HDDS-11753. Deprecate file per chunk layout from datanode code. #7654
Conversation
Change-Id: I2f1b168a5884abd5ad3e392dff3adcbc42edc6ee
Change-Id: I5cce02dd3dded77ca3bbeba12aac335043c38c7c
Change-Id: If8090f6dbdce3bb3e19ab6fa433b4ed011e6b951
|
This one supersedes #7585 |
| OzoneConfiguration.of(conf).set(ScmConfigKeys.OZONE_SCM_CONTAINER_LAYOUT_KEY, | ||
| "FILE_PER_BLOCK"); |
There was a problem hiding this comment.
If we are setting OZONE_SCM_CONTAINER_LAYOUT_KEY to FILE_PER_BLOCK in the code, we are actually removing support for FILE_PER_CHUNK not deprecating it.
With this change, there is no way to use FILE_PER_CHUNK anymore. Can we at least log a clear WARN message stating that the value of OZONE_SCM_CONTAINER_LAYOUT_KEY is getting overwritten in the code?
There was a problem hiding this comment.
This PR meant to disallow creating new FILE_PER_CHUNK containers, but it would still permit reading existing FILE_PER_CHUNK containers.
I'd love to get rid of FILE_PER_CHUNK code altogether but I'm not sure if anyone still have FILE_PER_CHUNK containers.
...tainer-service/src/main/java/org/apache/hadoop/ozone/container/keyvalue/KeyValueHandler.java
Outdated
Show resolved
Hide resolved
…ozone/container/keyvalue/KeyValueHandler.java Co-authored-by: Nandakumar Vadivelu <[email protected]>
Change-Id: I56ed5d2decb1354b24ffdf09e70ba5373df2e5ed
|
Merged. Thanks @nandakumar131 for the review. |
What changes were proposed in this pull request?
HDDS-11753. Deprecate file per chunk layout from datanode code.
Please describe your PR in detail:
What is the link to the Apache JIRA
https://issues.apache.org/jira/browse/HDDS-11753
How was this patch tested?
Unit test; one test case was added.