For dropping the `hadoop-2` profile, we need to discuss it further after releasing Apache Spark 3.4 and monitoring the adoption of 3.4.
For now, we have no plan of dropping the `hadoop-2` profile yet because it's known to be used used in the community against the custom Hadoop 2 distributions (in the previous dev mailing thread) Dongjoon. On Sun, Nov 6, 2022 at 4:03 PM Yang,Jie(INF) <yangji...@baidu.com> wrote: > Hi, Bjørn > > > > Currently, I don't think we can consider upgrading Guava for the following > reasons: > > > > 1. Although Spark 3.4.0 will no longer release the hadoop-2 distribution, > but the build and testing process is still running. We need to keep it and > will not consider upgrading Guava until it is really removed > > > > 2. In my impression, Hive 2.3 still dependency on the Guava 14.0.1, > Someone has tried to solve this problem before, but it was not completed, > and this is another issue we need to solve before upgrading Guava > > > > > > YangJie > > > > > > *发件人**: *Bjørn Jørgensen <bjornjorgen...@gmail.com> > *日期**: *2022年11月6日 星期日 22:17 > *收件人**: *dev <dev@spark.apache.org> > *主题**: *Upgrade guava to 31.1-jre and remove hadoop2 > > > > Hi, anyone that has tried to upgrade guava now after we stop supporting > hadoop2? > > And is there a plan for removing hadoop2 code from the code base? > > >
