cdc owner stuck after injection downstream failure 3 mins and recover

[Lily2025]

What did you do?

1、run workload
storage:"s3://benchmark/sysbench_64_7000w"
subType:"oltp_read_write"
db:"sysbench_64_7000w"
tableNum:64
tableSize:70000000
threads:32
ignoreErrors:"2013,1213,1105,1205,8022,8028,9004,9007,1062"
2、 injection downstream failure 3 mins and recover

What did you expect to see?

changefeed status are normal after fault recover

What did you see instead?

1、the status of changefeed task keep warning for 20 minutes after fault recover

2、the log lost during status of changefeed abnormal

Versions of the cluster

./cdc version
Release Version: v6.5.11
Git Commit Hash: 7ce6da9
Git Branch: HEAD
UTC Build Time: 2024-09-09 04:49:13
Go Version: go version go1.19.13 linux/amd64
Failpoint Build: false
2024-09-11T22:18:58.435+0800

current status of DM cluster (execute query-status <task-name> in dmctl)

No response

[Lily2025]

/remove-area dm
/area ticdc

[Lily2025]

/assign hicqu