InnoDB: page_cleaner: 1000ms intended loop took 23911ms.

MySQL & MariaDB Other MySQL® Questions
1

Seeing this message “InnoDB: page_cleaner: 1000ms intended loop took 23911ms. The settings might not be optimal. (flushed=200 and evicted=0, during the time.)” in the MySQL logs several times (v5.7.18).
Using SSD.
RAM: 64GB

Tried changing the following configuration in my.cnf but did not resolve this issue:
innodb_buffer_pool_size=50GB
innodb_lru_scan_depth = 512
innodb_write_io_threads=16
innodb_flush_neighbor=0
innodb_log_file_size=100GB
innodb_log_files_in_group=2
innodb_change_buffer_max_size=50
innodb_log_buffer_size=4GB
innodb_max_dirty_pages_pct_lwm=30
innodb_thread_concurrency=0
innodb_write_io_threads=64
innodb_concurrency_tickets=500
innodb_thread_concurrency=16
innodb_io_capacity=1000
innodb_io_capacity_max=2000

Please provide some pointers on how to resolve this issue. Thanks in advance.

2

Hi - did you see this page in the documentation? In case it might give you the pointer you need:

[url]XtraDB Performance Improvements for I/O-Bound Highly-Concurrent Workloads — Percona Server 5.7 Documentation

At the bottom of the page there’s also a link to a page on page cleaner thread tuning - althought it’s a 5.6 document maybe it would be useful?

3

Hi

Have you solved this problem? I also encountered the same problem.
Can you give me some advice?

Thanks.

4

Are you getting exactly the same messages as the original poster, keith?
If not, please do open up a new thread and we can take it from there. Thanks!