Reschedule Times Table for 5s plus N^4

By default, DelayedJob reschedules erroring jobs in 5 seconds + N ** 4, where N is the number of failed attempts.

These figures assume: 1) the job always runs on schedule at run_at and performs in 0 seconds. 2) your app has the default Delayed::Worker.max_attempts = 25 which is why the table says the job is not rescheduled for N=25 (check your app's Delayed::Worker.max_attempts as it may be different).

Number of failed attempts N	Delay until next (N+1) attempt 5s + N^4	Cumulative delay after 1st attempt (~approximation)
1	6s	6s
2	21s	27s
3	86s	113s (~2 mins)
4	261s	374s (~6 mins)
5	630s	1,004s (~18 mins)
6	1,301s	2,305s (~38 mins)
7	2,406s	4,711s (~1 hour)
8	4,101s	8,812s (~2 hours)
9	6,566s	15,378s (~4 hours)
10	10,005s	25,383s (~7 hours)
11	14,646s	40,029s (~11 hours)
12	20,741s	60,770s (~17 hours)
13	28,566s	89,336s (~1 day)
14	38,421s	127,757s (~1 day)
15	50,630s	178,387s (~2 days)
16	65,541s	243,928s (~3 days)
17	83,526s	327,454s (~4 days)
18	104,981s	432,435s (~5 days)
19	130,326s	562,761s (~7 days)
20	160,005s	722,766s (~8 days)
21	194,486s	917,252s (~11 days)
22	234,261s	1,151,513s (~13 days)
23	279,846s	1,431,359s (~17 days)
24	331,781s	1,763,140s (~20 days)
25	Job failed, is not rescheduled	-