Can't restore system tables separately without restoring user tables

[fubinzh]

Please answer these questions before submitting your issue. Thanks!

1. What did you do?
   If possible, provide a recipe for reproducing the error.

a. Run "br backup full" to create full backup, including system tables and user tables.
b. Run br restore full --filter "mysql*.*" to restore system tables only, (do not restore user tables)

2. What did you expect to see?
   System tables mysql.* should be restored correctly.

3. What did you see instead?

br failed to restore system tables. "Nothing to restore" seen in log.

4. What version of BR and TiDB/TiKV/PD are you using?

br 5.1.0 dev build.
[root@centos76_vm dev]# ./br -V
Release Version: v5.0.0-master-dirty
Git Commit Hash: b9c4ea6
Git Branch: backupmetav2
Go Version: go1.16.5
UTC Build Time: 2021-06-09 12:33:55
Race Enabled: false

5. Operation logs
   • Please upload br.log for BR if possible
   • Please upload tidb-lightning.log for TiDB-Lightning if possible
   • Please upload tikv-importer.log from TiKV-Importer if possible
   • Other interesting logs

[root@centos76_vm dev]# ./br restore full --pd "172.16.6.183:2379" --filter "mysql*.*" --storage "local:///minio/fubin/tidb510_br510" --log-file full_restore_2_system.log
[2021/06/10 11:38:52.086 +08:00] [INFO] [data_slow_query.go:144] ["Telemetry slow query stats initialized"] [currentSQBInfo={0.512:0,65.536:0,524.288:0,2097.152:0,134217.734:0,33554.434:0,1048.576:0,0.008:0,0.016:0,2.048:0,4.096:0,8.192:0,4194.304:0,0.002:0,16777.217:0,67108.867:0,0.004:0,0.064:0,16.384:0,32.768:0,131.072:0,262.144:0,+Inf:0,0.001:0,0.128:0,0.256:0,0.032:0,1.024:0,8388.608:0] [lastSQBInfo={1048.576:0,0.001:0,0.032:0,0.128:0,0.512:0,4.096:0,524.288:0,2097.152:0,33554.434:0,8.192:0,0.256:0,65.536:0,16777.217:0,0.064:0,2.048:0,+Inf:0,0.002:0,0.004:0,1.024:0,4194.304:0,134217.734:0,0.008:0,131.072:0,262.144:0,8388.608:0,0.016:0,16.384:0,32.768:0,67108.867:0]
Detail BR log in full_restore_2_system.log
[2021/06/10 11:38:52.229 +08:00] [INFO] [collector.go:66] ["Full restore success summary"] [total-ranges=0] [ranges-succeed=0] [ranges-failed=0] [total-take=119.928823ms] [Result="Nothing to restore"]

==full_restore_2_system.log==
[2021/06/10 11:38:52.204 +08:00] [INFO] [client.go:195] ["load backupmeta"] [databases=2] [jobs=0]
[2021/06/10 11:38:52.212 +08:00] [INFO] [import.go:242] ["multi ingest support"] [support=true]
[2021/06/10 11:38:52.218 +08:00] [WARN] [restore.go:530] ["set max-index-length to max(3072*4) to skip check index length in DDL"]
[2021/06/10 11:38:52.218 +08:00] [INFO] [restore.go:315] ["nothing to restore, all databases and tables are filtered out"]

6. Configuration of the cluster and the task

   • tidb-lightning.toml for TiDB-Lightning if possible
   • tikv-importer.toml for TiKV-Importer if possible
   • topology.yml if deployed by TiUP

7. Screenshot/exported-PDF of Grafana dashboard or metrics' graph in Prometheus if possible

[fubinzh]

Tried to restore single system table (mysql.user), also failed with below errors:

=====
[root@centos76_vm dev]# ./br restore table --pd "172.16.6.183:2379" --db mysql --table user --storage "local:///minio/fubin/tidb510_br510" --log-file system_full_backup2.restore_single_table.log
...
Detail BR log in system_full_backup2.restore_single_table.log
[2021/06/10 16:12:30.626 +08:00] [INFO] [collector.go:66] ["Table restore failed summary"] [total-ranges=0] [ranges-succeed=0] [ranges-failed=0]
Error: [database: mysql] has not been backup, please ensure you has input a correct database name: [BR:Common:ErrUndefinedDbOrTable]undefined restore databases or tables