Hi, I recently worked on the 3.2.2 release and encountered a couple of issues during the process.
1. First issue - Resolved The first issue has been addressed via PR #1366 [1]. This was related to javadoc generation locale settings. 2. Second issue - ratis-bom version not updated by make_rc.sh While running make_rc.sh 1-prepare-src and 2-verify-bin, I noticed that the ratis-bom module's version was not being updated along with other modules. This is because ratis-bom does not inherit from the parent ratis POM. Looking at PR #1335 [2] that introduced the BOM module, I noticed that in the first commit, ratis-bom was actually set as a child module of ratis. But later it was changed to be independent (no parent). I'm not sure about the rationale behind this design decision. I could add a separate versions:set command in make_rc.sh to update ratis-bom independently, but I wanted to ask the community: - Why was ratis-bom designed not to inherit from the parent POM? - What is the recommended way to keep ratis-bom version in sync during releases? I'd appreciate any guidance on this matter. Thanks, Xinyu Tan [1] https://github.com/apache/ratis/pull/1366 [2] https://github.com/apache/ratis/pull/1335 On 2026/03/07 04:08:51 Xinyu Tan wrote: > Hi Tsz-Wo, > > I've submitted a PR[1] to fix the javadoc locale issue. > > Once this PR is merged, I can proceed with preparing the 3.2.2 RC1 release. > > [1] https://github.com/apache/ratis/pull/1365 > > Best > ------------- > Xinyu Tan > > On 2026/03/06 18:52:25 Tsz Wo Sze wrote: > > Hi Xinyu, > > > > > As usual, I am willing to act as release manager for this release! > > > > Thanks for serving as the RM again! > > > > BTW, the generated javadoc for 3.2.0 and 3.2.1 is in Chinese [1, 2]. We > > should use English as default for future releases. Of course, we may > > include a Chinese version of the javadoc. > > > > Tsz-Wo > > [1] https://javadoc.io/doc/org.apache.ratis/ratis-server/3.2.0/index.html > > [2] https://javadoc.io/doc/org.apache.ratis/ratis-server/3.2.1/index.html > > > > On Fri, Mar 6, 2026 at 10:40 AM Tsz Wo Sze <[email protected]> wrote: > > > > > Hi Attila, > > > > > > > Can we revert RATIS-2382 and RATIS-2379 on the release-3.2.2 branch? ... > > > > > > This is a good idea. RATIS-2382 and RATIS-2379 should not block the > > > release. > > > > > > > Just found a unit test failure (RATIS-2422) introduced recently. ... > > > > > > Thanks for fixing the failure! > > > > > > Sorry for replying late. > > > > > > Tsz-Wo > > > > > > On Fri, Mar 6, 2026 at 3:36 AM Attila Doroszlai <[email protected]> > > > wrote: > > > > > >> > I have created the branch release-3.2.2 [1] and reverted those two > > >> > commits. This can be used to prepare the release. > > >> > > >> Just found a unit test failure (RATIS-2422) introduced recently. The > > >> fix is simple, will need to cherry-pick to release-3.2.2. > > >> > > >> -Attila > > >> > > > > > >
