[Building Sakai] Sakai 2.9.2 and Kernel 1.3.2 problems during load testing

[Matthew Jones]

I looked at this for a little today and don't see anything wrong with the
site, either with authenticated or unauthenticated users.

It's possible there's some edge cases it's missing? I see anything in a
typical case.


On Mon, Aug 5, 2013 at 8:02 PM, Matthew Jones <matthew at longsight.com> wrote:

> This feature detects if the remote site has the X-FRAME-OPTION headers and
> attempts to cache that value on the tool so it doesn't recheck. This is
> done if the popup button is not checked because it means that the site
> won't work embedded. I'm guessing the cache is there just in-case this
> option changes? There was probably a time when sites with this header set
> now didn't have it set in the past, but you wouldn't need to check very
> often.
>
> If you set the property Beth mentioned, then it returns false and it
> doesn't check/doesn't cache.
>
>
> On Mon, Aug 5, 2013 at 7:47 PM, Kusnetz, Jeremy <JKusnetz at apus.edu> wrote:
>
>>  We are seeing a much higher increase of that event since upgrading
>> too.  Can someone tell me what the effect of setting this parameter does?
>> ****
>>
>> ** **
>>
>> ****
>>
>> ** **
>>
>> -----Original Message-----
>> From: sakai-dev-bounces at collab.sakaiproject.org [mailto:
>> sakai-dev-bounces at collab.sakaiproject.org] On Behalf Of Beth Kirschner
>> Sent: Monday, August 05, 2013 5:43 PM
>> To: sakai-dev at collab.sakaiproject.org sakai
>> Subject: Re: [Building Sakai] Sakai 2.9.2 and Kernel 1.3.2 problems
>> during load testing
>>
>> ** **
>>
>> We've found the source of the high db cpu load in the web portlet code -
>> I've filed the following JIRA:****
>>
>>                 https://jira.sakaiproject.org/browse/SAK-23841****
>>
>> ** **
>>
>> An easy workaround for those who might see this problem is to set the
>> following in sakai.properties:****
>>
>>                 iframe.xframe.cachetime=0 ****
>>
>> ** **
>>
>> - Beth****
>>
>> ** **
>>
>> On Aug 1, 2013, at 6:02 PM, Steve Swinsburg <steve.swinsburg at gmail.com>
>> wrote:****
>>
>> ** **
>>
>> > It's possible to restructure that query to give the database a better
>> chance at optimising. It joins everything only to discard most of it at the
>> end. I know the execution time isn't the real issue but everywhere we can
>> optimise for performance we should.****
>>
>> > ****
>>
>> > Cheers,****
>>
>> > S****
>>
>> > ****
>>
>> > ****
>>
>> > ****
>>
>> > Sent from my iPad****
>>
>> > ****
>>
>> > On 02/08/2013, at 6:56, Beth Kirschner <bkirschn at umich.edu> wrote:****
>>
>> > ****
>>
>> >> Hi all,****
>>
>> >> ****
>>
>> >> We're running load tests against a Sakai 2.9.2 based build, with a few
>> kernel patches (mentioned at bottom of email) and seeing a 14 times
>> increase with the number of executions of the SAKAI_REALM_FUNCTION query
>> below. While we're digging into why, I thought I'd ask this esteemed
>> audience if they had any thoughts as to what might be causing this increase.
>> ****
>>
>> >> ****
>>
>> >> Thanks,****
>>
>> >> - Beth****
>>
>> >> ****
>>
>> >> ### Results from our Sakai 2.9.2 based build:****
>>
>> >> ****
>>
>> >>       Elapsed                  Elapsed Time****
>>
>> >>       Time (s)    Executions  per Exec (s)  %Total   %CPU    %IO
>> SQL Id****
>>
>> >> ---------------- -------------- ------------- ------ ------ ------
>> -------------****
>>
