Open Stack

On 8/16/16, 9:28 AM, "Mehdi Abaakouk" <sileht at sileht.net> wrote:

>Hi,
>
>Le 2016-08-15 04:50, Eric K a écrit :
>> Hi all, I'm running into an issue with oslo-messaging PRC call not
>> appearing to retry. If I do oslo_messaging.RPCClient(transport, target,
>> timeout=5, retry=10).call(self.context, method, **kwargs) using a topic
>> with no listeners, I consistently get the MessagingTimeout exception in
>> 5
>> seconds, with no apparent retry attempt. Any tips on whether this is a
>> user error or a bug or a feature? Thanks so much!
>
>About retry, from 
>http://docs.openstack.org/developer/oslo.messaging/rpcclient.html:
>
>"By default, cast() and call() will block until the message is
>successfully sent. However, the retry parameter can be used to have
>message sending fail with a MessageDeliveryFailure after the given
>number of retries."
>
>It looks like it retries in case of MessageDeliveryFailure not
>MessagingTimeout.
>
>Cheers,
>
>-- 
>Mehdi Abaakouk
>mail: sileht at sileht.net
>irc: sileht

Thank you, Mehdi! I¹ll assume that going forward.

>From the way the doc is worded and from the example that followed, it
sounded like setting the retry parameter changes the eventual exception
thrown after the retries are exhausted, not that it retries on that
exception.
"the retry parameter can be used to have message sending fail with a
MessageDeliveryFailure after the given number of retries."

I wonder what the true intended behavior is. Thanks again!

Cheers,

Eric