All of the APIs that boto uses to communicate with
cloud services are HTTP-based. That means that under the covers we are
communicating over sockets to distributed services. Sometimes these
services can be unresponsive or the communication layer between our
application and the service can be unreliable. To help manage this in a
reasonable way, timeouts are important. They allow our
application to detect a problem with the network or a service and attempt
to deal with it, or at least tell the user about it.
You can explicitly manage socket-level timeouts in
boto by adding a section to your boto config
file. The timeout is specified in seconds:
[Boto] http_socket_timeout = 5