RE: Asyn not automatically reconnecting

[Mark Rivers <rivers@cars.uchicago.edu>]

> Is it possible asyn is [un]intentionally not retrying the connection if it is closed gracefully?

In my tests the TCP server was running in another EPICS IOC, and I killed the server by typing "exit" at the IOC prompt.  It exited without a seg fault or other crash, so I think it closed the socket gracefully.  The IOC running StreamDevice immediately knew that the connection had been closed, without any retries, etc.  

This is the TCP server IOC running the echo server.

epics> exit
2018/02/27 13:29:41.281 echoListener: write error on: P5001:0: localhost:5001 disconnected:

This is the IOC running StreamDevice.  So it realizes the connection has been closed in under 1 second.

2018/02/27 13:29:42.259 corvette.cars.aps.anl.gov:5001 read 0
2018/02/27 13:29:42.259 Close corvette.cars.aps.anl.gov:5001 connection (fd 6): Read from broken connection
2018/02/27 13:29:42.259 13TEST:asyn1: exception 0, asynExceptionConnect
2018/02/27 13:29:42.259 ECHO read
2018/02/27 13:29:42.259 corvette.cars.aps.anl.gov:5001 write.
2018/02/27 13:29:42.259 corvette.cars.aps.anl.gov:5001 write 7
Hello\r\n
2018/02/27 13:29:42.259261 ECHO 13TEST:R:testOut: write failed because connection was closed by device
2018/02/27 13:29:42.264 ECHO addr -1 queueRequest priority 3 not lockHolder
2018/02/27 13:29:42.264 asynManager connect queueCallback port:ECHO
2018/02/27 13:29:42.264 Open connection to corvette.cars.aps.anl.gov:5001  reason:0  fd:-1

When the server IOC came back up it reconnected quickly.

Mark

> -----Original Message-----
> From: Daykin, Evan [mailto:daykin@frib.msu.edu]
> Sent: Tuesday, February 27, 2018 12:18 PM
> To: Mark Rivers <rivers@cars.uchicago.edu>
> Cc: tech-talk@aps.anl.gov
> Subject: RE: Asyn not automatically reconnecting
> 
> 
> Here's something interesting: asyn behaves correctly for us *only* when the client
> application is killed with SIGKILL, rather than SIGTERM. This results in TCP
> retransmissions and/or refused connections, then normal operation after a short while.  Is it
> possible asyn is [un]intentionally not retrying the connection if it is closed gracefully?