[dr-autosync] changfeed status change to error during patching pd

[mayjiang0203]

What did you do?

[2023/06/26 13:19:30.273 +08:00] [INFO] [tiup.go:422] ["run cmd"] [cmd="tiup cluster patch tidbcluster /tmp/2023-06-26T13:19:25+08:00.tiup.tar.gz -R pd -y"]

[2023/06/26 13:19:49.682 +08:00] [INFO] [ticdc.go:145] ["will run"] [cmd="tiup cdc:v6.5.3 cli changefeed list --server=http://ticdc1-peer.e2e-dr-auto-sync-5r-pref-tps-1809605-1-603:8300"]
2023-06-26T13:19:49.685+0800 INFO k8s/client.go:132 it should be noted that a long-running command will not be interrupted even the use case has ended. For more information, please refer to https://github.com/pingcap/test-infra/discussions/129
[
{
"id": "dr-autosyn-changefeed",
"namespace": "default",
"summary": {
"state": "error",
"tso": 442435313169596438,
"checkpoint": "2023-06-26 13:19:39.364",
"error": {
"time": "2023-06-26T13:19:41.10221041+08:00",
"addr": "ticdc1-peer:8300",
"code": "CDC:ErrReachMaxTry",
"message": "[CDC:ErrReachMaxTry]reach maximum try: 10, error: rpc error: code = Unknown desc = [PD:tso:ErrGenerateTimestamp]generate timestamp failed, requested pd is not leader of cluster: rpc error: code = Unknown desc = [PD:tso:ErrGenerateTimestamp]generate timestamp failed, requested pd is not leader of cluster"
}
}
}
]

What did you expect to see?

Changefeed should not report this error.

What did you see instead?

changefeed failed and can't recover automatically.

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):
[root@ticdc1-0 bin]# ./cdc version
Release Version: v6.5.3
Git Commit Hash: 1e2f277
Git Branch: heads/refs/tags/v6.5.3
UTC Build Time: 2023-06-05 10:19:02
Go Version: go version go1.19.9 linux/amd64
Failpoint Build: false

(paste TiCDC version here)