I am using webHCat to submit very simple queries via the /hive endpoints.
Then I used the /jobs endpoint to query if the job has been completed or
not (not via a callback from the hive endpoint).
I am noticing that this loop takes a little over 1 minutes,
Is there someway to tune this so that the
OK, more updates. Today I was running the query with Yarn and also turned
on DEBUG logging. Here's what I found from the task log for the dangling
task.
I noticed that after the RowContainer has been created by Hive, there are a
lot of IPC/RPC related logs (still printing), every 3 second apart. M
Moreover, the IP 10.224.174.71 is a different node, not the one executing
the reduce task. Why did that happen?
On Fri, Oct 20, 2017 at 3:37 PM, Daniel Bruce wrote:
> OK, more updates. Today I was running the query with Yarn and also turned
> on DEBUG logging. Here's what I found from the task l
