<html>
  <head>
    <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
  </head>
  <body text="#000000" bgcolor="#FFFFFF">
    No, the 32 bit limit indeed has an effect pretty soon. I had assumed
    they are "ticks", but they really are at the timer resolution (see
    DebugTasks records).<br>
    <br>
    Timer resolution is currently 1 MHz, so the counters will cycle over
    every 71.6 minutes.<br>
    <br>
    So if you don't enable the debug.tasks report, the fist manual
    report is likely to look strange.<br>
    <br>
    We can get a checkpoint every five minutes regardless of the report
    activation. Or we can add a measurement delay to the command.<br>
    <br>
    Regards,<br>
    Michael<br>
    <br>
    <br>
    <div class="moz-cite-prefix">Am 20.09.19 um 16:16 schrieb Michael
      Balzer:<br>
    </div>
    <blockquote type="cite"
      cite="mid:2f66409f-ef10-b513-5674-e0e740169da7@expeedo.de">
      <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
      The percentages are calculated from the time slice counter
      differences between two command invocations (or boot), but their
      32 bit limit shouldn't have an effect that soon.<br>
      <br>
      Also strange: your esp_timer SPIRAM allocation…<br>
      <br>
      Is that a build with toolkit -98?<br>
      <br>
      Regards,<br>
      Michael<br>
      <br>
      <br>
      <div class="moz-cite-prefix">Am 20.09.19 um 03:29 schrieb Mark
        Webb-Johnson:<br>
      </div>
      <blockquote type="cite"
        cite="mid:0D4D450E-631F-4A4C-B043-12633A4C7BBF@webb-johnson.net">
        <meta http-equiv="Content-Type" content="text/html;
          charset=UTF-8">
        Strange. Connected to my car, and from ovms_shell asked for
        ‘module tasks’:
        <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="">ovms>
                  module tasks</span></font></div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class="">...</span></font></div>
          </div>
          <div class="">
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class="">Vehicle
                  Response:</span></font></div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class=""> 
                  Number of Tasks = 17        Stack:  Now   Max Total  
                   Heap 32-bit SPIRAM C# PRI CPU%</span></font></div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class=""> 
                  3FFAFBF4    1 Blk esp_timer         400   656  4096  
                  35944    6442122956  0  22   2%</span></font></div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class=""> 
                  3FFBDBBC    2 Blk eventTask         440  1928  4608  
                    140      0      0  0  20   0%</span></font></div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class=""> 
                  3FFBFF00    3 Blk OVMS Events       444  3404  8192  
                  70816      0  21092  1   5   9%</span></font></div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class=""> 
                  3FFC3300    4 Blk OVMS DukTape      460  9548 12288  
                    188      0  31020  1   3  12%</span></font></div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class=""> 
                  3FFC4FF4    5 Blk OVMS CanRx        432   832  2048  
                   3144      0  32072  0  23 120%</span></font></div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class=""> 
                  3FFC5D3C    6 Blk ipc0              392   504  1024  
                   7804      0      0  0  24   2%</span></font></div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class=""> 
                  3FFC6340    7 Blk ipc1              396   444  1024  
                     12      0      0  1  24   0%</span></font></div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class=""> 
                  3FFC8194   10 Rdy IDLE0             416   512  1024  
                      0      0      0  0   0 362%</span></font></div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class=""> 
                  3FFC872C   11 Rdy IDLE1             408   504  1024  
                      0      0      0  1   0 387%</span></font></div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class=""> 
                  3FFC94C4   12 Blk Tmr Svc           352   912  3072  
                     88      0      0  0  20   1%</span></font></div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class=""> 
                  3FFC6794   17 Blk tiT               504  2536  3072  
                   7668      0      0  *  18   6%</span></font></div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class=""> 
                  3FFCF3E0   18 Blk OVMS SIMCOM       460  2412  4096  
                     40      0   3892  1   5   1%</span></font></div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class=""> 
