                                                        <br class="">
                                                        Some sdkconfig
                                                        changes are
                                                        necessary.<br
                                                          class="">
                                                        <br class="">
                                                        The build
                                                        including these
                                                        updates is on my
                                                        edge release as
3.2.002-258-g20ae554b.<br class="">
                                                        <br class="">
                                                        Btw: the network
                                                        restart strategy
                                                        seems to
                                                        mitigate issue
                                                        #241; I've seen
                                                        a major drop on
                                                        record
                                                        repetitions on
                                                        my server since
                                                        the rollout.<br
                                                          class="">
                                                        <br class="">
                                                        <br class="">
                                                        commit
                                                        99e4e48bdd40b7004c0976f51aba9e3da4ecab53<br
                                                          class="">
                                                        <br class="">
                                                           Module: add
                                                        per task CPU
                                                        usage
                                                        statistics, add
                                                        task stats
                                                        history records<br
                                                          class="">
                                                        <br class="">
                                                           To enable CPU
                                                        usage
                                                        statistics,
                                                        apply the
                                                        changes to
                                                        sdkconfig<br
                                                          class="">
                                                           included. The
                                                        CPU usage shown
                                                        by the commands
                                                        is calculated
                                                        against<br
                                                          class="">
                                                           the last task
                                                        status retrieved
                                                        (or system
                                                        boot).<br
                                                          class="">
                                                        <br class="">
                                                           Command
                                                        changes:<br
                                                          class="">
                                                           - "module
                                                        tasks" -- added
                                                        CPU (core) usage
                                                        in percent per
                                                        task<br class="">
                                                        <br class="">
                                                           New command:<br
                                                          class="">
                                                           - "module
                                                        tasks data" --
                                                        output task
                                                        stats in history
                                                        record form<br
                                                          class="">
                                                        <br class="">
                                                           New config:<br
                                                          class="">
                                                           - [module]
                                                        debug.tasks --
                                                        yes (default) =
                                                        send task stats
                                                        every 5 minutes<br
                                                          class="">
                                                        <br class="">
                                                           New history
                                                        record:<br
                                                          class="">
                                                           -
                                                        "*-OVM-DebugTasks"
                                                        v1:
                                                        <taskcnt,totaltime>
                                                        + per task:<br
                                                          class="">
       <tasknum,name,state,stack_now,stack_max,stack_total,<br
                                                          class="">
        heap_total,heap_32bit,heap_spi,runtime><br class="">
                                                             Note: CPU
                                                        core use
                                                        percentage =
                                                        runtime /
                                                        totaltime<br
                                                          class="">
                                                        <br class="">
                                                        commit
                                                        950172c216a72beb4da0bc7a40a46995a6105955<br
                                                          class="">
                                                        <br class="">
                                                           Build config:
                                                        default timer
                                                        service task
                                                        priority raised
                                                        to 20<br
                                                          class="">
                                                        <br class="">
                                                           Background:
                                                        the FreeRTOS
                                                        timer service
                                                        shall only be
                                                        used for very<br
                                                          class="">
                                                             short and
                                                        non-blocking
                                                        jobs. We
                                                        delegate event
                                                        processing to
                                                        our<br class="">
                                                             events
                                                        task, anything
                                                        else in timers
                                                        needs to run
                                                        with high<br
