How to clear the error 314101

I am trying to create a mechanism of robust communication between a cRIO and a desktop application HMI. I tried to use the network stream I could avoid a lot of pitfalls in the management of a TCP connection. My RT code must be written to operate, and must recover gracefully network disconnections and brutal power loss. I can simulate these things quite effectively pulling the Ethernet cable or hit the button to Abort in LV

My goal of RT has a drive with the endpoint name ' commlink/drive '. My HMI has a burner with the name "commlink/recorder. HMI tries to establish the connection: he gave the Player URL ' / //commlink/player. This set works very well on the first pitch, but not after abandonment to the RT code. On the next run, the RT code often receives the error 314101 (endpoint with the same name already exists.) to create network Stream Reader Endpoint VI. I can't understand how to clear this error in order to recreate my end point and restore the comm link. Does anyone know how? Calling the destroy Stream Endpoint VI with refnum workflow does not work, and there are no other screws or properties that look like they manage the connection management.

I know I'm coming to the end of the conversation, but maybe some of this information will still be useful. While network streams were designed to manage multiple connection/disconnection cycles with the underlying TCP/IP connection on the network seamlessly, they were not intended to perfectly manage and survive many lives or running instances application with endpoints. All this is still possible, it will take more work on the part of the implementer. An important thing to remember when using network flow, is that once you've managed to connect the two end points, the destruction of one of the endpoints (via normal stop or an accident) will require the destruction of the other end. This means that you cannot simply reuse on RT target endpoint to communicate with several sessions of HMI. Instead, you will have to destroy endpoint on the target of RT and create another endpoint (probably with the same name) for the next session.

To do this, I recommend a state machine that is basically:

Create an endpoint - you can do either with an infinite time-out or a continuous loop with a timeout if you need meet orders of demolition/judgment of the user.
Continuous read/write in the flow in a loop. If at any time you receive a fatal error of read/write, leave the loop.
Destroy the end point and return to the State first, where you can wait for the next connection.

I would avoid going down the road of trapping of the specific error codes and writing conditional logic for each of them. Network stream API was designed such that for most / all cases, you should should not do this. If you have a healthy stream that can still move the data, you will not be getting errors from the read or write call (apart from a few things matter as if we were trying to read from an endpoint unique writing of corner). If you get an error in reading or writing, it usually means your workflow is dead and you need to create a new one if you want to continue to communicate with the remote application.

If you follow the above, I think that you are 90-95% of the way. As the article you link above mentions, there are additional considerations to take into account if you need to tolerate the application crashes / accidents of remote endpoint. In these scenearios, you basically have one of the three results of the application that always uses the end point which don't collide:

Demand will start to return errors: In some scenarios, we can detect the remote application is crashed, abnormally terminated, or whatever. In these cases, the read/write will return an error and the same pattern described above should work.
The flow will introduce a disconnected state and stay offline: This would generally happen with a fall in demand. In these cases, we cannot tell if the connectivity loss is due to a problem with the application or poor network conditions / remote computer. If you need to detect and recover from this scenario, you can do as a section of the link recommend you and write a timer that resets the stream network if the flow remains disconnected for a period of time. This could be incorporated in step 2 of the above state machine quite easily.
The flow will continue to make a State connected, even if the remote application is no longer functional: This should rarely occur and would occur only when the application on the remote computer hung or an impasse, but the network computer stack is still in operational condition. In this case, we cannot differentiate between a suspended application and the other simply has all the data to send. If you need to detect and recover from this condition, the best thing to do is probably to create a second timer that resets the stream if it's too long you've changed successfully read/written all the data.

I must emphasize scenarios 2 and 3 can be greatly simplified if you care to detecting the crash & block until another application attempts to connect to the application that is still ongoing. For example, restarting the application has crashed/hung will recreate an end point and try to re-establish the connection of flow. At this point, the application which is still running will go account he communicated with endpoint remote must have crashed and raises an error. At this point, the same state machine mentioned above should be sufficient to re-establish the connection. I think that this part could be more difficult in LV 2010 since the first attempt to apply newly restarted it also throw an error when you try to reconnect to the remote application. This will occur until the endpoint in the live application was also destroyed which means that you would have to execute a loop on creating calling until she succeeded. In 2011, we changed the behavior so the call to create the newly restarted application would cause the remote application to start return errors, but now we are still trying to establish the connection on the newly restarted application for up to the timeout limit. If you use a state similar to the above machine, you should be able to reconnect without having to write a loop around the call to Create in the newly restarted application.

Tags: NI Software

How to clear the error 314101

Similar Questions

Maybe you are looking for