                  3FFD0C10   19 Blk wifi              460  2716  3584  
                  36356      0   2912  0  22  23%</span></font></div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class=""> 
                  3FFDBD70   20 Blk OVMS Vehicle      460  1308  6144  
                      0      0   2336  1  10  95%</span></font></div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class=""> 
                  3FFDF0FC   21 Rdy OVMS Console      488  1736  6144  
                      0      0     20  1   5   0%</span></font></div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class=""> 
                  3FFDF64C   22 Blk mdns              472  1576  4096  
                    108      0      0  0   1   1%</span></font></div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class=""> 
                  3FFE38F0   35 Rdy OVMS NetMan      1384  2984  8192  
                    144  25180   5004  1   5   3%</span></font></div>
          </div>
          <div class=""><font class="" face="Andale Mono"><span
                style="font-style: normal; font-size: 14px;" class=""><br
                  class="">
              </span></font></div>
          <div class="">
            <div class="">
              <div class=""><font class="" face="Andale Mono"><span
                    style="font-style: normal; font-size: 14px;"
                    class="">ovms> module tasks</span></font></div>
              <div class=""><font class="" face="Andale Mono"><span
                    style="font-style: normal; font-size: 14px;"
                    class="">...</span></font></div>
            </div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class="">Vehicle
                  Response:</span></font></div>
          </div>
          <div class="">
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class=""> 
                  Number of Tasks = 17        Stack:  Now   Max Total  
                   Heap 32-bit SPIRAM C# PRI CPU%</span></font></div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class=""> 
                  3FFAFBF4    1 Blk esp_timer         400   656  4096  
                  35944    6442122956  0  22   0%</span></font></div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class=""> 
                  3FFBDBBC    2 Blk eventTask         440  1928  4608  
                    140      0      0  0  20   0%</span></font></div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class=""> 
                  3FFBFF00    3 Blk OVMS Events       444  3404  8192  
                  70816      0  21092  1   5   0%</span></font></div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class=""> 
                  3FFC3300    4 Blk OVMS DukTape      460  9548 12288  
                    188      0  23712  1   3   1%</span></font></div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class=""> 
                  3FFC4FF4    5 Blk OVMS CanRx        432   832  2048  
                   3144      0  32072  0  23   4%</span></font></div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class=""> 
                  3FFC5D3C    6 Blk ipc0              392   504  1024  
                   7804      0      0  0  24   0%</span></font></div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class=""> 
                  3FFC6340    7 Blk ipc1              396   444  1024  
                     12      0      0  1  24   0%</span></font></div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class=""> 
                  3FFC8194   10 Rdy IDLE0             416   512  1024  
                      0      0      0  0   0  96%</span></font></div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class=""> 
                  3FFC872C   11 Rdy IDLE1             408   504  1024  
                      0      0      0  1   0  95%</span></font></div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class=""> 
                  3FFC94C4   12 Blk Tmr Svc           352   912  3072  
                     88      0      0  0  20   0%</span></font></div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class=""> 
                  3FFC6794   17 Blk tiT               504  2536  3072  
                   7668      0      0  *  18   0%</span></font></div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class=""> 
                  3FFCF3E0   18 Blk OVMS SIMCOM       668  2412  4096  
                     40      0   3892  1   5   0%</span></font></div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class=""> 
                  3FFD0C10   19 Blk wifi              460  2716  3584  
                  36356      0   2912  0  22   0%</span></font></div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class=""> 
                  3FFDBD70   20 Blk OVMS Vehicle      460  1308  6144  
                      0      0   2336  1  10   3%</span></font></div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class=""> 
