Missed charging slot

WindyMiller

Last night one charging slot was missed. I had set 2 charging slots at a minimum of 30%. At 2.30am the charge was above 30% so the slot was correctly held. At 4.30am the charge was below 30% and should have charged back to 30% but the slot wasn't programmed to charge or hold. In fact there is nothing in the log at all which is unusual as there is usually an entry to say what happened with the slot. It was as if the charge slot wasn't selected even though its outline was blue. Please can you see if I have mis-programmed something (I just ran out of charge during the morning peak). Thanks.

elaborate-routine

admin

WindyMiller We had a Hold from 3:00-6:30.

WindyMiller

These were the 2 active schedules

WW had chosen the 2 cheapest slots as 2.30am and 4.30am. The 2.30am was above the charge threshold but the 4.30am should have triggered as it was below the threshold. Or does it only apply to the first active slot in a window?

admin

WindyMiller Ah thanks, I see, we'll check it in a bit more detail.

WindyMiller

admin Thanks. I have a similar setup for tonight. The 2 chosen cheap slots are 00:30am and 04:30am. I expect the 00:30am not to trigger but the 04:30am probably should.

WindyMiller

Neither slot triggered last night as the charge was above the 30% level but it is interesting in that both slots produced a hold record in the log.

I do not understand why the 2nd slot produced no effect (action or entry in the log) the previous night as the only difference was that the 2nd slot was below the 30% threshold?

admin

WindyMiller Yeah, seems like some weird anomaly. We may need to add additional logging still.

WindyMiller

I had another missed slot last night. The 4.30-6am slot didn't charge. No entries in either log (WW or GE) at 4.30am other than the clear-down of the previous export slot at 4.28am. This time it was a charge to 100% independent of current charge levels.

admin

WindyMiller Thanks, we'll take a look.

OK, found the culprit, looks like database command timeouts. We've optimised several queries, and especially as we're checking IOG schedules fairly often, and have a lot of influx from IOG users, this seems to have been contributing to the problem.

Let us know if you see any other anomalies, thanks!