You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@drill.apache.org by "John Omernik (JIRA)" <ji...@apache.org> on 2017/05/02 17:52:04 UTC

[jira] [Commented] (DRILL-4258) Create New SYS tables: cpu, queries, fragments, threads, threadtraces, connections

    [ https://issues.apache.org/jira/browse/DRILL-4258?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15993390#comment-15993390 ] 

John Omernik commented on DRILL-4258:
-------------------------------------

Let me say this is a great idea. Specifically we were talking about the sys.functions one on the call today. There have been a number of times as a user and administrator of a Drill cluster where I only found out about a Drill functions via the user list, a code review, or a pull request.   That's not acceptable. If we had sys.functions, with both inputs to a functions, outputs, examples, and human readable descriptions, we'd be able to use this as a check box for new functions to be added. (Did you do your sys.functions documentation). 

In addition, it's a great resource for a analyst... instead of doing the things I mentioned above to find the functions, they could do "select * from sys.functions" or select * from sys.functions where function like '%SPLIT%' to get the information they need. All part of their analysis rather than as a separate interrupt based activity. 

+1 for this idea  (I which I had more than 1 + to add !)


> Create New SYS tables: cpu, queries, fragments, threads, threadtraces, connections
> ----------------------------------------------------------------------------------
>
>                 Key: DRILL-4258
>                 URL: https://issues.apache.org/jira/browse/DRILL-4258
>             Project: Apache Drill
>          Issue Type: New Feature
>          Components: Execution - Monitoring
>            Reporter: Jacques Nadeau
>
> cpu: Drillbit, # Cores, CPU consumption (with different windows?)
> queries: Foreman, QueryId, User, SQL, Start Time, rows processed, query plan, # nodes involved, number of running fragments, memory consumed
> fragments: Drillbit, queryid, major fragmentid, minorfragmentid, coordinate, memory usage, rows processed, start time
> threads: name, priority, state, id, thread-level cpu stats
> threadtraces: threads, stack trace
> connections: client, server, type, establishedDate, messagesSent, bytesSent



--
This message was sent by Atlassian JIRA
(v6.3.15#6346)