Bug #59799 handler statistics for the performance schema engine
Submitted: 28 Jan 2011 18:00 Modified: 10 Feb 2011 15:15
Reporter: Marc Alff Email Updates:
Status: Closed Impact on me:
Category:MySQL Server: Performance Schema Severity:S3 (Non-critical)
Version:5.5.6 OS:Any
Assigned to: Marc Alff CPU Architecture:Any

[28 Jan 2011 18:00] Marc Alff
The performance schema storage engine does not update the thread session
variables, for handler statistics.

How to repeat:
show status like "%handler%";
select * from performance_schema.setup_instruments;
show status like "%handler%";
see that counters did not increase, even when data was returned.

Suggested fix:
call ha_statistic_increment(&SSV::ha_read_next_count) and related
[28 Jan 2011 18:41] Bugs System
A patch for this bug has been committed. After review, it may
be pushed to the relevant source trees for release in the next
version. You can access the patch from:


3569 Marc Alff	2011-01-28
      Bug#59799 handler statistics for the performance schema engine
      Before this fix, the performance schema storage engine did not
      update the handler_* statistics, so that:
        show status like "%handler%";
      returned under evaluated values.
      With this fix, calls to rnd_next(), rnd_pos(), etc
      methods in the performance schema engine now also properly
      update the corresponding handler_* statistics.
      Tested manually.
      No automated scripts provided, as the value reported for
      a session in these counters is hard to predict exactly,
      as it depends too much on the environment.
[28 Jan 2011 18:56] Bugs System
Pushed into mysql-trunk 5.6.2 (revid:marc.alff@oracle.com-20110128185545-knrd930f29jy22b2) (version source revid:marc.alff@oracle.com-20110128184059-k2jlz36la1hkis5a) (merge vers: 5.6.2) (pib:24)
[10 Feb 2011 15:15] Paul Dubois
Noted in 5.6.2 changelog.

Performance Schema did not update status handler status variables, so
SHOW STATUS LIKE '%handler%' produced undercounted values.