                                                          class="">
                                                             priority.<br
                                                          class="">
                                                        <br class="">
                                                        commit
                                                        31ac19d187480046c16356b80668de45cacbb83d<br
                                                          class="">
                                                        <br class="">
                                                           DukTape: add
                                                        build config for
                                                        task priority,
                                                        default lowered
                                                        to 3<br class="">
                                                        <br class="">
                                                           Background:
                                                        the DukTape
                                                        garbage
                                                        collector shall
                                                        run on lower<br
                                                          class="">
                                                             priority
                                                        than tasks like
                                                        SIMCOM &
                                                        events<br
                                                          class="">
                                                        <br class="">
                                                        commit
                                                        e0a44791fbcfb5a4e4cad24c9d1163b76e637b4f<br
                                                          class="">
                                                        <br class="">
                                                           Server V2:
                                                        use
                                                        esp_log_timestamp
                                                        for timeout
                                                        detection,<br
                                                          class="">
                                                             add timeout
                                                        config, limit
                                                        data records
                                                        & size per
                                                        second<br
                                                          class="">
                                                        <br class="">
                                                           New config:<br
                                                          class="">
                                                           - [server.v2]
                                                        timeout.rx --
                                                        timeout in
                                                        seconds, default
                                                        960<br class="">
                                                        <br class="">
                                                        commit
                                                        684a4ce9525175a910040f0d1ca82ac212fbf5de<br
                                                          class="">
                                                        <br class="">
                                                           Notify: use
                                                        esp_log_timestamp
                                                        for creation
                                                        time instead of
                                                        monotonictime<br
                                                          class="">
                                                             to harden
                                                        against timer
                                                        service
                                                        starvation /
                                                        ticker event
                                                        drops<br
                                                          class="">
                                                        <br class="">
                                                        <br class="">
                                                        Regards,<br
                                                          class="">
                                                        Michael<br
                                                          class="">
                                                        <br class="">
                                                        <br class="">
                                                        Am 07.09.19 um
                                                        10:55 schrieb
                                                        Michael Balzer:<br
                                                          class="">
                                                        <blockquote
                                                          type="cite"
                                                          class="">I
                                                          think the RTOS
                                                          timer service
                                                          task starves.
                                                          It's running
                                                          on core 0 with
                                                          priority 1.<br
                                                          class="">
                                                          <br class="">
                                                          Taks on core 0
                                                          sorted by
                                                          priority:<br
                                                          class="">
                                                          <br class="">
                                                          Number of
                                                          Tasks = 20
                                                               Stack:
                                                           Now   Max
                                                          Total    Heap
                                                          32-bit SPIRAM
                                                          C# PRI<br
                                                          class="">
                                                          3FFC84A8  6
                                                          Blk ipc0
                                                                       388
                                                            500  1024
                                                             7788      0
                                                               0  0  24<br
                                                          class="">
                                                          3FFC77F0  5
                                                          Blk OVMS CanRx
                                                                 428
                                                            428  2048
                                                             3052      0
                                                           31844  0  23<br
                                                          class="">
                                                          3FFAFBF4  1
                                                          Blk esp_timer
                                                                  400
                                                            656  4096
                                                            35928    644
                                                           25804  0  22<br
                                                          class="">
                                                          3FFD3240 19
                                                          Blk wifi
                                                                       460
