Hi Adarsh,

Below are those 2 parameters


cursor_sharing=FORCE
_b_tree_bitmap_plans=False

Refer Page 39 of Performance tuning document.

Regards/Vaibhav

On Thu, Oct 25, 2012 at 8:58 PM, Adarsh <[email protected]> wrote:

> ** Hi Vaibhav
>
> I am not able to view the Parameters which you have suggested . could you
> please send it again..
>
>
> Regards,
> Adarsh
>
> On Fri, Oct 26, 2012 at 7:17 AM, vaibhav wadekar <
> [email protected]> wrote:
>
>> ** Adarsh,
>>
>>
>> Based on this,it appears that it is using a bitmap index and I woud
>> recommend you to try below two parameters
>>
>>
>>
>>
>> So setting above two parameter will help you resolve the problem.
>>
>> Hope this helps.
>>
>> Regards/Vaibhav
>>
>>
>> Do you know the Cursor-Sharing Parameter set to what value? Did you try
>> setting that to FORCE and setting the
>>
>> On Thu, Oct 25, 2012 at 2:45 AM, Adarsh <[email protected]>wrote:
>>
>>> ** Hi Joe,
>>>
>>> We did not find anything in the Remedy logs .. but below are findings
>>> from DB end
>>>
>>> Please find the TOP I/O consuming SQL's in the database.
>>>
>>>
>>> ##################################################################
>>>
>>> SQL_ID  6zqfj98g199pp, child number 1
>>> -------------------------------------
>>> SELECT * FROM ( SELECT  T2118.C1,C1000000161,C1000000000,C303497300,C100
>>> 0000164,C7,C1000000218,C1000005261,C1000000217,C1000003009,C1000000018,C
>>> 1000000019,C1000000150,C179,C1000000422,C1000000715,C4,C1000000079,C1000
>>> 000162,C1000002613,C1000005785,C1000005791,C1000000560,C1000000151,C1000
>>> 000099,C1000000056,C1000000001,C1000000080,C1000000020,C1000000163,C3,C3
>>> 03497400,C923051203 FROM T2118 WHERE ((T2118.C7 < :"SYS_B_0") AND
>>> (:"SYS_B_1" = :"SYS_B_2") AND (:"SYS_B_3" = :"SYS_B_4")) ORDER BY
>>> C1000000161 DESC, :"SYS_B_5" ASC ) WHERE ROWNUM <= :"SYS_B_6"
>>>
>>> Plan hash value: 4273177787
>>>
>>>
>>> -----------------------------------------------------------------------------------------
>>> | Id  | Operation               | Name  | Rows  | Bytes |TempSpc| Cost
>>> (%CPU)| Time     |
>>>
>>> -----------------------------------------------------------------------------------------
>>> |   0 | SELECT STATEMENT        |       |       |       |       |
>>> 172K(100)|          |
>>> |*  1 |  COUNT STOPKEY          |       |       |       |
>>> |            |          |
>>> |   2 |   VIEW                  |       |   334K|   968M|       |
>>> 172K  (1)| 00:34:30 |
>>> |*  3 |    SORT ORDER BY STOPKEY|       |   334K|   332M|   373M|
>>> 172K  (1)| 00:34:30 |
>>> |*  4 |     FILTER              |       |       |       |
>>> |            |          |
>>> |*  5 |      TABLE ACCESS FULL  | T2118 |   334K|   332M|       |
>>> 99321   (1)| 00:19:52 |
>>>
>>> -----------------------------------------------------------------------------------------
>>>
>>> ##################################################################
>>>
>>> SQL_ID  6ttp45bbmxgr5, child number 1
>>> -------------------------------------
>>> SELECT T2723.C1 FROM T2723 WHERE (((:"SYS_B_0" = T2723.C179) OR
>>> (:"SYS_B_1" = T2723.C490009000)) AND (:"SYS_B_2" = T2723.C490008000)
>>> AND (T2723.C300364200 = :"SYS_B_3")) ORDER BY C300364900 ASC,C1 DESC
>>>
>>> Plan hash value: 3787592043
>>>
>>>
>>> ----------------------------------------------------------------------------
>>> | Id  | Operation          | Name  | Rows  | Bytes | Cost (%CPU)|
>>> Time     |
>>>
>>> ----------------------------------------------------------------------------
>>> |   0 | SELECT STATEMENT   |       |       |       | 52221
>>> (100)|          |
>>> |   1 |  SORT ORDER BY     |       |     1 |   118 | 52221   (2)|
>>> 00:10:27 |
>>> |*  2 |   TABLE ACCESS FULL| T2723 |     1 |   118 | 52220   (2)|
>>> 00:10:27 |
>>>
>>> ----------------------------------------------------------------------------
>>>
