The alert sending interval is inconsistent with the document description

What did you do?
I create a receiver to receive the alerts from alertmanager.
I set these variables like that:

1. global group_wait: 15m
2. global group_interval: 15m
3. global repeat_interval: 20m
4. receiver repeat_interval: 20m

What did you expect to see?
I except receive the same alert in 30 minutes.

What did you see instead? Under which circumstances?
I always receive the same alert in 30 minutes instead of 20 minutes.

Environment

• System information:

Linux 5.10.25-linuxkit x86_64

• Alertmanager version:

  alertmanager, version 0.19.0 (branch: HEAD, revision: 7aa5d19fea3f58e3d27dbdeb0f2883037168914a)
  build user: root@587d0268f963
  build date: 20190903-15:01:40
  go version: go1.12.8

• Prometheus version:

  prometheus, version 2.12.0 (branch: HEAD, revision: 43acd0e2e93f9f70c49b2267efa0124f1e759e86)
  build user: root@7a9dbdbe0cc7
  build date: 20190818-13:53:16
  go version: go1.12.8

Start by reproducing the alert timing with the listed global and receiver intervals on Alertmanager 0.19.0, then compare the observed behavior with the documented repeat_interval semantics. Done means the interval discrepancy is explained and either the behavior or documentation is corrected, with regression coverage if the repository provides a relevant test location.