                                                           2716  3584
                                                            43720      0
                                                              20  0  22<br
                                                          class="">
                                                          3FFC03C4  2
                                                          Blk eventTask
                                                                  448
                                                           1984  4608
                                                              104      0
                                                               0  0  20<br
                                                          class="">
                                                          3FFC8F14 17
                                                          Blk tiT
                                                                        500
                                                           2308  3072
                                                             6552      0
                                                               0  *  18<br
                                                          class="">
                                                          3FFE14F0 26
                                                          Blk OVMS COrx
                                                                  456
                                                            456  4096
                                                                0      0
                                                               0  0   7<br
                                                          class="">
                                                          3FFE19D4 27
                                                          Blk OVMS COwrk
                                                                 476
                                                            476  3072
                                                                0      0
                                                               0  0   7<br
                                                          class="">
                                                          3FFCBC34 12
                                                          Blk Tmr Svc
                                                                    352
                                                            928  3072
                                                               88      0
                                                               0  0   1<br
                                                          class="">
                                                          3FFE7708 23
                                                          Blk mdns
                                                                       468
                                                           1396  4096
                                                              108      0
                                                               0  0   1<br
                                                          class="">
                                                          <br class="">
                                                          I don't think
                                                          it's our
                                                          CanRx, as that
                                                          only fetches
                                                          and queues CAN
                                                          frames, the
                                                          actual work is
                                                          done by the
                                                          listeners. The
                                                          CO tasks only
                                                          run for
                                                          CANopen jobs,
                                                          which are few
                                                          for normal
                                                          operation.<br
                                                          class="">
                                                          <br class="">
                                                          That leaves
                                                          the system
                                                          tasks, with
                                                          main suspect
                                                          -once again-
                                                          the wifi blob.<br
                                                          class="">
                                                          <br class="">
                                                          We need to
                                                          know how much
                                                          CPU time the
                                                          tasks actually
                                                          use now. I
                                                          think I saw
                                                          some option
                                                          for this in
                                                          the FreeRTOS
                                                          config.<br
                                                          class="">
                                                          <br class="">
                                                          Regards,<br
                                                          class="">
                                                          Michael<br
                                                          class="">
                                                          <br class="">
                                                          <br class="">
                                                          Am 06.09.19 um
                                                          23:15 schrieb
                                                          Michael
                                                          Balzer:<br
                                                          class="">
                                                          <blockquote
                                                          type="cite"
                                                          class="">The
                                                          workaround is
                                                          based on the
                                                          monotonictime
                                                          being updated
                                                          per second, as
                                                          do the history
                                                          record
                                                          offsets.<br
                                                          class="">
                                                          <br class="">
                                                          Apparently,
                                                          that mechanism
                                                          doesn't work
                                                          reliably. That
                                                          may be an
                                                          indicator for
                                                          some bigger
                                                          underlying
                                                          issue.<br
                                                          class="">
                                                          <br class="">
                                                          Example log
                                                          excerpt:<br
                                                          class="">
                                                          <br class="">
                                                          2019-09-06
                                                          22:07:48.126919
                                                          +0200 info
                                                           main: #173 C
                                                          MITPROHB rx
