dongjoon-hyun commented on code in PR #45075:
URL: https://github.com/apache/spark/pull/45075#discussion_r1500902404
##########
docs/core-migration-guide.md:
##########
@@ -36,6 +36,8 @@ license: |
- Since Spark 4.0, Spark uses `ReadWriteOncePod` instead of `ReadWriteOnce`
access mode in persistence volume claims. To restore the legacy behavior, you
can set `spark.kubernetes.legacy.useReadWriteOnceAccessMode` to `true`.
+- Since Spark 4.0, Spark uses `~/.ivy2.5.2` as Ivy user directory by default
to isolate the existing systems from Apache Ivy's incompatibility. To restore
the legacy behavior, you can set `spark.jars.ivy` to `~/.ivy2`.
Review Comment:
I also thought like that. Something like `.ivy2.5.2+`.
After receiving your comment, I'm rethinking about that.
The bottom line is that the compatibility and release cycle depends on the
Apache Ivy community, not Apache Spark community.
- `.ivy2.5.2` literally means Apache Ivy format written by Apache Ivy 2.5.2 .
- If Apache Ivy 2.5.3 is not going to introduce any new change, it's still
Apache Ivy 2.5.2-format.
- If Apache Ivy 2.5.3 breaks the format again, we need to use `.ivy2.5.3`
at that time.
- In addition, if we use `ivy2.5.2_or_higher`, it could be an over-claim
because Apache Spark community is unable to guarantee any compatibility for
Apache Ivy 2.5.3 or higher which implies the naming.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
