[ http://issues.apache.org/jira/browse/AXISCPP-492?page=comments#action_60180 ] Fred Preston commented on AXISCPP-492: --------------------------------------
Hi All, The reason it can get stuck when reading data is if the message on the wire is not what HTTPTransport is waiting for (i.e a double CR/LF) then the loop(s) waiting for this specific data string will keep reading in the message until there is no message left and will then get stuck in this loop trying to read data from an empty channel for ever more. I have introduced a countdown into these loops that will count down from 100 (decrementing only when the length of the Rx data is 0, otherwise the count is set back to 100 (so on longer messages, the countdown will not reach zero whilst there is still information to be read)). Thus, if stuck in one of these loops with no more data to read, the countdown will eventually reach 0 and the loop will be exited. This should take care of this 'sticking' problem. Regards, Fred Preston. > Channel gets stuck in perpetual loop > ------------------------------------ > > Key: AXISCPP-492 > URL: http://issues.apache.org/jira/browse/AXISCPP-492 > Project: Axis-C++ > Type: Bug > Components: Transport (axis3) > Versions: current (nightly) > Reporter: Adrian Dick > > I have noticed the Channel can get stuck in a loop when a communication > problem occurs. > When a communication failure occurs while reading in a message, the Channel > can end up in a perpetual loop trying to read the incoming message. While > there is a read timeout before trying again, there doesn't appear to be any > logic to determine how many attempts have been made, and to give up. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://issues.apache.org/jira/secure/Administrators.jspa - If you want more information on JIRA, or have a bug to report see: http://www.atlassian.com/software/jira