>>> ##################################################################
>>>
>>> SQL_ID  6vq7xsws6ga2q, child number 1
>>> -------------------------------------
>>> SELECT T1581.C1 FROM T1581 WHERE ((T1581.C179 IS NULL) AND
>>> ((T1581.C302943300 = :"SYS_B_0") OR (T1581.C301721000 = :"SYS_B_1")))
>>> ORDER BY C3 DESC, :"SYS_B_2" ASC
>>>
>>> Plan hash value: 736766182
>>>
>>>
>>> ----------------------------------------------------------------------------
>>> | Id  | Operation          | Name  | Rows  | Bytes | Cost (%CPU)|
>>> Time     |
>>>
>>> ----------------------------------------------------------------------------
>>> |   0 | SELECT STATEMENT   |       |       |       | 57186
>>> (100)|          |
>>> |   1 |  SORT ORDER BY     |       |     1 |    58 | 57186   (1)|
>>> 00:11:27 |
>>> |*  2 |   TABLE ACCESS FULL| T1581 |     1 |    58 | 57185   (1)|
>>> 00:11:27 |
>>>
>>> ----------------------------------------------------------------------------
>>>
>>> ##################################################################
>>>
>>> SQL_ID  gbzhnmm1db1tw, child number 1
>>> -------------------------------------
>>> SELECT * FROM ( SELECT  T2118.C1,C1000000161,C1000000000,C303497300,C100
>>> 0000164,C7,C1000000218,C1000005261,C1000000217,C1000003009,C1000000018,C
>>> 1000000019,C1000000150,C179,C1000000422,C1000000715,C4,C1000000079,C1000
>>> 000162,C1000002613,C1000005785,C1000005791,C1000000560,C1000000151,C1000
>>> 000099,C1000000056,C1000000001,C1000000080,C1000000020,C1000000163,C3,C3
>>> 03497400,C923051203 FROM T2118 WHERE ((T2118.C7 < :"SYS_B_00") AND
>>> ((T2118.C1000000079 = :"SYS_B_01") OR (T2118.C1000000427 = :"SYS_B_02")
>>> OR (T2118.C1000000079 = :"SYS_B_03") OR (T2118.C1000000427 =
>>> :"SYS_B_04") OR (T2118.C1000000079 = :"SYS_B_05") OR (T2118.C1000000427
>>> = :"SYS_B_06") OR (T2118.C1000000079 = :"SYS_B_07") OR
>>> (T2118.C1000000427 = :"SYS_B_08") OR (T2118.C1000000079 = :"SYS_B_09")
>>> OR (T2118.C1000000427 = :"SYS_B_10") OR (T2118.C1000000079 =
>>> :"SYS_B_11") OR (T2118.C1000000427 = :"SYS_B_12") OR (T2118.C1000000079
>>> = :"SYS_B_13") OR (T2118.C1000000427 = :"SYS_B_14") OR
>>> (T2118.C1000000079 = :"SYS_B_15") OR (T2118.C100000042
>>>
>>> Plan hash value: 2646718021
>>>
>>>
>>> --------------------------------------------------------------------------------------------------------------
>>> | Id  | Operation                       | Name               | Rows  |
>>> Bytes |TempSpc| Cost (%CPU)| Time     |
>>>
>>> --------------------------------------------------------------------------------------------------------------
>>> |   0 | SELECT STATEMENT                |                    |
>>> |       |       | 92257 (100)|          |
>>> |*  1 |  COUNT STOPKEY                  |                    |
>>> |       |       |            |          |
>>> |   2 |   VIEW                          |                    | 75850 |
>>> 219M|       | 92257   (1)| 00:18:28 |
>>> |*  3 |    SORT ORDER BY STOPKEY        |                    | 75850
>>> |    67M|    74M| 92257   (1)| 00:18:28 |
>>> |*  4 |     FILTER                      |                    |
>>> |       |       |            |          |
>>> |   5 |      TABLE ACCESS BY INDEX ROWID| T2118              | 75850
>>> |    67M|       | 77445   (1)| 00:15:30 |
>>> |*  6 |       INDEX SKIP SCAN           | I2118_1000000427_1 | 75850
>>> |       |       |  3505   (1)| 00:00:43 |
>>>
>>> --------------------------------------------------------------------------------------------------------------
>>>
>>> ##################################################################
>>>
>>>
>>>
>>>
>>>
>>> On Wed, Oct 24, 2012 at 11:15 PM, Joe Martin D'Souza 
>>> <[email protected]>wrote:
>>>
>>>> I'd also suggest to run the same SQL that is executed (capture it from
>>>> the SQL logs) directly against the DB and measure the response time. If
>>>> there is a significant difference in timing where the DB search is returned
