agent report checkpoint is 0 to the coordinator

[3AceShowHand]

What did you do?

enable pull-based-sink.

What did you expect to see?

agent report normal checkpoint to the coordinator.

What did you see instead?

[2022/12/03 11:02:15.769 +00:00] [INFO] [replication_manager.go:216] ["schedulerv3: ignore table status no table found"] [namespace=default] [changefeed=pd-leader-switch-sync] [message="{\"table_id\":81,\"state\":4,\"checkpoint\":{\"resolved_ts\":437797607323729952},\"stats\":{\"region_count\":1,\"current_ts\":437797607364624384,\"stage_checkpoints\":{\"puller-egress\":{\"checkpoint_ts\":437797607363313759,\"resolved_ts\":437797607323729952},\"puller-ingress\":{\"checkpoint_ts\":437797607323730050,\"resolved_ts\":437797607323729952},\"sink\":{\"resolved_ts\":437797607323729952},\"sorter-egress\":{\"checkpoint_ts\":437797605069553686,\"resolved_ts\":437797605069553686},\"sorter-ingress\":{\"checkpoint_ts\":437797607363313759,\"resolved_ts\":437797607323729952}},\"barrier_ts\":437797605069553687}}"]

From the log above, there is no checkpoint reported, which means it's 0, this is impossible.

Versions of the cluster

Upstream TiDB cluster version (execute SELECT tidb_version(); in a MySQL client):

(paste TiDB cluster version here)

Upstream TiKV version (execute tikv-server --version):

(paste TiKV version here)

TiCDC version (execute cdc version):

master