[
https://issues.apache.org/jira/browse/HDFS-9405?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Xiao Chen updated HDFS-9405:
----------------------------
Attachment: HDFS-9405.05.patch
Thanks for explaining [~liuml07], I misread your previous comment. Agree the
extra {{spy()}} is not needed. Patch 5 reflects this.
Surprisingly, I discovered another problem in
{{TestEncryptionZonesWithKMS#testCreateEZPopulatesEDEKCache}} (the case above
my test case):
In HDFS-7209, we added this test, and method {{ValueQueue#getSize}} for testing
purpose. The problem is, when we do {{keyQueues.get(keyName).size()}} to verify
the size, the loading cache itself will always populate the cache because of
the call to {{get}}. In other words, the test always passes.... I've verified
that by backing up the change and running the test - it's still green.
So, patch 5 here also changed the {{getSize}} method to check size without
loading the cache. Verified both tests behave in a fail-before, pass-after
manner.
> When starting a file, NameNode should generate EDEK in a separate thread
> ------------------------------------------------------------------------
>
> Key: HDFS-9405
> URL: https://issues.apache.org/jira/browse/HDFS-9405
> Project: Hadoop HDFS
> Issue Type: Improvement
> Components: encryption, namenode
> Affects Versions: 2.7.1
> Reporter: Zhe Zhang
> Assignee: Xiao Chen
> Attachments: HDFS-9405.01.patch, HDFS-9405.02.patch,
> HDFS-9405.03.patch, HDFS-9405.04.patch, HDFS-9405.05.patch
>
>
> {{generateEncryptedDataEncryptionKey}} involves a non-trivial I/O operation
> to the key provider, which could be slow or cause timeout. It should be done
> as a separate thread so as to return a proper error message to the RPC caller.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
