[
https://issues.apache.org/jira/browse/SPARK-2584?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14273762#comment-14273762
]
Ilya Ganelin edited comment on SPARK-2584 at 1/12/15 4:47 PM:
--------------------------------------------------------------
Hi Andrew, question about this. When you say "we drop it from memory" what
mechanism are you talking about? It's illegal to change the persistence level
of an already persisted RDD and if you call unpersist() it's dropped from both
memory and disk storage. How would an RDD be "dropped" from memory? I'm just
trying to reproduce the issue before creating a fix.
was (Author: ilganeli):
Hi Andrew, question about this. When you say "we drop it from memory" what
mechanism are you talking about? It's illegal to change the persistence level
of an already persisted RDD and if you call unpersist() it's dropped from both
memory and disk storage. How would an RDD be "dropped" from memory?
> Do not mutate block storage level on the UI
> -------------------------------------------
>
> Key: SPARK-2584
> URL: https://issues.apache.org/jira/browse/SPARK-2584
> Project: Spark
> Issue Type: Bug
> Components: Spark Core, Web UI
> Affects Versions: 1.0.1
> Reporter: Andrew Or
>
> If a block is stored MEMORY_AND_DISK and we drop it from memory, it becomes
> DISK_ONLY on the UI. We should preserve the original storage level proposed
> by the user, in addition to the change in actual storage level.
--
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
