Re: [Dev] [BPS] - Performance test - Long running

2015-10-15 Thread Hasitha Aravinda 
Hi Chathurika,

On Fri, Oct 16, 2015 at 9:47 AM, Chathurika De Silva 
wrote:

> Furthermore, following WARNING is observed frequently in the cluster at
> the 4th day of the long running test
>
> TID: [-1] [] [2015-10-16 03:58:45,455]  WARN
> {org.apache.ode.scheduler.simple.SimpleScheduler} -  Dispatching jobs with
> more than 5 minutes delay. Either the server was down for some time or the
> job load is greater than available capacity
> {org.apache.ode.scheduler.simple.SimpleScheduler}
>
>
​ODE gives this warning when there is a high load on BPS server. Because of
the load, some immediate jobs are getting queued in scheduler. As Warning
msg says, there are jobs which are older than 5 mins in job queue.

To over come this issue, you can increase number of ODE scheduler threads
along with number of DB connections allocated to BPS data source and
allocated sessions from database server. However due to database
contentions, you can't increase this number beyond a certain
limit.
​This value is mainly depend on performance of your DB and BPS vms.

​


> Also although i clean up the database once in every day at the 4th day of
> the test, the database ran out of space and had to alter by adding another
> datafile. What is the recommended frequency for the cleanup?
>
>
​Based on your test parameters, you are generating a heavy load. I would
suggest to clean db every day. Use DB clean scripts for this. Disable ODE
clean up task in bps.xml as it consumes ODE scheduler threads to clean up
db.

Thanks,
Hasitha.
 ​

> Please find the test parameters in [1]
>
> [1] https://wso2.org/jira/browse/BPS-912
>
>
> On Thu, Oct 15, 2015 at 3:14 PM, Chathurika De Silva  > wrote:
>
>> Hi
>>
>> While running the $subject on a manager, 2 worker cluster, noticed a
>> dis-ambiguity reported in [1]
>>
>> Can you please provide an explanation on this? Your help is greatly
>> appreciated.
>>
>> [1] https://wso2.org/jira/browse/BPS-912
>>
>> Thank You
>> Erandi
>>
>> --
>> *Chathurika Erandi*
>> Software Engineer,
>> WSO2 Inc.
>>
>> *E-mail:* chathuri...@wso2.com
>> *Mobile: *+94714 328 612
>> *Blog*: chathurikaerandi.blogspot.com
>>
>
>
>
> --
> *Chathurika Erandi*
> Software Engineer,
> WSO2 Inc.
>
> *E-mail:* chathuri...@wso2.com
> *Mobile: *+94714 328 612
> *Blog*: chathurikaerandi.blogspot.com
>



-- 
--
Hasitha Aravinda,
Senior Software Engineer,
WSO2 Inc.
Email: hasi...@wso2.com
Mobile : +94 718 210 200
___
Dev mailing list
Dev@wso2.org
http://wso2.org/cgi-bin/mailman/listinfo/dev

Re: [Dev] [BPS] - Performance test - Long running

2015-10-15 Thread Chathurika De Silva 
Furthermore, following WARNING is observed frequently in the cluster at the
4th day of the long running test

TID: [-1] [] [2015-10-16 03:58:45,455]  WARN
{org.apache.ode.scheduler.simple.SimpleScheduler} -  Dispatching jobs with
more than 5 minutes delay. Either the server was down for some time or the
job load is greater than available capacity
{org.apache.ode.scheduler.simple.SimpleScheduler}

Also although i clean up the database once in every day at the 4th day of
the test, the database ran out of space and had to alter by adding another
datafile. What is the recommended frequency for the cleanup?

Please find the test parameters in [1]

[1] https://wso2.org/jira/browse/BPS-912


On Thu, Oct 15, 2015 at 3:14 PM, Chathurika De Silva 
wrote:

> Hi
>
> While running the $subject on a manager, 2 worker cluster, noticed a
> dis-ambiguity reported in [1]
>
> Can you please provide an explanation on this? Your help is greatly
> appreciated.
>
> [1] https://wso2.org/jira/browse/BPS-912
>
> Thank You
> Erandi
>
> --
> *Chathurika Erandi*
> Software Engineer,
> WSO2 Inc.
>
> *E-mail:* chathuri...@wso2.com
> *Mobile: *+94714 328 612
> *Blog*: chathurikaerandi.blogspot.com
>



-- 
*Chathurika Erandi*
Software Engineer,
WSO2 Inc.

*E-mail:* chathuri...@wso2.com
*Mobile: *+94714 328 612
*Blog*: chathurikaerandi.blogspot.com
___
Dev mailing list
Dev@wso2.org
http://wso2.org/cgi-bin/mailman/listinfo/dev

Re: [Dev] [BPS] - Performance test - Long running

2015-10-15 Thread Vinod Kavinda 
Hi Chathurika,
The first error you have reported is a known issue in hazelcast if you
interrupted a node. Please refer [2] for more info. There is a fix given.
No idea either we have included it in latest fixes of hazelcast.

The second warning message is logged, when a batch of jobs are enqueued
after a certain max delay time. Just to inform that they are delayed jobs.
The warning message explains the reasons itself.


