Good thinking! And I already tried and it didn’t work… but what is it they say about great minds?! 
Here is a brand-new piston, and the log from its last two executions (both of which failed as described above). I’ve added emphasis in the logs where teh questionable behavior is happening.
I’m running on the na04-useast2 shard.
And last but certainly not least, thank you for trying to help! This failure mode has completely ruined home automation for me and I’d like to get it back. Bonus points if we can figure out how/why it’s doing this… I can’t tell anything from the logs.
11/21/2018, 7:30:35 AM +493ms
+0ms ╔Received event [Home].time/recovery = 1542803435492 with a delay of 0ms
+599ms ║RunTime Analysis CS > 250ms > PS > 307ms > PE > 43ms > CE
+602ms ║Runtime (42859 bytes) successfully initialized in 307ms (v0.3.108.20180906) (601ms)
+603ms ║╔Execution stage started
+605ms ║╚Execution stage complete. (2ms)
+607ms ║Setting up scheduled job for Thu, Nov 22 2018 @ 7:30:00 AM EST (in 86363.901s)
+617ms ╚Event processed successfully (617ms)
11/21/2018, 7:30:35 AM +207ms
+0ms ╔Received event [Home].execute = recovery with a delay of 129ms
+148ms ║RunTime Analysis CS > 21ms > PS > 65ms > PE > 63ms > CE
+151ms ║Runtime (42844 bytes) successfully initialized in 65ms (v0.3.108.20180906) (149ms)
+152ms ║╔Execution stage started
+162ms ║║Comparison (time) 27035366 happens_daily_at (time) 27000000 = false (1ms)
+164ms ║║Condition #31 evaluated false (5ms)
+166ms ║║Cancelling statement #31’s schedules…
+170ms ║║Requesting time schedule wake up at Thu, Nov 22 2018 @ 7:30:00 AM EST
+174ms ║║Condition group #30 evaluated false (state did not change) (15ms)
+177ms ║╚Execution stage complete. (25ms)
+179ms ║Setting up scheduled job for Thu, Nov 22 2018 @ 7:30:00 AM EST (in 86364.615s)
+187ms ╚Event processed successfully (187ms)
As I mentioned above, the logs for events on SmartThings and webCoRE don’t align. Not a single one of them. Clearly above the time condition failed, therefore no commands were sent to lights. However, one light was left on this morning. Three lights met the condition of the piston at 7:30am (level below 99% and switch off)… yet only one was turned on. And it was never turned off; at least not by the piston, it was manually switched off by my (way out of patience) wife at 7:45am.