Adjust perfschema autosizing for Percona Server mutexes

Bug #1264952 reported by Laurynas Biveinis
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
Percona Server moved to https://jira.percona.com/projects/PS
Fix Released
Low
Laurynas Biveinis
5.1
Invalid
Undecided
Unassigned
5.5
Invalid
Undecided
Unassigned
5.6
Fix Released
Low
Laurynas Biveinis

Bug Description

Currently sizing_growth testcase is recorded as

status
FAILED: heuristics need to be updated

which is caused by Percona Server hitting four instead of three mutexes in MySQL for opening a connection. The three MySQL mutexes being wait/synch/mutex/mysys/my_thread_var::mutex, wait/synch/mutex/mysys/THR_LOCK::mutex, wait/synch/mutex/sql/THD::LOCK_thd_data, and the fourth Percona Server mutex being wait/synch/mutex/sql/THD::LOCK_temporary_tables.

The comments in the testcase suggest trivially updating PFS autosizing heuristics in pfs_autosize.cc to account for four mutexes, so that the autosized performance_schema_max_mutex_instances presumably would do a bit better job.

I guess this has a negligible impact, but it's an easy fix and fixing it is probably a good idea as the current sizing_growth.result is recorded very counter-intuitively.

Related branches

tags: added: merge-regression performance
tags: added: low-hanging-fruit
Revision history for this message
Shahriyar Rzayev (rzayev-sehriyar) wrote :

Percona now uses JIRA for bug reports so this bug report is migrated to: https://jira.percona.com/browse/PS-2056

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.