JCO_SYSTEM_FAILURE

[Former Member]

We get this error when sending messages:

<?xml version="1.0" encoding="UTF-8" standalone="yes" ?>

- <!-- Request Message Mapping

-->

- <SAP:Error xmlns:SAP="http://sap.com/xi/XI/Message/30" xmlns:SOAP="http://schemas.xmlsoap.org/soap/envelope/" SOAP:mustUnderstand="">

<SAP:Category>XIServer</SAP:Category>

<SAP:Code area="MAPPING">JCO_SYSTEM_FAILURE</SAP:Code>

<SAP:P1><--Localization failed: ResourceBundle='com.sap.engine.services.rfcengine.RFCRes</SAP:P1>

<SAP:P2 />

<SAP:P3 />

<SAP:P4 />

<SAP:AdditionalText />

<SAP:ApplicationFaultMessage namespace="" />

<SAP:Stack>&quot;SYSTEM FAILURE&quot; during JCo call. &lt;--Localization failed: ResourceBundle='com.sap.engine.services.rfcengine.RFCRes</SAP:Stack>

<SAP:Retry>A</SAP:Retry>

</SAP:Error>

There seems to be an issue with RFC communication, but we can't seem to find the problem. This above error is logged a number of time, after which the server process will die. Each morning, the error is the same and we have to restart the Java Engine in order to get XI up and running again.

What RFC communication is meant here? Is it XI Java to XI ABAP, or XI ABAP to ECC ABAP? How can we differenciate between these forms of communication?

[Former Member]

Recreating the RFC connection seems to have solved this issue.

[Former Member]

Hi,

Is this is a problem with many flows or only 1-2 messages are failing.

apart from increaseing CPIC value as suggested in other post also log OSS call with SAP.

To me it seems that there are problem on your J2EE machine so SAP support will check your environment and will give you solution. It is quite possible that whole machine restart required. For RFC communication you need to set up environment variable also CPIC_MAX_CONN and check following for any hint.

SMGW - gateway monitor

SM04 - user monitor. Check if any user is consuming lot of RFC connection if yes then you need to eliminate root cause why that particular user is consuming lot of RFC connection.

Another reason could be incorrect written Java lookup code which open connection but didn't close properly.

Give more info to help you on this.

Regards,

Gourav

[Former Member]

HI W. Jeuken ,

There are two stack in XI 1) java 2) ABAP

these stacks foe communication use JCO (RFC)

for this gateway connections are set up for each RFC call

now it depends on your requirement and performance setting hoe many connections have you set up there.

now when there is a RFC call and the service is engaged. for example you have given MAX 5 connection and at that time all 5 connection are engaged so it will refuse the connection and you will get ERROr JCO_SYSTEM_FAILURE

since these are syn messages so you cant restart them you have to resend them

Everyday when you are restating J2EE engine all resources are freed and your messages atarts processing

this is your temporary solution

what i suggest you that increase your max no of connection

Thanks

sandeep sharma

PS:if helpful kindly reward points

[Former Member]

Hello,

The parameter gw/max_conn is set to the recommended value of 2000 and icm/max_conn is set to 500. What other parameter(s) could be of influence here?

Regards, Wilbert

[GabrielSagaya]

Did u do any changes to the interface mapping?

Everytime u do some changes to the interface mapping or the message mapping program,u have to again select the interface mapping in the receiver determination.

RFC communication is synchronous one

it can be R/3 (ABAP)>XI->JDBC

it can be ECC(Abap) >XI->JDBC

[Former Member]

Hi,

Please look in tho these links.May be they can help u.

**Reward if helpfull**

[aashish_sinha]

Hi,

check your sm59 destination, i guess there is some problem, Delete AI_RUNTIME_JCOSERVER and try again.

regards

Aashish Sinha

PS : reward points if helpful

[Former Member]

Hello,

That would imply it's the XI Java to XI ABAP communication that's causing the problem?

Why would the deletion of a RFC connection solve the issue?

Regards, Wilbert