0 Replies Latest reply on Jun 21, 2013 8:50 AM by Dan McCabe

    RhoElements.. how exactly are we supposed to use logging?

    Dan McCabe

      After installing from RhoElements 2.2.1.13 WM Runtime.cab, the default config.xml looks like this:

       

        <Logger>

          <LogProtocol   value="FILE"/>

          <LogPort       value="80"/>

          <LogURI        value="file://%INSTALLDIR%\Log.txt"/>

          <LogError      value="1"/>

          <LogWarning    value="1"/>

          <LogInfo       value="1"/>

          <LogUser       value="1"/>

          <LogMemory     value="0"/>

          <LogMemPeriod  value="5000"/>

          <LogMaxSize    value="10000000"/>

        </Logger>

       

      From this config, and without reading the docs, (because who reads the docs on all the config options until they have a problem?) I would expect that I'd get a log file called Log.txt that would contain the most recent 10000000 bytes.

       

      Then, if a problem cropped up, I'd have something useful to look at at the bottom of the log file.

       

      It seems that I'm completely wrong on several accounts:

      First, the log file does not go to Log.txt, but rather rholog.txt.

       

      Second, the docs say that LogMaxSize is in kb, which explains why all my deployed devices are running out of storage space and popping up messages to the users.  (GPS position reports really fill up the log)   This does not appear to be a sane default given the storage typically available on Moto WinMo devices where this cab file would be used.

       

      Third, if the docs are correct, then when the max size is reached logging will stop.

       

      If that is really true, then it seems that there is no point in leaving the logging on at all on deployed devices.

      By the time there is a problem, the logging may have been stopped for weeks or months because the size limit was reached.

       

      Is that actually the case?

       

      The only way I can see around this is to delete the log file in startupctl and assume that that will happen often enough to work out.

      Of course, since users are always trained to reboot before calling for help, I can't really just delete the log file.. have to move it first and hope that they don't reboot twice...ughhh.

       

      Dan