Customer-specific fixes on 10.2 LTS only if every LTS customer wants them - otherwise it is a multi-year branch and breaks bug-for-bug compatibility with RH

November 3, 2026 at 5:13 PMtechnicalmedium

Situation

Ryan raised a customer-requested fix for 10.2 LTS that Melissa was advocating. Peter gave the test: if every 10.2 LTS customer wants the fix, do it; if not, it is another branch to maintain for years and very likely not worth it, and doing it means CIQ is no longer bug-for-bug compatible with Red Hat - we need to be clear about who we are. He framed it as thinking out loud and said he would support whatever Ryan and the other party agreed.

Reasoning

Universality is the test, not urgency: a fix every LTS customer wants is product, a fix one wants is a fork carried for years. Bug-for-bug compatibility is what LTS customers buy, so diverging for one account changes the promise for all. Peter supplied the criterion and left the verdict with Ryan.

Additional Context

Related to the 10/30 Rocky Linux Pro rename decision, which also put product identity ahead of a local request.

Observed Evidence

If every 10.2 LTS customer wants that fix, cool; if not, another branch to maintain for years, very likely not worth it; not bug for bug compatible with RH; be clear about who we are; happy to support if you both agree.

Matching Patterns

30%
Reclassify to Route - Decide Who Owns It, Then Give a Criterion Not a Verdict(criterion given, verdict left with Ryan)

Confidence Breakdown

32/35
Evidence
12/30
Pattern
16/20
Source
10/15
Corroboration

Reasoning Depth Analysis

Org Signal:LTS identity outranks individual customer asks
Who Affected:All 10.2 LTS customers; CS advocating for the requester
Precedent:Fixes that diverge from RH need universal demand
Consequences:Avoids a long-lived single-customer branch
Timing:LTS 10.2 GA approaching

Source

reflection

AI Confidence

70%

Related Context

💬
DM Peter / Ryan

slack

My question will be if every 10.2 LTS customer wants that fix. If so, cool. If not, it is another branch to maintain for years and very very likely not worth it at all.

Outcome

No outcome recorded yet.

Decision ID: 6fd5520f-3293-43cf-8be5-14732fdcf84b