[Rt-devel] Performance Problem

[Tina Schade]

Hello,

I have an big performace problem. I work with RT 3.0.10, SearchBuilder 0.99 
and Postgres 7.3.4. 
Every time I logged in RT needs 1 minute minimum to build the page after 
login. The part to show my own 10 tickets is ok, but for the part to show the 
requested tickets take the longest time. I print out the querys and I notice 
that the querys difference is the "=" and the "ILIKE". Here are the queries:

SELECT main.* FROM Tickets main WHERE ((main.EffectiveId = main.id)) AND 
((main.Status != 'deleted')) AND ((main.Type = 'ticket')) AND ( ( 
(main.Status = 'new')OR(main.Status = 'open') ) AND ( (main.Owner = '32') ) ) 
ORDER BY main.Priority DESC LIMIT 10

and:
SELECT DISTINCT main.* FROM Tickets main , Groups Groups_1, CachedGroupMembers 
CachedGroupMembers_2, Users Users_3 WHERE ((CachedGroupMembers_2.MemberId 
ILIKE Users_3.id)) AND ((Groups_1.id ILIKE CachedGroupMembers_2.GroupId)) AND 
((main.EffectiveId = main.id)) AND ((main.Status != 'deleted')) AND 
((main.Type = 'ticket')) AND ((main.id ILIKE Groups_1.Instance)) AND ( ( 
(main.Status = 'new')OR(main.Status = 'open') ) AND ( ( (Users_3.EmailAddress 
ILIKE 'T.Schade at chili-radiology.com')AND(Groups_1.Domain = 
'RT::Ticket-Role')AND(Groups_1.Type = 'Requestor') ) ) ) ORDER BY 
main.Priority DESC LIMIT 10

I don´t understand, why you compare ids with "ILIKE" and not with "=" 
(CachedGroupMembers_2.MemberId ILIKE Users_3.id, Groups_1.id ILIKE 
CachedGroupMembers_2.GroupId). 
Does only I have this problem that it takes such a long time to build a 
RT-page? Does I have configure RT in a wrong way that it is so slow?

This fact, that the pagebuilding takes a long time, isn´t only after login, it 
is also when I want to show a ticket and the requestor have some other 
tickets. Also to show the tickets history.

How can I get a better performace?
I also searched for "ILIKE", but I can´t find it in RT or DBIx::SearchBuilder. 
Where does the "ILIKE" be generated?

I would be very happy if someone can help me and say me what I have to change 
to get a better output.

Thanks,
Tina