                                                          msg h
964,0,RT-BAT-C,5,86400,2,1,3830,3795,3830,-10,25,25,25,0<br class="">
                                                          2019-09-06
                                                          22:09:03.089031
                                                          +0200 info
                                                           main: #173 C
                                                          MITPROHB rx
                                                          msg h
964,-10,RT-BAT-C,5,86400,2,1,3830,3795,3830,-10,25,25,25,0<br class="">
                                                          2019-09-06
                                                          22:09:05.041574
                                                          +0200 info
                                                           main: #173 C
                                                          MITPROHB rx
                                                          msg h
964,-20,RT-BAT-C,5,86400,2,1,3830,3795,3830,-10,25,25,25,0<br class="">
                                                          2019-09-06
                                                          22:09:05.052644
                                                          +0200 info
                                                           main: #173 C
                                                          MITPROHB rx
                                                          msg h
964,-30,RT-BAT-C,5,86400,2,1,3830,3795,3830,-10,25,25,25,0<br class="">
                                                          2019-09-06
                                                          22:09:05.063617
                                                          +0200 info
                                                           main: #173 C
                                                          MITPROHB rx
                                                          msg h
964,-49,RT-BAT-C,5,86400,2,1,3830,3795,3830,-10,25,25,25,0<br class="">
                                                          2019-09-06
                                                          22:09:05.077527
                                                          +0200 info
                                                           main: #173 C
                                                          MITPROHB rx
                                                          msg h
964,-59,RT-BAT-C,5,86400,2,1,3830,3795,3830,-10,25,25,25,0<br class="">
                                                          2019-09-06
                                                          22:09:05.193775
                                                          +0200 info
                                                           main: #173 C
                                                          MITPROHB rx
                                                          msg h
964,-70,RT-BAT-C,5,86400,2,1,3830,3795,3830,-10,25,25,25,0<br class="">
                                                          2019-09-06
                                                          22:09:13.190645
                                                          +0200 info
                                                           main: #173 C
                                                          MITPROHB rx
                                                          msg h
964,-80,RT-BAT-C,5,86400,2,1,3830,3795,3830,-10,25,25,25,0<br class="">
                                                          2019-09-06
                                                          22:09:22.077994
                                                          +0200 info
                                                           main: #173 C
                                                          MITPROHB rx
                                                          msg h
964,-90,RT-BAT-C,5,86400,2,1,3830,3795,3830,-10,25,25,25,0<br class="">
                                                          2019-09-06
                                                          22:09:54.590300
                                                          +0200 info
                                                           main: #173 C
                                                          MITPROHB rx
                                                          msg h
964,-109,RT-BAT-C,5,86400,2,1,3830,3795,3830,-10,25,25,25,0<br class="">
                                                          2019-09-06
                                                          22:11:10.127054
                                                          +0200 info
                                                           main: #173 C
                                                          MITPROHB rx
                                                          msg h
964,-119,RT-BAT-C,5,86400,2,1,3830,3795,3830,-10,25,25,25,0<br class="">
                                                          2019-09-06
                                                          22:11:16.794200
                                                          +0200 info
                                                           main: #173 C
                                                          MITPROHB rx
                                                          msg h
964,-130,RT-BAT-C,5,86400,2,1,3830,3795,3830,-10,25,25,25,0<br class="">
                                                          2019-09-06
                                                          22:11:22.455652
                                                          +0200 info
                                                           main: #173 C
                                                          MITPROHB rx
                                                          msg h
964,-140,RT-BAT-C,5,86400,2,1,3830,3795,3830,-10,25,25,25,0<br class="">
                                                          2019-09-06
                                                          22:12:49.423412
                                                          +0200 info
                                                           main: #173 C
                                                          MITPROHB rx
                                                          msg h
964,-150,RT-BAT-C,5,86400,2,1,3830,3795,3830,-10,25,25,25,0<br class="">
                                                          2019-09-06
                                                          22:12:49.442096
                                                          +0200 info
                                                           main: #173 C
                                                          MITPROHB rx
                                                          msg h
964,-169,RT-BAT-C,5,86400,2,1,3830,3795,3830,-10,25,25,25,0<br class="">
                                                          2019-09-06
                                                          22:12:49.461941
                                                          +0200 info
                                                           main: #173 C
                                                          MITPROHB rx
                                                          msg h
964,-179,RT-BAT-C,5,86400,2,1,3830,3795,3830,-10,25,25,25,0<br class="">
                                                          2019-09-06
                                                          22:14:39.828133
                                                          +0200 info
                                                           main: #173 C
                                                          MITPROHB rx
                                                          msg h
964,-190,RT-BAT-C,5,86400,2,1,3830,3795,3830,-10,25,25,25,0<br class="">
                                                          2019-09-06
                                                          22:14:39.858144
                                                          +0200 info
                                                           main: #173 C
                                                          MITPROHB rx
                                                          msg h
964,-200,RT-BAT-C,5,86400,2,1,3830,3795,3830,-10,25,25,25,0<br class="">
                                                          2019-09-06
                                                          22:14:52.020319
                                                          +0200 info
                                                           main: #173 C
                                                          MITPROHB rx
                                                          msg h
964,-210,RT-BAT-C,5,86400,2,1,3830,3795,3830,-10,25,25,25,0<br class="">
                                                          2019-09-06
