This from Michael Nugent @ Zvents:
I was wondering if there was anything we could do with our consulting
hours to increase the stability of the hyperrecord/thriftbroker
communication. As it is now, if the connection between these is
severed, the hyperrecord side needs to be restart after thriftbroker
is up and running. Is there any way we can add an option for this to
communicate via a named pipe instead of a TCP socket? Or have
hyperrecord recover more stably?
Me:
Possibly. Why does the connection between the application and the thriftbroker
get severed?
Michael:
Yesterday, for example, we were doing network maintenance. The loss
of link on the adapter caused the TCP socket to drop. Most
applications deal with this gracefully and reconnect at their earliest
convenience.
Another example is when thriftbroker goes a bit haywire and needs to
be restarted. Decoupling the link between these so that hyperrecord
could experience a brief outage and reconnect would be a great
benefit.
Does that answer what you're looking for? Is there a way to do this
already that I just don't know about?
Original issue reported on code.google.com by nuggetwh...@gmail.com on 17 Jan 2012 at 11:38
Original issue reported on code.google.com by
nuggetwh...@gmail.comon 17 Jan 2012 at 11:38