socket and Ctl-C workaround?

Tony Nelson *firstname*nlsnews at georgea*lastname*.com
Thu Jun 15 21:34:07 CEST 2006


I've been trying to figure out why Ctl-C sometimes doesn't interrupt 
yum.  It appears to be unresolved Python bug 926423, unresolved proposed 
patch 1102879, don't know if anything ever came of it.  Note that I 
cannot ask all yum users to apply the patch.  I'm not sure I should be 
getting rid of the timeout, either.

When a Ctl-C is lost, sometimes it is silently eaten, and other times it 
becomes socket.error(11,...) (EWOULDBLOCK).  As (AFAICT) the socket in 
question is a blocking socket, I could use that nonsense exception to 
detect at least some of the lost Ctl-Cs (eww).  The socket does have a 
timeout.

Note that I am using FC5 Linux, Python 2.4.2.
________________________________________________________________________
TonyN.:'                        *firstname*nlsnews at georgea*lastname*.com
      '                                  <http://www.georgeanelson.com/>



More information about the Python-list mailing list