[CDC:ErrCheckClusterVersionFromPD]failed to request PD%!(EXTRA string=response status: 503 Service Unavailable)

[Tammyxia]

What did you do?

• 4 pd, 2 tidb, 7 tikv, 2 cdc
• test infra case: 360 pd_leader_switch_changefeed_admin_sync
• when pd leader switch one by one,meanwhile changefeed pause/resume

What did you expect to see?

No response

What did you see instead?

• cli chnangefeed resume failed:

2022-01-14T19:16:51.039Z INFO host/ticdc.go:220 Execute command on ticdc {"command": "/cdc cli changefeed resume "--pd=http://upstream-pd.cdc-qihoo-360-testbed-tps-600887-1-433:2379\" "--changefeed-id=qihoo360-mysql-task"", "timeout": "10s", "ticdc peer": 0}
[2022/01/14 19:16:51.050 +00:00] [INFO] [pd.go:32] ["pd leader"] [leader-name=upstream-pd-2]
[2022/01/14 19:16:51.553 +00:00] [INFO] [pd.go:46] ["transfer pd leader success"] [leader-name=upstream-pd-1]
[2022/01/14 19:16:52.215 +00:00] [ERROR] [ticdc.go:61] ["run changefeed operation failed"] [output="{stdout: Usage:\n cdc cli changefeed resume [flags]\n\nFlags:\n -c, --changefeed-id string Replication task (changefeed) ID\n -h, --help help for resume\n --no-confirm Don't ask user whether to ignore ineligible table\n\nGlobal Flags:\n --ca string CA certificate path for TLS connection\n --cert string Certificate path for TLS connection\n -i, --interact Run cdc cli with readline\n --key string Private key path for TLS connection\n --log-level string log level (etc: debug|info|warn|error) (default "warn")\n --pd string PD address, use ',' to separate multiple PDs (default "http://127.0.0.1:2379"), stderr: Error: [CDC:ErrCheckClusterVersionFromPD]failed to request PD%!(EXTRA string=response status: 503 Service Unavailable)\n[CDC:ErrCheckClusterVersionFromPD]failed to request PD%!(EXTRA string=response status: 503 Service Unavailable), ExitCode: 1"] [stack="github.com/pingcap/test-infra/caselib/pkg/host.(*TiCDCHost).ChangeFeed\n\t/Users/lixia/source-code/test-infra/caselib/pkg/host/ticdc.go:61\ngithub.com/pingcap/test-infra/caselib/pkg/steps.(*changeFeedTask).Execute\n\t/Users/lixia/source-code/test-infra/caselib/pkg/steps/changefeed.go:49\ngithub.com/pingcap/test-infra/caselib/pkg/steps.(*loopTask).Execute\n\t/Users/lixia/source-code/test-infra/caselib/pkg/steps/task.go:95\ngithub.com/pingcap/test-infra/caselib/pkg/steps.(*Parallel).Execute.func1\n\t/Users/lixia/source-code/test-infra/caselib/pkg/steps/step.go:69\ngolang.org/x/sync/errgroup.(*Group).Go.func1\n\t/Users/lixia/go/pkg/mod/golang.org/x/sync@v0.0.0-20210220032951-036812b2e83c/errgroup/errgroup.go:57"]

Versions of the cluster

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

5.4.0

TiCDC version (execute cdc version):

5.4.0