Li Lu commented on YARN-3814:

bq. We have a JIRA for aggregation implementation for reader side. As we do not 
have parts of aggregation write path and we havent discussed timeline reader 
API for it, I think we can do it in the aggregation related JIRA. Because we 
were initially thinking of having different REST endpoints for different levels 
of aggregations. Thoughts ?
That's reasonable. Yes let's not fixing it here. However, my feeling is it will 
be way too many fragments if we address the aggregation REST APIs separately. 
Maybe we can add the RESTful endpoints when we add our aggregation readers? 
This will speed up the whole process of reviewing and committing. We don't 
really need a separate JIRA for each set of RESTful APIs since this part is 
relatively trivial. 

bq. The intention here is that if user queries for an entity which does not 
exist, the entity file would not exist. FileNotFoundException has been 
swallowed to return null which will then be converted to HTTP 404 in 
OK, but still we need some rationales in the code as comment for why we can 
swallow the exception. It will be also helpful if we output something about the 
404? I think it will be quite hard to debug if we hit an exception which cause 
a null return value and then a 404. 

> REST API implementation for getting raw entities in TimelineReader
> ------------------------------------------------------------------
>                 Key: YARN-3814
>                 URL: https://issues.apache.org/jira/browse/YARN-3814
>             Project: Hadoop YARN
>          Issue Type: Sub-task
>          Components: timelineserver
>    Affects Versions: YARN-2928
>            Reporter: Varun Saxena
>            Assignee: Varun Saxena
>         Attachments: YARN-3814-YARN-2928.01.patch, 
> YARN-3814-YARN-2928.02.patch, YARN-3814-YARN-2928.03.patch, 
> YARN-3814-YARN-2928.04.patch, YARN-3814.reference.patch

This message was sent by Atlassian JIRA

Reply via email to