[
https://issues.apache.org/jira/browse/SPARK-7642?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Xiangrui Meng updated SPARK-7642:
---------------------------------
Attachment: 1.4 data distribution.png
1.3 data distribution.png
Attached the storage tab of a same RDD in 1.3 and 1.4.
> Missing 1 worker on standalone clusters.
> ----------------------------------------
>
> Key: SPARK-7642
> URL: https://issues.apache.org/jira/browse/SPARK-7642
> Project: Spark
> Issue Type: Bug
> Components: Deploy, Spark Core
> Affects Versions: 1.4.0
> Reporter: Xiangrui Meng
> Priority: Blocker
> Attachments: 1.3 data distribution.png, 1.4 data distribution.png
>
>
> Saw this weird issue during performance test. I have a 16-node (plus 1
> master) standalone cluster on EC2. I saw 16 works on the master page (:8080).
> When I run a job, in the executor tab I saw 17 executors (including the
> driver) in 1.3. However, this number becomes 16 (also including the driver)
> in 1.4. So one worker is missing from the cluster in 1.4.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
