recover table by job ID would cause panic

[wjhuang2016]

Bug Report

Please answer these questions before submitting your issue. Thanks!

1. Minimal reproduce step (Required)

func (e *DDLExec) executeRecoverTable(s *ast.RecoverTableStmt) error {
	dom := domain.GetDomain(e.Ctx())
	var job *model.Job
	var err error
	var tblInfo *model.TableInfo
	// Let check table first. Related isssue #46296.
	if s.Table != nil {
		job, tblInfo, err = e.getRecoverTableByTableName(s.Table)
	} else {
		job, tblInfo, err = e.getRecoverTableByJobID(s, dom)
	}
	if err != nil {
		return err
	}
	// Check the table ID was not exists.
	tbl, ok := dom.InfoSchema().TableByID(tblInfo.ID)
	if ok {
		return infoschema.ErrTableExists.GenWithStack("Table '%-.192s' already been recover to '%-.192s', can't be recover repeatedly", s.Table.Name.O, tbl.Meta().Name.O)
	}

ErrTableExists uses s.Table, but it's nil in recover table by job ID.

2. What did you expect to see? (Required)

N/A

3. What did you see instead (Required)

N/A

4. What is your TiDB version? (Required)

master

[crazycs520]

Nice catch, I will fix it.

This will only occur in some abnormal situations, and the panic will only affect the current query and will not cause tidb-server to panic.

And the minimal reproduce step is:

test> create table t (a int);
Query OK, 0 rows affected
Time: 0.085s
test> drop table t;
Query OK, 0 rows affected
Time: 0.330s
test> admin show ddl jobs 1; 
+--------+---------+------------+---------------+--------------+-----------+----------+-----------+----------------------------+----------------------------+----------------------------+--------+
| JOB_ID | DB_NAME | TABLE_NAME | JOB_TYPE      | SCHEMA_STATE | SCHEMA_ID | TABLE_ID | ROW_COUNT | CREATE_TIME                | START_TIME                 | END_TIME                   | STATE  |
+--------+---------+------------+---------------+--------------+-----------+----------+-----------+----------------------------+----------------------------+----------------------------+--------+
| 106    | test    | t          | drop table    | none         | 2         | 104      | 0         | 2024-10-29 19:48:08.829000 | 2024-10-29 19:48:08.879000 | 2024-10-29 19:48:09.180000 | synced |
+--------+---------+------------+---------------+--------------+-----------+----------+-----------+----------------------------+----------------------------+----------------------------+--------+
1 row in set
Time: 0.009s
test> recover table by job 106;
Query OK, 0 rows affected
Time: 0.224s
test> recover table by job 106;
(1105, 'runtime error: invalid memory address or nil pointer dereference')