                                                          22:14:54.452637
                                                          +0200 info
                                                           main: #173 C
                                                          MITPROHB rx
                                                          msg h
964,-229,RT-BAT-C,5,86400,2,1,3830,3795,3830,-10,25,25,25,0<br class="">
                                                          2019-09-06
                                                          22:15:12.613935
                                                          +0200 info
                                                           main: #173 C
                                                          MITPROHB rx
                                                          msg h
964,-239,RT-BAT-C,5,86400,2,1,3830,3795,3830,-10,25,25,25,0<br class="">
                                                          2019-09-06
                                                          22:15:35.223845
                                                          +0200 info
                                                           main: #173 C
                                                          MITPROHB rx
                                                          msg h
964,-250,RT-BAT-C,5,86400,2,1,3830,3795,3830,-10,25,25,25,0<br class="">
                                                          2019-09-06
                                                          22:16:09.255059
                                                          +0200 info
                                                           main: #173 C
                                                          MITPROHB rx
                                                          msg h
964,-260,RT-BAT-C,5,86400,2,1,3830,3795,3830,-10,25,25,25,0<br class="">
                                                          2019-09-06
                                                          22:17:31.919754
                                                          +0200 info
                                                           main: #173 C
                                                          MITPROHB rx
                                                          msg h
964,-270,RT-BAT-C,5,86400,2,1,3830,3795,3830,-10,25,25,25,0<br class="">
                                                          2019-09-06
                                                          22:19:23.366267
                                                          +0200 info
                                                           main: #173 C
                                                          MITPROHB rx
                                                          msg h
964,-289,RT-BAT-C,5,86400,2,1,3830,3795,3830,-10,25,25,25,0<br class="">
                                                          2019-09-06
                                                          22:21:57.344609
                                                          +0200 info
                                                           main: #173 C
                                                          MITPROHB rx
                                                          msg h
964,-299,RT-BAT-C,5,86400,2,1,3830,3795,3830,-10,25,25,25,0<br class="">
                                                          2019-09-06
                                                          22:23:40.082406
                                                          +0200 info
                                                           main: #31 C
                                                          MITPROHB rx
                                                          msg h
964,-1027,RT-BAT-C,5,86400,2,1,3830,3795,3830,-10,25,25,25,0<br class="">
                                                          2019-09-06
                                                          22:25:58.061883
                                                          +0200 info
                                                           main: #31 C
                                                          MITPROHB rx
                                                          msg h
964,-1040,RT-BAT-C,5,86400,2,1,3830,3795,3830,-10,25,25,25,0<br class="">
                                                          <br class="">
                                                          <br class="">
                                                          This shows the
                                                          ticker was
                                                          only run 299
                                                          times from
                                                          22:07:48 to
                                                          22:21:57.<br
                                                          class="">
                                                          <br class="">
                                                          After 22:21:57
                                                          the workaround
                                                          was triggered
                                                          and did a
                                                          reconnect.
                                                          Apparently
                                                          during that
                                                          network
                                                          reinitialization
                                                          of 103
                                                          seconds, the
                                                          per second
                                                          ticker was run
                                                          628 times.<br
                                                          class="">
                                                          <br class="">
                                                          That can't be
                                                          catching up on
                                                          the event
                                                          queue, as that
                                                          queue has only
                                                          20 slots. So
                                                          something
                                                          strange is
                                                          going on here.<br
                                                          class="">
                                                          <br class="">
                                                          Any ideas?<br
                                                          class="">
                                                          <br class="">
                                                          Regards,<br
                                                          class="">
                                                          Michael<br
                                                          class="">
                                                          <br class="">
                                                          <br class="">
                                                          Am 06.09.19 um
                                                          08:04 schrieb
                                                          Michael
                                                          Balzer:<br
                                                          class="">
                                                          <blockquote
                                                          type="cite"
                                                          class="">Mark
                                                          & anyone
                                                          else running a
                                                          V2 server,<br
                                                          class="">
                                                          <br class="">
                                                          as most cars
                                                          don't send
                                                          history
                                                          records, this
                                                          also needs the
                                                          change to the
                                                          server I just
                                                          pushed, i.e.
