GitHub user xuang7 edited a comment on the discussion: Skipping release 1.2.1?

Thanks for starting this discussion. I agree that 1.3.0 should be the priority, 
and I'm not proposing to split effort between the two. I'd still like to keep 
1.2.1 as a low-cost fallback for a few reasons.

One concern is that new deployments of 1.2.0 are affected by the removal of the 
MinIO images, which 1.2.1 aims to address. Its scope is relatively small: since 
1.2.0, release/v1.2 has 50 commits touching 120 files, mostly covering the 
switch to RustFS and licensing fixes that 1.3.0 also needs. In comparison, 
release/v1.3 has 1,189 commits touching 2,047 files, including new services and 
modules (such as Jupyter and notebook migration), so its first RC and audit may 
take longer and could surface additional issues.

The extra work for 1.2.1 should be relatively limited, assuming no further 
major fixes are needed. Fixes land on main and are backported to both branches, 
so the remaining work on the 1.2 side would likely involve resolving a few 
backport conflicts and preparing another RC.

My suggestion is to focus on 1.3.0 for now and revisit the decision based on 
its progress. If 1.3.0 is ready for a vote within the next few weeks, we could 
skip 1.2.1 and add a note on the download page directing 1.2.0 users to 1.3.0, 
as @zyratlo suggested. If 1.3.0 takes longer than expected, we could consider 
proceeding with 1.2.1 RC2 to provide a working release in the meantime.

As the 1.2 release manager, I'm happy to follow the community's consensus on 
how best to proceed.

GitHub link: 
https://github.com/apache/texera/discussions/9305#discussioncomment-19405056

----
This is an automatically sent email for [email protected].
To unsubscribe, please send an email to: [email protected]

Reply via email to