ijc
Well Known Member
Dear List,
I need some urgent help with the JDECOMCONNECTOR. We have an application using this to call JDE business functions through VB
We are constantly getting the message
10054-recv returned 10054 (WSAECONNRESET): Connection was reset by peer
The message will be received constantly until I get an error "JDENET internal connection array full". This is because the connections still consume memory (hence the machine getting slower) and a connection even when it has been "disconnected" (JDE has a limit of 400 which I have upped to 800 for the moment).
Below is a couple of calls from the the JD Edwards Knowledge Garden. It acknowledges a problem with Windows but basically says it won't be fixed in JDE because the problem is caused by a remote disconnect
<Q>
10054-recv returned 10054 (WSAECONNRESET): Connection was reset by peer
Could not pass message to jdenet_k Kernel, Message aborted, type=2147483647
<A>
SAR 5914493:
No plans to change status of this sar. Development's reply: The above
statement is telling us that a client has closed its side of the socket
connection to the server without notifying it properly. The explanation as to
why this happens has been explained in detail in the * section of this SAR. In
the above statement we give out which client closed its socket as in
Host=0xa60cc68 (a-10,60- 96,cc-204,68-104; 10.96.204.104) . This is the IP
address of the client in hexadecimal representation.
<Q>
RE: Enterprise server log. Seeing errors in server log. Error messages:
"duplicate socket connection found and Freed" "
10054-RECV returned 10054 (WFAECONNRESTE) ; connection was reset by pier."
10054-recv returned 10054 (wsaeconnreset): connection was reset by peer
<A>
There was an older SAR (#3343287) that describes this type of error.
A ORIGINAL REQUEST
JDENet is not freeing user environments on the server when the OneWorld
client abends. Apparently, NT tcp/ip sends a different event than Unix & 400.
JDENet handles this problem fine on Unix & 400.
A Definition of the Error
WSAECONNRESET
(10054)
Connection reset by peer.
A existing connection was forcibly closed by the remote host. This normally
results if the peer application on the remote host is suddenly stopped, the
host is rebooted, or the remote host used a "hard close" (see setsockopt for
more information on the SO_LINGERoption on the remote socket.)
As far as we can make out we are not abending. I have had a NETMON on & all looks good. I can see the user log on & off through the security history.
The VB code looks as per JDE samples.
We can "replicate" the problem by logging on/logging off the app and shutting down the connector globe in the tray. To my mind I shouldn't be getting the error in this instance as I should be logged off (although I'm aware the globe is not shut down automatically be JDE)
All users are logged on through the connector with the same profile and may also have Oneworld open.
Has anyone come across this or similiar,
Any help appreciated,
Ian
I need some urgent help with the JDECOMCONNECTOR. We have an application using this to call JDE business functions through VB
We are constantly getting the message
10054-recv returned 10054 (WSAECONNRESET): Connection was reset by peer
The message will be received constantly until I get an error "JDENET internal connection array full". This is because the connections still consume memory (hence the machine getting slower) and a connection even when it has been "disconnected" (JDE has a limit of 400 which I have upped to 800 for the moment).
Below is a couple of calls from the the JD Edwards Knowledge Garden. It acknowledges a problem with Windows but basically says it won't be fixed in JDE because the problem is caused by a remote disconnect
<Q>
10054-recv returned 10054 (WSAECONNRESET): Connection was reset by peer
Could not pass message to jdenet_k Kernel, Message aborted, type=2147483647
<A>
SAR 5914493:
No plans to change status of this sar. Development's reply: The above
statement is telling us that a client has closed its side of the socket
connection to the server without notifying it properly. The explanation as to
why this happens has been explained in detail in the * section of this SAR. In
the above statement we give out which client closed its socket as in
Host=0xa60cc68 (a-10,60- 96,cc-204,68-104; 10.96.204.104) . This is the IP
address of the client in hexadecimal representation.
<Q>
RE: Enterprise server log. Seeing errors in server log. Error messages:
"duplicate socket connection found and Freed" "
10054-RECV returned 10054 (WFAECONNRESTE) ; connection was reset by pier."
10054-recv returned 10054 (wsaeconnreset): connection was reset by peer
<A>
There was an older SAR (#3343287) that describes this type of error.
A ORIGINAL REQUEST
JDENet is not freeing user environments on the server when the OneWorld
client abends. Apparently, NT tcp/ip sends a different event than Unix & 400.
JDENet handles this problem fine on Unix & 400.
A Definition of the Error
WSAECONNRESET
(10054)
Connection reset by peer.
A existing connection was forcibly closed by the remote host. This normally
results if the peer application on the remote host is suddenly stopped, the
host is rebooted, or the remote host used a "hard close" (see setsockopt for
more information on the SO_LINGERoption on the remote socket.)
As far as we can make out we are not abending. I have had a NETMON on & all looks good. I can see the user log on & off through the security history.
The VB code looks as per JDE samples.
We can "replicate" the problem by logging on/logging off the app and shutting down the connector globe in the tray. To my mind I shouldn't be getting the error in this instance as I should be logged off (although I'm aware the globe is not shut down automatically be JDE)
All users are logged on through the connector with the same profile and may also have Oneworld open.
Has anyone come across this or similiar,
Any help appreciated,
Ian
