JCO: connection to partner broken

[Former Member]

Hi

Can anyone help us? We have a J2EE Engine 6.20, using SAP JCo 2.0.x to make a connection to another 6.20-System in the same net zone (=no firewall in between). From time to time (unpredictable!), we are recieving an error as stated below:

-

---

com.sap.mw.jco.JCO$Exception: (104) RFC_ERROR_SYSTEM_FAILURE: CPIC-CALL: 'CMRCV'

LOCATION CPIC (TCP/IP) on local host

ERROR connection to partner broken

TIME Wed Oct 27 14:11:24 2004

RELEASE 620

COMPONENT NI (network interface)

VERSION 36

RC -6

MODULE ninti.c

LINE 1091

DETAIL NiPRead (xxx.xxx.xxx.xx/3300)

SYSTEM CALL recv

ERRNO 10054

ERRNO TEXT WSAECONNRESET: Connection reset by peer

COUNTER 3

-

---

All involed Systems are based on Windows NT. Has anyone made a similar experience? Does anyone know the reason for this error and how to fix it?

Regards,

SyBOR AG

Klaus Imfeld

[Oliver_S]

did you patch to the newest JCO version for 6.20?

CPIC-CALL: 'CMRCV' no data received means normally means "no data received". It is also possible that connection pool is full or timed out and partner server terminated the session. to knwo exactly what happened you have to make a JCO connection trace and a RFC trace on R/3.

see note 598891 or jco documentation.

https://websmp205.sap-ag.de/~form/handler?_APP=01100107900000000342&_EVENT=REDIR&_NNUM=598891&_NLANG...

[Former Member]

I met the error 10054 too,

some people say it's may caused by the network environment

change,you add a new router/switch,or a new firewall.i 'm not sure

[Former Member]

Hi, All:

Any thing news for this topic ?

it's really a hard nut to crack. if you have any ideal or suggestion,pls say your opinion.

thanks a lot.

[Former Member]

hi, everyone,

is there any new info about this error msg? since i met similar problem when trying to call SAP from external java program. the problem occurs randomly.

thanks!

alan

[Former Member]

Hi Alan,

Check out OSS notes # 26086, 155147

and, especially, 413330.

In a few words it makes sense to test the connection quality between source and destination via niping, ping, tracert and adjust a number of retransmissions if required.

Regards,

Mike

[Former Member]

I'm getting a simliar issue.

com.sap.mw.jco.JCO$Exception: (104) RFC_ERROR_SYSTEM_FAILURE: CPIC-CALL: 'CMRCV' no data received

java.lang.Exception: com.sap.mw.jco.JCO$Exception: (104) RFC_ERROR_SYSTEM_FAILURE: CPIC-CALL: 'CMRCV' no data received at jrun__internalorderlist2ejsp16._jspService(jrun__internalorderlist2ejsp16.java:201) at jrun.jsp.runtime.HttpJSPServlet.service

In our case, the R/3 system is 4.7/6.2, and we recently upgraded from JCO 1.3 to 2.1.6 (mostly to fix problems with utf8 support). Things were fine with 1.3. The upgrade to 2.1.6 seems to be the trigger for these "random" errors.

[Former Member]

Were you ever able to find the source of this problem?

Dan

[Former Member]

How was the bandwidth usage /io saturation of source / destination during the connection attempt.

Was there a dialog process available for RFC calls during the connection attempt ?

Did you experience micro network interruptions / collisions ?

Enjoy

[gregorw]

Hello Klaus,

was the application server or message server shut down or was the system rebooted?

Regards

Gregor

[Former Member]

Hello Gregor

No, there was absolutly no shutdown and no system reboot. That's why it looks so strange to me.

Klaus

[former_member584282]

We are also facing similar problem. We are running 4.7, and some of our users [ not all ] get the same message and this happens at random interval and with random users. At that time other users does not have any problems. And these are just regular SAP transactions that are giving this error.

Microsoft describes this error as follows

"This normally results if the peer application on the remote hos is suddenly stopped, the host is rebooted, the host or remote network interface is disabled, orthe remote host uses a hard close. This error may also result if a connection was broken due to keep-alive activity detecting a failure while one or more operations are in progress. Operations that were in progress fail with WSAENETRESET. Subsequent operations fail with WSAECONNRESET."

Would love to read a solution about this from you all.

Thanks,

Amit Purohit.