                  3FFDF0FC   21 Blk OVMS Console      552  1736  6144  
                      0      0     20  1   5   0%</span></font></div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class=""> 
                  3FFDF64C   22 Blk mdns              472  1576  4096  
                    108      0      0  0   1   0%</span></font></div>
            <div class=""><font class="" face="Andale Mono"><span
                  style="font-style: normal; font-size: 14px;" class=""> 
                  3FFE38F0   35 Rdy OVMS NetMan      1896  3240  8192  
                    400  25180   5392  1   5   0%</span></font></div>
          </div>
        </blockquote>
        <div class="">
          <div><br class="">
          </div>
          <div>The second one is what I normally see, but the first is
            bizarre. Any idea what the time frame for the CPU% is?</div>
          <div><br class="">
          </div>
          <div>Regards, Mark</div>
          <div><br class="">
          </div>
          <div>
            <blockquote type="cite" class="">
              <div class="">
                <div text="#000000" bgcolor="#FFFFFF" class="">
                  <blockquote type="cite" class="">
                    <blockquote type="cite" class="">
                      <blockquote type="cite" class="">
                        <div class="">
                          <div class="">
                            <blockquote type="cite" class="">
                              <div class="">
                                <div class="">
                                  <blockquote type="cite" class=""> </blockquote>
                                </div>
                              </div>
                            </blockquote>
                          </div>
                        </div>
                      </blockquote>
                    </blockquote>
                    <blockquote type="cite"
                      cite="mid:7a3bb65a-9a54-6f67-c794-4b6f8d4ca2b8@expeedo.de"
                      class=""> </blockquote>
                  </blockquote>
                  <blockquote type="cite"
                    cite="mid:76a60138-8d46-0703-d9d0-6b2f5daa20da@expeedo.de"
                    class="">
                    <blockquote type="cite"
                      cite="mid:7a3bb65a-9a54-6f67-c794-4b6f8d4ca2b8@expeedo.de"
                      class=""> </blockquote>
                  </blockquote>
                </div>
              </div>
            </blockquote>
            <div class="">
              <div text="#000000" bgcolor="#FFFFFF" class="">
                <blockquote type="cite"
                  cite="mid:7a3bb65a-9a54-6f67-c794-4b6f8d4ca2b8@expeedo.de"
                  class="">
                  <blockquote type="cite"
                    cite="mid:BFED773E-0CFE-4EBB-A3B4-9B44FE4A783D@webb-johnson.net"
                    class="">
                    <div class="">
                      <div class="">
                        <blockquote type="cite" class="">
                          <div class="">
                            <div class="">
                              <blockquote type="cite" 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 class="">
                  <pre class="moz-signature" cols="144">-- 
Michael Balzer * Helkenberger Weg 9 * D-58256 Ennepetal
Fon 02333 / 833 5735 * Handy 0176 / 206 989 26</pre>
                </blockquote>
              </div>
            </div>
          </div>
        </div>
        <br>
        <fieldset class="mimeAttachmentHeader"></fieldset>
        <pre class="moz-quote-pre" wrap="">_______________________________________________
OvmsDev mailing list
<a class="moz-txt-link-abbreviated" href="mailto:OvmsDev@lists.openvehicles.com" moz-do-not-send="true">OvmsDev@lists.openvehicles.com</a>
<a class="moz-txt-link-freetext" href="http://lists.openvehicles.com/mailman/listinfo/ovmsdev" moz-do-not-send="true">http://lists.openvehicles.com/mailman/listinfo/ovmsdev</a>
</pre>
      </blockquote>
      <br>
      <pre class="moz-signature" cols="160">-- 
Michael Balzer * Helkenberger Weg 9 * D-58256 Ennepetal
Fon 02333 / 833 5735 * Handy 0176 / 206 989 26
</pre>
      <br>
      <fieldset class="mimeAttachmentHeader"></fieldset>
      <pre class="moz-quote-pre" wrap="">_______________________________________________
OvmsDev mailing list
<a class="moz-txt-link-abbreviated" href="mailto:OvmsDev@lists.openvehicles.com">OvmsDev@lists.openvehicles.com</a>
<a class="moz-txt-link-freetext" href="http://lists.openvehicles.com/mailman/listinfo/ovmsdev">http://lists.openvehicles.com/mailman/listinfo/ovmsdev</a>
</pre>
    </blockquote>
    <br>
    <pre class="moz-signature" cols="160">-- 
Michael Balzer * Helkenberger Weg 9 * D-58256 Ennepetal
Fon 02333 / 833 5735 * Handy 0176 / 206 989 26
</pre>
  </body>
</html>