[2] - https://github.com/hazelcast/hazelcast/issues/1815

Regards,
Vinod Kavinda

On Fri, Oct 16, 2015 at 9:47 AM, Chathurika De Silva 
wrote:

> Furthermore, following WARNING is observed frequently in the cluster at
> the 4th day of the long running test
>
> TID: [-1] [] [2015-10-16 03:58:45,455]  WARN
> {org.apache.ode.scheduler.simple.SimpleScheduler} -  Dispatching jobs with
> more than 5 minutes delay. Either the server was down for some time or the
> job load is greater than available capacity
> {org.apache.ode.scheduler.simple.SimpleScheduler}
>
> Also although i clean up the database once in every day at the 4th day of
> the test, the database ran out of space and had to alter by adding another
> datafile. What is the recommended frequency for the cleanup?
>
> Please find the test parameters in [1]
>
> [1] https://wso2.org/jira/browse/BPS-912
>
>
> On Thu, Oct 15, 2015 at 3:14 PM, Chathurika De Silva  > wrote:
>
>> Hi
>>
>> While running the $subject on a manager, 2 worker cluster, noticed a
>> dis-ambiguity reported in [1]
>>
>> Can you please provide an explanation on this? Your help is greatly
>> appreciated.
>>
>> [1] https://wso2.org/jira/browse/BPS-912
>>
>> Thank You
>> Erandi
>>
>> --
>> *Chathurika Erandi*
>> Software Engineer,
>> WSO2 Inc.
>>
>> *E-mail:* chathuri...@wso2.com
>> *Mobile: *+94714 328 612
>> *Blog*: chathurikaerandi.blogspot.com
>>
>
>
>
> --
> *Chathurika Erandi*
> Software Engineer,
> WSO2 Inc.
>
> *E-mail:* chathuri...@wso2.com
> *Mobile: *+94714 328 612
> *Blog*: chathurikaerandi.blogspot.com
>
> ___
> Dev mailing list
> Dev@wso2.org
> http://wso2.org/cgi-bin/mailman/listinfo/dev
>
>


-- 
Vinod Kavinda
Software Engineer
*WSO2 Inc. - lean . enterprise . middleware .*
Mobile : +94 (0) 712 415544
Blog : http://soatechflicks.blogspot.com/
___
Dev mailing list
Dev@wso2.org
http://wso2.org/cgi-bin/mailman/listinfo/dev

Re: [Dev] [BPS] - Performance test - Long running

2015-10-15 Thread Vinod Kavinda 
Hi Chathurika,

Missed the last part.

The required frequency for database cleanup depends on the request
frequency and the space available for DB. So it is hard to give a
recommended frequency.

Regards,
Vinod Kavinda

On Fri, Oct 16, 2015 at 10:18 AM, Vinod Kavinda  wrote:

> Hi Chathurika,
> The first error you have reported is a known issue in hazelcast if you
> interrupted a node. Please refer [2] for more info. There is a fix given.
> No idea either we have included it in latest fixes of hazelcast.
>
> The second warning message is logged, when a batch of jobs are enqueued
> after a certain max delay time. Just to inform that they are delayed jobs.
> The warning message explains the reasons itself.
>
>
> [2] - https://github.com/hazelcast/hazelcast/issues/1815
>
> Regards,
> Vinod Kavinda
>
> On Fri, Oct 16, 2015 at 9:47 AM, Chathurika De Silva  > wrote:
>
>> Furthermore, following WARNING is observed frequently in the cluster at
>> the 4th day of the long running test
>>
>> TID: [-1] [] [2015-10-16 03:58:45,455]  WARN
>> {org.apache.ode.scheduler.simple.SimpleScheduler} -  Dispatching jobs with
>> more than 5 minutes delay. Either the server was down for some time or the
>> job load is greater than available capacity
>> {org.apache.ode.scheduler.simple.SimpleScheduler}
>>
>> Also although i clean up the database once in every day at the 4th day of
>> the test, the database ran out of space and had to alter by adding another
>> datafile. What is the recommended frequency for the cleanup?
>>
>> Please find the test parameters in [1]
>>
>> [1] https://wso2.org/jira/browse/BPS-912
>>
>>
>> On Thu, Oct 15, 2015 at 3:14 PM, Chathurika De Silva <
>> chathuri...@wso2.com> wrote:
>>
>>> Hi
>>>
>>> While running the $subject on a manager, 2 worker cluster, noticed a
>>> dis-ambiguity reported in [1]
>>>
>>> Can you please provide an explanation on this? Your help is greatly
>>> appreciated.
>>>
>>> [1] https://wso2.org/jira/browse/BPS-912
>>>
>>> Thank You
>>> Erandi
>>>
>>> --
>>> *Chathurika Erandi*
>>> Software Engineer,
>>> WSO2 Inc.
>>>
>>> *E-mail:* chathuri...@wso2.com
>>> *Mobile: *+94714 328 612
>>> *Blog*: chathurikaerandi.blogspot.com
>>>
>>
>>
>>
>> --
>> *Chathurika Erandi*
>> Software Engineer,
>> WSO2 Inc.
>>
>> *E-mail:* chathuri...@wso2.com
>> *Mobile: *+94714 328 612
>> *Blog*: chathurikaerandi.blogspot.com
>>
>> ___
>> Dev mailing list
>> Dev@wso2.org
>> http://wso2.org/cgi-bin/mailman/listinfo/dev
>>
>>
>
>
> --
> Vinod Kavinda
> Software Engineer
> *WSO2 Inc. - lean . enterprise . middleware .*
> Mobile : +94 (0) 712 415544
> Blog : http://soatechflicks.blogspot.com/
>
>


-- 
Vinod Kavinda
Software Engineer
*WSO2 Inc. - lean . enterprise . middleware .*
Mobile : +94 (0) 712 415544
Blog : http://soatechflicks.blogspot.com/
___
Dev mailing list
Dev@wso2.org
http://wso2.org/cgi-bin/mailman/listinfo/dev