Has anyone experienced an issue where when the slow log is enabled with the following options:
root@(none)> show global variables like "%slow_query%";
+-----------------------------------+-------------------------------+
| Variable_name | Value |
+-----------------------------------+-------------------------------+
| slow_query_log | ON |
| slow_query_log_always_write_time | 10.000000 |
| slow_query_log_file | /var/log/mysql/mysql-slow.log |
| slow_query_log_use_global_control | |
+-----------------------------------+-------------------------------+
4 rows in set (0.00 sec)
I get a weird issue where the slow log file is being spammed with the same thing over and over again, without a single slow log query:
Time Id Command Argument
/usr/sbin/mysqld, Version: 8.0.25-15 (Percona Server (GPL), Release '15', Revision 'a558ec2'). started with:
Tcp port: 3306 Unix socket: /var/run/mysqld/mysqld.sock
These three lines are being spammed constantly, about 100 kilobytes per second. Not only that, because of this the replication was lagging, the seconds behind master was constantly creeping up. I was noticing some odd usage of the OS disks (the MySQL datadir is on a different array) where the /var/log/mysql directory resides, I suppose it was flushing to disk each and every one of these useless messages. Once I disabled slow_query_log the SBM started dropping very fast and has no problems keeping up.
Am I hitting a bug with Percona MySQL, I can't find any reference anywhere of this?