[openstack-dev] [eventlet] should we use spawn instead of spawn_n?

Russell Bryant rbryant at redhat.com
Mon Dec 2 13:58:05 UTC 2013


On 11/29/2013 01:01 AM, Jian Wen wrote:
> eventlet.spawn_n is the same as eventlet.spawn, but it’s not possible
> to know how the function terminated (i.e. no return value or exceptions)[1].
> If an exception is raised in the function, spawn_n prints a stack trace.
> The stack trace will not be written to the log file. It will be lost if we
> restart the daemon.
> 
> Maybe we need to replace spawn_n with spawn. If an exception is raised
> in the
> function, we can log it if needed. Any thoughts?
> 
> related bug: https://bugs.launchpad.net/neutron/+bug/1254984
> 
> [1] http://eventlet.net/doc/basic_usage.html

In most cases the use of spawn_n is intentional.  There certainly should
not be a mass search and replace of this.

In the case of the patch you submitted [1], I think the change makes
sense.  The code was already waiting for all operations to finish.
Adding some error handling seems like a good idea.

[1] https://review.openstack.org/#/c/58668/
-- 
Russell Bryant



More information about the OpenStack-dev mailing list