Electron data usage problem redux

Troubleshooting
24

You’re too quick @RWB! I had other things to attend to, but I’m back and just re-tested my original code with and without the extra 5 second delay. (I’ll get to the new firmware soon).

Results with my original code and 0.6.0-rc.2:

  1. :scream: :hankey: because I can’t find a facepalm or head-bangingin-into-wall gif. I’ve been missing part of the data when looking at getDataUsage(). I had been resetting the counter early in code and therefore missed the usage associate with network connection. Duh. When looking at all usage after board reset, it adds up! Thanks to Hologram MVNO session usage stats, I can confirm that getDataUsage() matched the carrier usage. It’s about the same as Particle. Every publish was burning 953 bytes of data. Good news overall, but I’m still sorry I didn’t have that straight.

  2. In previous tests, I did in fact use much less data. It’s still not clear what changed between then and now, but I’m done trying to solve that puzzle.

  3. If I add the 5 second delay to my original code, the first connection uses 941 bytes and subsequent connections use only 288 bytes. My data string is about 85 bytes of that. 288 bytes @ 10 minute interval is about 1.24 MB per month, which is on par with my previous results (before this whole mess). That’s still more than @RWB claimed here Best Methods for Conserving Electron Current and Data Usage, but still much better.

Kudos to @RWB and @ScruffR for helping me figure this out! :muscle:

2 Likes