6A 67 34 Fault Events analisis

First findings:

• some of the last successful issued temp basal pulse rates are unending if calculated to seconds like, 81,8181818181 and 3,2233333
• It almost seems once a normal basal can run, a 6A will occur after a while.
• 6A and 34 errors are often in segment 2 and after 1 or 2 hours.
• It seems if there is a temp basal starting < 100 seconds around a segment change, a fault can occur within 2 hours of the next segment.
• If a long temp basal is running with continuously reissuing the same value, that can add to the hours of when a failure occurs in a segment.
• There are 2 occasions where the fault happens at 10AM.
• I tried to look at a race condition by counting the number of pulses staring with half the period. Many cases have around 2-12 seconds where it could issue a next pulse in the last succesful issued temp basal. Sometimes this value only is true if the pulse starts at the end or beginning of the period, not halfway.
• Some of the earlier logs only contained 24 hours of dosages.

Xcode Issue Reports:

• 6A-Fault-event-1h-segment-2---temp-basal-overlaps-segment

• 6a-fault-event-1-h-segment-1-less-then-100-seconds-temp-basal-start

• 6A Fault event at 2h segment 2

• 6A Fault event at 3 hour segment 1

• 67-Fault-event-after-2h-in-segment-2-temp-basal-started-too-soon

• 34-Fault-event-at-2h-segment-2-temp-basal

• 6A-Fault-event-before-PPPP-and-XXXXXXXX-ZZZZZZZZ-fixes