                                                          server version
                                                          2.4.2.<br
                                                          class="">
                                                          <br class="">
                                                          <a
href="https://github.com/openvehicles/Open-Vehicle-Monitoring-System/commits/master"
                                                          class=""
                                                          moz-do-not-send="true">https://github.com/openvehicles/Open-Vehicle-Monitoring-System/commits/master</a>
                                                          <<a
href="https://github.com/openvehicles/Open-Vehicle-Monitoring-System/commits/master"
                                                          class=""
                                                          moz-do-not-send="true">https://github.com/openvehicles/Open-Vehicle-Monitoring-System/commits/master</a>><br
                                                          class="">
                                                          <br class="">
                                                          Regards,<br
                                                          class="">
                                                          Michael<br
                                                          class="">
                                                          <br class="">
                                                          <br class="">
                                                          Am 05.09.19 um
                                                          19:55 schrieb
                                                          Michael
                                                          Balzer:<br
                                                          class="">
                                                          <blockquote
                                                          type="cite"
                                                          class="">I've
                                                          pushed the
                                                          nasty
                                                          workaround:
                                                          the v2 server
                                                          checks for no
                                                          RX over 15
                                                          minutes, then
                                                          restarts the
                                                          network (wifi
                                                          & modem)
                                                          as configured
                                                          for autostart.<br
                                                          class="">
                                                          <br class="">
                                                          Rolled out on
                                                          my server in
                                                          edge as
                                                          3.2.002-237-ge075f655.<br
                                                          class="">
                                                          <br class="">
                                                          Please test.<br
                                                          class="">
                                                          <br class="">
                                                          Regards,<br
                                                          class="">
                                                          Michael<br
                                                          class="">
                                                          <br class="">
                                                          <br class="">
                                                          Am 05.09.19 um
                                                          01:58 schrieb
                                                          Mark
                                                          Webb-Johnson:<br
                                                          class="">
                                                          <blockquote
                                                          type="cite"
                                                          class="">
                                                          <blockquote
                                                          type="cite"
                                                          class="">Mark,
                                                          you can check
                                                          your server
                                                          logs for
                                                          history
                                                          messages with
                                                          ridiculous
                                                          time offsets:<br
                                                          class="">
                                                          [sddexter@ns27
                                                          server]$ cat
                                                          log-20190903 |
                                                          egrep "rx msg
                                                          h
[0-9]+,-[0-9]{4}" | wc -l<br class="">
                                                          455283<br
                                                          class="">
                                                          </blockquote>
                                                          <br class="">
                                                          I checked my
                                                          logs and see
                                                          12 vehicles
                                                          showing this.
                                                          But, 2 only
                                                          show this for
                                                          a debugcrash
                                                          log (which is
                                                          expected, I
                                                          guess, if the
                                                          time is not
                                                          synced at
                                                          report time).
                                                          I’ve got 4
                                                          cars with the
                                                          offset >
                                                          10,000.<br
                                                          class="">
                                                          <br class="">
                                                          Regards, Mark.<br
                                                          class="">
                                                          <br class="">
                                                          <blockquote
                                                          type="cite"
                                                          class="">On 4
                                                          Sep 2019, at
                                                          4:45 AM,
                                                          Michael Balzer
                                                          <<a
                                                          href="mailto:dexter@expeedo.de"
                                                          class=""
                                                          moz-do-not-send="true">dexter@expeedo.de</a>
                                                          <<a
                                                          href="mailto:dexter@expeedo.de"
                                                          class=""
                                                          moz-do-not-send="true">mailto:dexter@expeedo.de</a>>>
                                                          wrote:<br
                                                          class="">
                                                          <br class="">
                                                          Everyone,<br
                                                          class="">
                                                          <br class="">
                                                          I've pushed a
                                                          change that
                                                          needs some
                                                          testing.<br
                                                          class="">
                                                          <br class="">
                                                          I had the
                                                          issue myself
                                                          now parking at
                                                          a certain
                                                          distance from
                                                          my garage wifi
                                                          AP, i.e. on
                                                          the edge of
                                                          "in", after
                                                          wifi had been
                                                          disconnected
                                                          for some
                                                          hours, and
                                                          with the
                                                          module still
                                                          connected via
                                                          modem. The
                                                          wifi blob had
                                                          been trying to
                                                          connect to the
                                                          AP for about
                                                          two hours.<br
                                                          class="">
                                                          <br class="">
                                                          As seen
                                                          before, the
                                                          module saw no
                                                          error, just
                                                          the server
                                                          responses and
                                                          commands
                                                          stopped coming
                                                          in. I noticed
                                                          the default
                                                          interface was
                                                          still "st1"
                                                          despite wifi
                                                          having been
                                                          disconnected
                                                          and modem
                                                          connected. The
                                                          DNS was also
                                                          still
                                                          configured for
                                                          my wifi
                                                          network, and
                                                          the interface
                                                          seemed to have
                                                          an IP address
                                                          -- but wasn't
                                                          pingable from
                                                          the wifi
                                                          network.<br
                                                          class="">
                                                          <br class="">
                                                          A power cycle
                                                          of the modem
                                                          solved the
                                                          issue without
                                                          reboot. So the
                                                          cause may be
                                                          in the
                                                          modem/ppp
                                                          subsystem, or
                                                          it may be
                                                          related (in
                                                          some weird
                                                          way) to the
                                                          default
                                                          interface /
                                                          DNS setup.<br
                                                          class="">
                                                          <br class="">
                                                          More tests
                                                          showed the
                                                          default
                                                          interface
                                                          again/still
                                                          got set by the
                                                          wifi blob
                                                          itself at some
                                                          point,
                                                          overriding our
                                                          modem
                                                          prioritization.
