[openstack-dev] [Monasca] Selectively publish logs to topics

Sundaram, Venkat (Venkat Sundaram (TSV)) ts.venkat at hpe.com
Wed Jun 1 15:25:56 UTC 2016


Thanks, I like the way you break it into two topics. I agree with you on the first one to make it similar to dimensions (global and per message).
On the second one, I see your point about including the topic as part of the resource URI. I am ok with it if others don’t see any issues there, just that it would be more significant (new API endpoint).

Yes, you are right about the retention part, that was mean’t only as an example. I removed it from the blueprint now.


On 5/31/16, 7:08 AM, "Witek Bedyk" <witold.bedyk at est.fujitsu.com> wrote:

>Hi Venkat,
>thank you for submitting the blueprint [1]. It covers actually two 
>topics, both of them a valuable functional extension:
>1) submitting additional (apart from dimensions) information with the logs
>2) specifying a specific output topic
>ad. 1
>I think we should keep it generic to allow the operator add any 
>information one needs. I like the idea of adding the 'attributes' 
>dictionary, but we would need it per message, not only per request (the 
>same story as we had with global and local dimensions).
>ad. 2
>As we want to change the target where the API writes the data, we could 
>use perhaps the path parameter for that. The request could look like:
>POST /v3.0/logs/topics/{kafka_topic_name}
>I don't think we should send 'retention' with every request, instead the 
>Kafka topic should be configured accordingly, but I understand it was 
>just an example, right?
>FUJITSU Enabling Software Technology GmbH
>Schwanthalerstr. 75a, 80336 München
>Telefon: +49 89 360908-547
>Telefax: +49 89 360908-8547
>COINS: 7941-6547
>Sitz der Gesellschaft: München
>AG München, HRB 143325
>Geschäftsführer: Dr. Yuji Takada, Hans-Dieter Gatzka, Christian Menk

More information about the OpenStack-dev mailing list