>>>> much quicker, it would immediately tell you if it’s a problem at the DB
>>>> level or something within Remedy.. if within Remedy it could be factors
>>>> like (or a combination of factors like) number of records returned, whether
>>>> or not chunking is enabled on the AR side, etc. etc..
>>>>
>>>> You could find out the same thing by taking a combined SQL and API log,
>>>> and analyzing that log..
>>>>
>>>> Joe
>>>>
>>>> -----Original Message----- From: Longwing, LJ CTR MDA/IC
>>>> Sent: Wednesday, October 24, 2012 11:34 AM Newsgroups:
>>>> public.remedy.arsystem.general
>>>>
>>>> To: [email protected]
>>>> Subject: Re: Sarch on HPD:Help Desk from taking a very long time.
>>>>
>>>> And if you do a Show Plan, or, in whatever way is supported by your DB,
>>>> what does it say it's doing for those 25 seconds?
>>>>
>>>> -----Original Message-----
>>>> From: Action Request System discussion list(ARSList) [mailto:
>>>> [email protected]] On Behalf Of Adarsh
>>>> Sent: Wednesday, October 24, 2012 9:30 AM
>>>> To: [email protected]
>>>> Subject: Re: Sarch on HPD:Help Desk from taking a very long time.
>>>>
>>>> ** I have following indexes on the HPD form
>>>>
>>>>
>>>> On Wed, Oct 24, 2012 at 8:45 PM, Adarsh <[email protected]>
>>>> wrote:
>>>>
>>>>
>>>> Thanks Tauf,
>>>>
>>>> I am trying to search all open tickets (status < "Resolved") assigned
>>>> to a particular group
>>>>
>>>> 'Status*' < "Resolved" AND 'Assigned Group*+' = "Service Desk"
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> On Wed, Oct 24, 2012 at 8:14 PM, Tauf Chowdhury <[email protected]>
>>>> wrote:
>>>>
>>>>
>>>> ** There are various factors such as how much data you are searching
>>>> on, indexing, etc...
>>>> What kind of searches are you doing? Are you searching for words in the
>>>> summary or Notes field?
>>>> What if you left everything blank and only select Status = "In
>>>> Progress" ? Does the data come back quicker? The status field is indexed by
>>>> default (I think) so just check that and report back. We would need more
>>>> info to help you though.
>>>>
>>>>
>>>> On Wed, Oct 24, 2012 at 10:40 AM, Adarsh <[email protected]>
>>>> wrote:
>>>>
>>>>
>>>> ** Hi ,
>>>>
>>>>
>>>> We are using Remedy 7.6.04 SP2 , currently we are facing a weird issue,
>>>> The application is working perfectly fine except for searches on HPD:Help
>>>> Desk form,
>>>>
>>>> everything from database end  has been checked, we did a re-org of the
>>>> tables and also rebuilt the indexes but still the issue is not resolved.
>>>>
>>>> its takes around 25 seconds to refresh the data..
>>>>
>>>> Has anyone faced such an issue..  ??? :-(
>>>>
>>>>
>>>> Regards,
>>>> Adarsh
>>>>
>>>>
>>>>
>>>> _attend WWRUG12 www.wwrug.com ARSlist: "Where the Answers Are"_
>>>>
>>>>
>>>>
>>>>
>>>> --
>>>> Tauf Chowdhury
>>>>
>>>>
>>>> _attend WWRUG12 www.wwrug.com ARSlist: "Where the Answers Are"_
>>>>
>>>>
>>>>
>>>> _attend WWRUG12 www.wwrug.com ARSlist: "Where the Answers Are"_
>>>>
>>>> ______________________________**______________________________**
>>>> ___________________
>>>> UNSUBSCRIBE or access ARSlist Archives at www.arslist.org
>>>> attend wwrug12 www.wwrug12.com ARSList: "Where the Answers Are"
>>>> ______________________________**______________________________**
>>>> ___________________
>>>> UNSUBSCRIBE or access ARSlist Archives at www.arslist.org
>>>> attend wwrug12 www.wwrug12.com ARSList: "Where the Answers Are"
>>>>
>>>
>>> _attend WWRUG12 www.wwrug.com ARSlist: "Where the Answers Are"_
>>>
>>
>> _attend WWRUG12 www.wwrug.com ARSlist: "Where the Answers Are"_
>>
>
> _attend WWRUG12 www.wwrug.com ARSlist: "Where the Answers Are"_
>

_______________________________________________________________________________
UNSUBSCRIBE or access ARSlist Archives at www.arslist.org
attend wwrug12 www.wwrug12.com ARSList: "Where the Answers Are"

Reply via email to