                                                          The events we
                                                          didn't handle
                                                          up to now were
"sta.connected" and "sta.lostip", so I added these, and the bug didn't
                                                          show up again
                                                          since then.
                                                          That doesn't
                                                          mean anything,
                                                          so we need to
                                                          test this.<br
                                                          class="">
                                                          <br class="">
                                                          The default
                                                          interface
                                                          really
                                                          shouldn't
                                                          affect inbound
                                                          packet routing
                                                          of an
                                                          established
                                                          connection,
                                                          but there
                                                          always may be
                                                          strange bugs
                                                          lurking in
                                                          those libs.<br
                                                          class="">
                                                          <br class="">
                                                          The change
                                                          also
                                                          reimplements
                                                          the wifi
                                                          signal
                                                          strength
                                                          reading, as
                                                          the tests also
                                                          showed that
                                                          still wasn't
                                                          working well
                                                          using the CSI
                                                          callback. It
                                                          now seems to
                                                          be much more
                                                          reliable.<br
                                                          class="">
                                                          <br class="">
                                                          Please test
                                                          & report.
                                                          The single
                                                          module will be
                                                          hard to test,
                                                          as the bug
                                                          isn't
                                                          reproducable
                                                          easily, but
                                                          you can still
                                                          try if wifi /
                                                          modem
                                                          transitions
                                                          work well.<br
                                                          class="">
                                                          <br class="">
                                                          Mark, you can
                                                          check your
                                                          server logs
                                                          for history
                                                          messages with
                                                          ridiculous
                                                          time offsets:<br
                                                          class="">
                                                          [sddexter@ns27
                                                          server]$ cat
                                                          log-20190903 |
                                                          egrep "rx msg
                                                          h
[0-9]+,-[0-9]{4}" | wc -l<br class="">
                                                          455283<br
                                                          class="">
                                                          The bug now
                                                          severely
                                                          affects the V2
                                                          server
                                                          performance,
                                                          as the server
                                                          is single
                                                          threaded and
                                                          doesn't scale
                                                          very well to
                                                          this kind of
                                                          bulk data
                                                          bursts,
                                                          especially
                                                          when coming
                                                          from multiple
                                                          modules in
                                                          parallel. So
                                                          we really need
                                                          to solve this
                                                          now. Slow
                                                          reactions or
                                                          connection
                                                          drops from my
                                                          server lately
                                                          have been due
                                                          to this bug.
