Errors? Bad Flash? Maybe Not...
Posted: Thu Jul 24, 2014 3:42 am
Based on previous posts I was convinced that new operational errors were caused by failure of my expensive SLC flash after 270 rounds, and I needed to replace it. Not so sure about that now. I share the following experience in case you have a perplexing new operational issue with MH that you cannot resolve.
MH 4.9y spontaneously began to have problems reading the VP2 USB logger. It would intermittently catch a few readings (Sensors page), but data was never or rarely logged (Inspect Data). Bad data logger? Weatherlink read it just fine. I reset the VP2 console, and repeatedly saved the correct Device location in the Weather Station setup, but this had no affect. The Meteohub log indicated that the system kept cycling through "child process received termination signal " followed by restart of the data logger initialization.
I re-imaged the CF followed by updates, and reloaded a recent copy of the MH configuration (prior to onset of the problems). Same problem.
I then imaged a new CF, followed by updates and the archived MH config. Same problem. Was it the data logger, the MH config file, or the ALIX.1D?
I went back to the old CF (with the original config), and Reset the MH configuration. Entered all of my station's parameters from scratch. It has functioned perfectly for 24 hours since then.
So consider that if there is an operational problem, clear out all of the settings and start again from scratch without loading a backup .conf. Was it originally a corrupt CF memory location that subsequently corrupted the MH config file? Cannot say for sure. But I'll keep the original CF going as long as I can.
MH 4.9y spontaneously began to have problems reading the VP2 USB logger. It would intermittently catch a few readings (Sensors page), but data was never or rarely logged (Inspect Data). Bad data logger? Weatherlink read it just fine. I reset the VP2 console, and repeatedly saved the correct Device location in the Weather Station setup, but this had no affect. The Meteohub log indicated that the system kept cycling through "child process received termination signal " followed by restart of the data logger initialization.
I re-imaged the CF followed by updates, and reloaded a recent copy of the MH configuration (prior to onset of the problems). Same problem.
I then imaged a new CF, followed by updates and the archived MH config. Same problem. Was it the data logger, the MH config file, or the ALIX.1D?
I went back to the old CF (with the original config), and Reset the MH configuration. Entered all of my station's parameters from scratch. It has functioned perfectly for 24 hours since then.
So consider that if there is an operational problem, clear out all of the settings and start again from scratch without loading a backup .conf. Was it originally a corrupt CF memory location that subsequently corrupted the MH config file? Cannot say for sure. But I'll keep the original CF going as long as I can.