TiCDC: BR failed to restore due to the existence of the CDC task.

[King-Dylan]

What did you do?

Run a BR restore job on the TiDB cluster with all TiCDC changefeeds stopped beforehand.

What did you expect to see?

The BR job can run successfully.

What did you see instead?

The BR restore job is failing with the following error messages:
[2023/09/22 22:56:13.506 +00:00] [ERROR] [restore.go:64] ["failed to restore"] [error="failed to check task exists: found CDC changefeed(s): cluster/namespace: __backup__/0/tidb/cdc changefeed(s): [test-task], please stop changefeed(s) before restore"] [errorVerbose="found CDC changefeed(s): cluster/namespace: __backup__/0/tidb/cdc changefeed(s): [test-task], please stop changefeed(s) before restore\ngithub.com/pingcap/tidb/br/pkg/task.checkTaskExists\n\t/mnt/tidb/sql/br/pkg/task/stream.go:1082\ngithub.com/pingcap/tidb/br/pkg/task.RunRestore\n\t/mnt/tidb/sql/br/pkg/task/restore.go:576\nmain.runRestoreCommand\n\t/mnt/tidb/sql/br/cmd/br/restore.go:63\nmain.newDBRestoreCommand.func1\n\t/mnt/tidb/sql/br/cmd/br/restore.go:183\ngithub.com/spf13/cobra.(*Command).execute\n\t/go/pkg/mod/github.com/spf13/[email protected]/command.go:916\ngithub.com/spf13/cobra.(*Command).ExecuteC\n\t/go/pkg/mod/github.com/spf13/[email protected]/command.go:1044\ngithub.com/spf13/cobra.(*Command).Execute\n\t/go/pkg/mod/github.com/spf13/[email protected]/command.go:968\nmain.main\n\t/mnt/tidb/sql/br/cmd/br/main.go:58\nruntime.main\n\t/usr/local/go/src/runtime/proc.go:250\nruntime.goexit\n\t/usr/local/go/src/runtime/asm_amd64.s:1598\nfailed to check task exists"] [stack="main.runRestoreCommand\n\t/mnt/tidb/sql/br/cmd/br/restore.go:64\nmain.newDBRestoreCommand.func1\n\t/mnt/tidb/sql/br/cmd/br/restore.go:183\ngithub.com/spf13/cobra.(*Command).execute\n\t/go/pkg/mod/github.com/spf13/[email protected]/command.go:916\ngithub.com/spf13/cobra.(*Command).ExecuteC\n\t/go/pkg/mod/github.com/spf13/[email protected]/command.go:1044\ngithub.com/spf13/cobra.(*Command).Execute\n\t/go/pkg/mod/github.com/spf13/[email protected]/command.go:968\nmain.main\n\t/mnt/tidb/sql/br/cmd/br/main.go:58\nruntime.main\n\t/usr/l[ocal/go/src/runtime/proc.go:250"]](http://ngithub.com/spf13/cobra.(*Command)

Versions of the cluster

Dataflow Engine version (run tiflow version):

v7.1.0

[King-Dylan]

/assign sdojjy

[jackysp]

/found gs

[sdojjy]

From the Ticdc view, it's by design, BR should not check the backup data.