File: group_commit_force_recovery.test

package info (click to toggle)
mariadb 1%3A10.11.11-0%2Bdeb12u1
  • links: PTS, VCS
  • area: main
  • in suites: bookworm
  • size: 597,064 kB
  • sloc: ansic: 2,386,076; cpp: 1,663,071; asm: 378,311; perl: 62,203; pascal: 38,769; sh: 37,071; java: 33,919; sql: 19,830; yacc: 19,727; xml: 10,509; python: 9,522; ruby: 8,544; cs: 5,855; makefile: 5,793; ada: 1,700; lex: 1,207; javascript: 1,039; objc: 80; tcl: 73; awk: 46; php: 22
file content (21 lines) | stat: -rw-r--r-- 731 bytes parent folder | download | duplicates (5) 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
 
# MDEV-24302 RESET MASTER hangs as Innodb does not report on binlog checkpoint
# Testing binlog checkpoint notification works under stringent condition
# set by innodb_force_recovery = 2.

--source include/have_innodb.inc
--source include/have_binlog_format_mixed.inc

# Binlog checkpoint notification consumers such as RESET MASTER
# receive one when lsn_0 at the time of the request is finally gets flushed
#   flush_lsn >= lsn_0
# The bug situation was that when lsn_0 reflects a write of an internal innodb trx
# and RESET MASTER was not followed by any more user transaction
# it would hang.

CREATE TABLE t1(a int) ENGINE=InnoDB;
INSERT INTO t1 SET a=1;
RESET MASTER;

# final cleanup
DROP TABLE t1;
--echo End of the tests.