>> >>        2,676.6        748,379          0.00   17.1   99.5     .0
>> 3ryj6azagfbwv****
>>
>> >> Module: JDBC Thin Client****
>>
>> >> select DISTINCT FUNCTION_NAME from SAKAI_REALM_FUNCTION SRF inner ****
>>
>> >> join SAKAI_REA LM_RL_FN SRRF on SRF.FUNCTION_KEY = SRRF.FUNCTION_KEY *
>> ***
>>
>> >> inner join SAKAI_REALM_ROL E SRR on SRRF.ROLE_KEY = SRR.ROLE_KEY ****
>>
>> >> inner join SAKAI_REALM SR on SRRF.REALM_KE Y = SR.REALM_KEY where ****
>>
>> >> SRR.ROLE_NAME = :1 and SR.REALM_ID IN (:2 ,:3 ,:4 )****
>>
>> >> ****
>>
>> >> ****
>>
>> >> ### Results from our Sakai 2.9.1 based build****
>>
>> >> ****
>>
>> >>    176.6       53,040       0.00    1.8      177.6   99.4     .0
>> 3ryj6azagfbwv****
>>
>> >> Module: JDBC Thin Client****
>>
>> >> select DISTINCT FUNCTION_NAME from SAKAI_REALM_FUNCTION SRF inner ****
>>
>> >> join SAKAI_REA LM_RL_FN SRRF on SRF.FUNCTION_KEY = SRRF.FUNCTION_KEY *
>> ***
>>
>> >> inner join SAKAI_REALM_ROL E SRR on SRRF.ROLE_KEY = SRR.ROLE_KEY ****
>>
>> >> inner join SAKAI_REALM SR on SRRF.REALM_KE Y = SR.REALM_KEY where ****
>>
>> >> SRR.ROLE_NAME = :1 and SR.REALM_ID IN (:2 ,:3 ,:4 )****
>>
>> >> ****
>>
>> >> ****
>>
>> >> ## Patches previously run in production****
>>
>> >> KNL-1056        Need more efficient caching of notifyingMemoryStore
>> object ****
>>
>> >> KNL-1016        Allow multiple site types for isCourseSite(),
>> isProjectSite(), isPortfolioSite() ****
>>
>> >> ****
>>
>> >> ## Patches new to this build****
>>
>> >> KNL-989         Add tool groups****
>>
>> >> KNL-1060        Cache invalidation for grants isn't happening when
>> editing a special realm****
>>
>> >> KNL-990         Joinable Groups****
>>
>> >> KNL-885         The group property to enable Site Info to see groups
>> should be moved out of site-manage and into kernel-api****
>>
>> >> KNL-874         Replace getFormattedMessage array argument with a
>> varargs ****
>>
>> >> ****
>>
>> >> _______________________________________________****
>>
>> >> sakai-dev mailing list****
>>
>> >> sakai-dev at collab.sakaiproject.org****
>>
>> >> http://collab.sakaiproject.org/mailman/listinfo/sakai-dev****
>>
>> >> ****
>>
>> >> TO UNSUBSCRIBE: send email to
>> sakai-dev-unsubscribe at collab.sakaiproject.org with a subject of
>> "unsubscribe"****
>>
>> ** **
>>
>> _______________________________________________****
>>
>> sakai-dev mailing list****
>>
>> sakai-dev at collab.sakaiproject.org****
>>
>> http://collab.sakaiproject.org/mailman/listinfo/sakai-dev****
>>
>> ** **
>>
>> TO UNSUBSCRIBE: send email to
>> sakai-dev-unsubscribe at collab.sakaiproject.org with a subject of
>> "unsubscribe"****
>>
>> This message is private and confidential. If you have received it in
>> error, please notify the sender and remove it from your system.
>>
>> _______________________________________________
>> sakai-dev mailing list
>> sakai-dev at collab.sakaiproject.org
>> http://collab.sakaiproject.org/mailman/listinfo/sakai-dev
>>
>> TO UNSUBSCRIBE: send email to
>> sakai-dev-unsubscribe at collab.sakaiproject.org with a subject of
>> "unsubscribe"
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://collab.sakaiproject.org/pipermail/sakai-dev/attachments/20130805/e21b60cc/attachment.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/png
Size: 21485 bytes
Desc: not available
Url : http://collab.sakaiproject.org/pipermail/sakai-dev/attachments/20130805/e21b60cc/attachment.png