                                                          If this change
                                                          doesn't solve
                                                          it, we'll need
                                                          to add some
                                                          reboot trigger
                                                          on "too many
                                                          server v2
                                                          notification
                                                          retransmissions"
                                                          -- or maybe a
                                                          modem power
                                                          cycle will do,
                                                          that wouldn't
                                                          discard the
                                                          data.<br
                                                          class="">
                                                          <br class="">
                                                          Thanks,<br
                                                          class="">
                                                          Michael<br
                                                          class="">
                                                          <br class="">
                                                          <br class="">
                                                          Am 03.09.19 um
                                                          07:46 schrieb
                                                          Mark
                                                          Webb-Johnson:<br
                                                          class="">
                                                          <blockquote
                                                          type="cite"
                                                          class="">No
                                                          problem. We
                                                          can hold. I
                                                          won’t commit
                                                          anything for
                                                          the next few
                                                          days (and
                                                          agree to
                                                          hold-off on
                                                          Markos’s
                                                          pull). Let me
                                                          know when you
                                                          are ready.<br
                                                          class="">
                                                          <br class="">
                                                          Regards, Mark.<br
                                                          class="">
                                                          <br class="">
                                                          <blockquote
                                                          type="cite"
                                                          class="">On 3
                                                          Sep 2019, at
                                                          1:58 AM,
                                                          Michael Balzer
                                                          <<a
                                                          href="mailto:dexter@expeedo.de"
                                                          class=""
                                                          moz-do-not-send="true">dexter@expeedo.de</a>>
                                                          <<a
                                                          href="mailto:dexter@expeedo.de"
                                                          class=""
                                                          moz-do-not-send="true">mailto:dexter@expeedo.de</a>>
                                                          wrote:<br
                                                          class="">
                                                          <br class="">
                                                          Mark, please
                                                          wait.<br
                                                          class="">
                                                          <br class="">
                                                          I may just
                                                          have found the
                                                          cause for
                                                          issue #241, or
                                                          at least
                                                          something I
                                                          need to
                                                          investigate
                                                          before
                                                          releasing.<br
                                                          class="">
                                                          <br class="">
                                                          I need to dig
                                                          into my logs
                                                          first, and try
                                                          something.<br
                                                          class="">
                                                          <br class="">
                                                          Regards,<br
                                                          class="">
                                                          Michael<br
                                                          class="">
                                                          <br class="">
                                                          <br class="">
                                                          Am 02.09.19 um
                                                          12:23 schrieb
                                                          Michael
                                                          Balzer:<br
                                                          class="">
                                                          <blockquote
                                                          type="cite"
                                                          class="">Nothing
                                                          open from my
                                                          side at the
                                                          moment.<br
                                                          class="">
                                                          <br class="">
                                                          I haven't had
                                                          the time to
                                                          look in to
                                                          Markos pull
                                                          request, but
                                                          from a first
                                                          check also
                                                          think that's
                                                          going too deep
                                                          to be included
                                                          in this
                                                          release.<br
                                                          class="">
                                                          <br class="">
                                                          Regards,<br
                                                          class="">
                                                          Michael<br
                                                          class="">
                                                          <br class="">
                                                          <br class="">
                                                          Am 02.09.19 um
                                                          04:15 schrieb
                                                          Mark
                                                          Webb-Johnson:<br
                                                          class="">
                                                          <blockquote
                                                          type="cite"
                                                          class="">I
                                                          think it is
                                                          well past time
                                                          for a 3.2.003
                                                          release.
                                                          Things seems
                                                          table in edge
                                                          (although some
                                                          things only
                                                          partially
                                                          implemented).<br
                                                          class="">
                                                          <br class="">
                                                          Anything
                                                          people want to
                                                          include at the
                                                          last minute,
                                                          or can we go
                                                          ahead and
                                                          build?<br
                                                          class="">
                                                          <br class="">
                                                          Regards, Mark.<br
                                                          class="">
                                                          <br class="">
                                                          </blockquote>
                                                          </blockquote>
                                                          </blockquote>
                                                          </blockquote>
                                                          </blockquote>
                                                          </blockquote>
                                                          </blockquote>
                                                          </blockquote>
                                                          </blockquote>
                                                        </blockquote>
                                                      </blockquote>
                                                    </blockquote>
                                                  </div>
                                                </div>
                                              </blockquote>
                                            </div>
                                          </div>
                                        </blockquote>
                                        <br>
                                      </div>
                                    </div>
                                  </blockquote>
                                </div>
                              </div>
                            </blockquote>
                          </div>
                        </div>
                      </blockquote>
                    </div>
                  </div>
                </blockquote>
              </div>
            </div>
          </blockquote>
        </div>
      </div>
    </blockquote>
    <br>
    <pre class="moz-signature" cols="144">-- 
Michael Balzer * Helkenberger Weg 9 * D-58256 Ennepetal
Fon 02333 / 833 5735 * Handy 0176 / 206 989 26
</pre>
  </body>
</html>