Loading ...
Sorry, an error occurred while loading the content.

Re: Heavyweather Problems

Expand Messages
  • jjantti2
    ... I did pick up the statement and that is exactly where I started to think the problem a bit deeper. My best guess based on the statement about the Weather
    Message 1 of 30 , Jan 20, 2006
    • 0 Attachment
      --- In wuhu_software_group@yahoogroups.com, "Hall Blankenship" <

      > I don't know if anyone picked up on a statement I made in my earlier
      > post that the program Weather Display (trial ver 10.31K) also
      > displays similiar blue screen reboots but much less often. It just
      > rebooted about 20 minutes ago but had been active for almost 2 days.
      > As another note, I am also sending the data out in the APRS format
      > via internet connection. I don't think that has had any effect on
      > the ultimate results because the reboots were occurring regularly
      > prior to beginning the ARES posting process. I have to say that
      > Weather Display has been stable enough to often get close to 24
      > hours run time before rebooting where Heavy Weather Beta 2.0 would
      > usually only run 4 or 5 hours.

      I did pick up the statement and that is exactly where I started to
      think the problem a bit deeper.

      My best guess based on the statement about the Weather Display
      software is that WD is fetching data less intesively. This results in
      added stability. My HeavyWeather software is set to fetch data to
      history.dat file every minute. I have not tested in any other fetch
      time yet, since the server is currently in production stage. When the
      data fetch time is set to 1 minute, HeavyWeather communicates with the
      weather station approximately every two seconds. This is based from
      the status bar on HeavyWeather; After it reaches "Ready." state, after
      two seconds it starts communications again. The whole fetch process
      takes three seconds, in which the processor usage jumps to 100% for
      about half a second. Both the CPU and it's Kernel-level CPU usage goes
      up to 100% indicating some sort of hardware call, that takes all the
      attention from the CPU.

      Eventually this can lead to system instability. My server experienced
      a crash recently and I didn't notice it for one day, but once I got it
      back up and running, I found out that all the data between the crash
      and logging in of the computer was lost. This is strange since my 3600
      should have had all the data available for burst send when
      Heavyweather came back online. This can only mean that the last system
      call sent from the computer was nothing short of "send the data" and
      when the computer crashed, the weather station kept sending data to
      the serial interface, but no one was listening and not saving them to
      it's memory. This is an error condition I have never been able to
      reproduce and ever since the crash, the server has been up and running.

      One strange thing in all of this is that my weather station is plugged
      to the server's COM port without any adapters.

      > I also have to say that I have run GPS data continuously with no
      > problems via the same USB to 232 adapter I am now trying to use with
      > Weather Display and Heavy Weather so the adapter works at least on
      > some levels.

      This is because the GPS receiver is using standard methods in
      communicating with the adapter device. The serial communication with
      the station is whole lot different. I have no way of determining the
      exact cause, but I suspect the problem to be with both the
      communication method and drivers.

      My 3600 is a whole lot different compared to 2300-series, but at least
      narrowing the cause of the crash can help us all.

      Best Regards,
      Juha
      WU : IUUSIMAA2
      CWOP: CW4473
    Your message has been successfully submitted and would be delivered to recipients shortly.