0005207: Improve performance of "My View"

ID	Project	Category	View Status	Date Submitted	Last Update
0005207	mantisbt	performance	public	2005-02-03 10:33	2005-04-25 07:47

Reporter	polzin	Assigned To	thraxisp
Priority	normal	Severity	minor	Reproducibility	N/A
Status	closed	Resolution	fixed
Product Version	git trunk
Fixed in Version	1.0.0a2

Summary	0005207: Improve performance of "My View"
Description	I noticed that a relevant portion of time for "My view" is spend on SELECT last_modified FROM mantis_bugnote_table WHERE bug_id='2103' ORDER BY last_modified DESC and an additional COUNT SQL-Statement. But this information is not used on My view, isn´t it? Thus, commenting out grab the bugnote count `$bugnote_count = bug_get_bugnote_count( $v_id );` [...] if ( $bugnote_count > 0 ) { $v_bugnote_updated = bug_get_newest_bugnote_timestamp( $v_id ); } improved the perfomance noticeably.
Additional Information	I also noticed that the bug_get_bugnote_status calls on view_all_inc.php contribute significantly to the sql-time. I tried adding an index on bug_id and the sql-times for a view_all_inc dropped from 2 sec to 0.2 sec (50 issues per page, ca. 6000 bugnotes in the database). (This was depended on the mysql-version. Older versions did not use the index...) Is this worth, adding an INDEX?
Tags	No tags attached.

vboctor 2005-04-20 09:56 manager ~0009847	I have committed to CVS an upgrade step which adds an index on the bug_id field in the bugnotes table.

polzin 2005-04-21 06:46 reporter ~0009849	Still the question remains: Why is the bugnote count computed at all, if the value is not used!? Or did I miss something?

thraxisp 2005-04-21 11:02 reporter ~0009852	This line (or derivative thereof) is commented out in the latest CVS HEAD.

polzin 2005-04-21 11:56 reporter ~0009854	Sorry, this was my mistake. I had a sticky tag in my cvs. :-) Thanks for patching. From my point of view this issue is resolved now.

grab the bugnote count