Message Log not obtained

[itabhishek9]

Hi SDNites,

I am currently working on Dual Stack. I am able to do all monitoring from SXMB_MONI but when I go to Message monitor in adapter engine, I am not able to view the Message log there though as per configuration it should maintain the details for a day. But I can see that message log for SOAP adapter only is visible,

19.02.2016 12:50:04.697 Information The message was successfully retrieved from the receive queue

19.02.2016 12:50:04.702 Information Delivering to channel: communicaion channel name

19.02.2016 12:50:04.702 Information MP: processing local module localejbs/sap.com/com.sap.aii.af.soapadapter/XISOAPAdapterBean

19.02.2016 12:50:04.702 Information Message status set to DLNG

19.02.2016 12:50:04.703 Information SOAP: Request message entering the adapter processing with user J2EE_GUEST

Can you please let me know why the same is happening as my objective is to make the sysytem compatible with Single stack going forward and all monitoring can be done with no hassles.

Also do we need to set some configuration parameters to make this happen and if yes can you please help me with details.

Regards,

Abhi

[itabhishek9]

Hi Praveen,

Is there any memory size or the no. of messages which fall under the limit and when can I see the same.

Regards,

Abhi

[former_member182412]

Hi Abhishek,

I think it uses internally default value as 1000 messages, it is not configurable. The below is the example similar parameter for synchronous messages.

Regards,

Praveen.

[former_member186851]

Praveen I guess these parameters are decided based on memory and hardware installations of PI.

So to increase the options would be to increase the hardware right?

[itabhishek9]

@ Praveen - Can you please let me know the path view the above property.

Also please let me know your inputs on the below.

1. Parameter xiadapter.inbound.persistDuration.default and xiadapter.outbound.persistDuration.default are for persistance of message log in both adapter engine and Integration engine or not. If not can you help me with the parameters individually for the both the engines.

2. If we increase the no. of days from default 1 day to 30 days, what will be the impact on performace and memory and how can I measure it.

Regards,

Abhi

[former_member182412]

Hi Abhishek,

Change the property messaging.auditLog.memoryCache to false of messaging system service then you can be able to see the audit log no other configuration required. Try this option first, if does not work then let us know.

Configuration Management -> Infrastructure -> Java System Properties -> Services -> XPI Service : Messaging System

Regards,

Praveen.

[itabhishek9]

No pipeline steps dosen't have those properties.

[itabhishek9]

Thanks for the replies.

@Raghu - The scenario which I have mentioned is built has no ICO configuration and is built using the conventional Pipeline approach. Please let me know your inputs.

@Praveen - In both QA and PRD system the paramter you mentioned messaging.auditLog.memoryCache is maintianed as true. Also the File to split file scenario I mentioned has success in both QA and PRD.Please let me know why I am not able to view message log only in PRD but able to vew in QA. (Note : both message has been processed today and is in the limit of 24 hrs.)

Also all please let me know your views on,

1.Parameter xiadapter.inbound.persistDuration.default and xiadapter.outbound.persistDuration.default are for persistance of message log in both adapter engine and Integration engine or not. If not can you help me with the parameters individually for the both the engines.

2. If we increase the no. of days from default 1 day to 30 days, what will be the impact on performace and memory and how can I measure it.

Regards,

Abhi

[former_member186851]

Hello Abhishek,

Its been long since I worked in Classical scenario.

We had kind of same issue,After enabling we were ok.

Just check anyother Pipeline step has similar properties to be enabled.

[former_member182412]

Hi Abhishek,

In QA you have less messages processed but in production you processed more messages that is why you cannot see the audit log entries in production because after certain number of messages the audit log entries will  be deleted from the memory.

Regards,

Praveen

[itabhishek9]

Thanks Inaki / Vishu for the inputs.

I have already gone through the links shared above. I have following question here,

1. Parameter xiadapter.inbound.persistDuration.default and xiadapter.outbound.persistDuration.default are for persistance of message log in both adapter engine and Integration engine or not. If not can you help me with the parameters individually for the both the engines.

2. If we increase the no. of days from default 1 day to 30 days, what will be the impact on performace and memory and how can I measure it.

I have executed a File to File scenario (Sender file is split into multiple receiver file) in PRD enevironment and am not able to view the Message log for the same while if I execute the same in my QA environment I am able to view the message log. Can you let me know if there is any paramter which is affecting the same. What I have noticed in the past is that message log gets disappeared in 10-15 mins approx. (Note : paramters xiadapter.inbound.persistDuration.default and xiadapter.outbound.persistDuration.default are set for 1 day)

Regards,

Abhi

[former_member186851]

Abhishek,

Check your ICO(Advanced Settings)and make sure you enabled the below configs for payload view.

[former_member182412]

Hi Abhishek,

By default audit log entries are temporarily stored in memory not in the database.

The >= PI 7.1 AF Messaging System does not, by default, persist the audit logs of successfully processed asynchronous messages.

Instead the audit logs for successful asynchronous messages are stored in an internal memory cache which is flushed on system restart. The logs, while stored in the cache, are available for monitoring but cache entries are evicted - using a 'least recently used' (LRU) removal strategy - when the cache is full. The longevity of cache entries is dependent on message throughput.

The audit logs for unsuccessfully processed messages continue to be persisted and are therefore available for troubleshooting.

If you want to change this default behavior you need to change the parameter messaging.auditLog.memoryCache to false of XPI Service : Messaging System

Configuration Management -> Infrastructure -> Java System Properties -> Services -> XPI Service : Messaging System

Check below sap note for more details.

1314974 - Successful asynchronous messages - audit log persistence

Regards,

Praveen.

[iaki_vila]

Hi Abhishek,

About the persitence of the PI messages in the adapter engine depends about your PI version and you can't do the same for PI dual or single stack, because in dual stack you will have logs on the abap engine with its own persistence but in single stack you only will have adapter engine and you can configure the persistence at channel level, adapter and so on.

Vladilm Klimow have explained this behaviour better than me now in the next links:

For single stack:

For dual stack:How to persist the messages on Adapter engine? | SCN

Summing up, it is not useless to know the persistence parametrization in one PI version if you are planning to migrate to upper one.

Regards.

[vishnu_pallamreddy]

Hi Abhishek,

please check what are all the channels u have through channel Monitoring, if you have IDOC and HTTP  and XI adapters , those are not visible in Channel monitoring.