Sangjin Lee commented on YARN-3908:

bq. 2. We don't want to store the metric type, do we?

Maybe I was mistaken when I read your comment that said "I also realized that 
the metric type is not persisted too." I took it to mean that you're suggesting 
that we persist it. I also had an offline chat with [~vrushalic], and she 
clarified that we probably do not need to persist the metric type and that we 
can distinguish between a single value metric vs. time series as you described.

We still need to make some changes to ensure that the hbase writer sets the 
right min version when it writes a single-value metric (currently it's not 
being done).

> Bugs in HBaseTimelineWriterImpl
> -------------------------------
>                 Key: YARN-3908
>                 URL: https://issues.apache.org/jira/browse/YARN-3908
>             Project: Hadoop YARN
>          Issue Type: Sub-task
>          Components: timelineserver
>            Reporter: Zhijie Shen
>            Assignee: Vrushali C
>         Attachments: YARN-3908-YARN-2928.001.patch, 
> YARN-3908-YARN-2928.002.patch, YARN-3908-YARN-2928.003.patch
> 1. In HBaseTimelineWriterImpl, the info column family contains the basic 
> fields of a timeline entity plus events. However, entity#info map is not 
> stored at all.
> 2 event#timestamp is also not persisted.

This message was sent by Atlassian JIRA

Reply via email to