|
| 1 | +# ALESCo Meeting Minutes (2025-04-02) |
| 2 | +Minutes recorded by Jonathan Wright. |
| 3 | + |
| 4 | +Edited by Jonathan Wright for publishing. |
| 5 | + |
| 6 | +## Members |
| 7 | +### ALESCo Member Attendees |
| 8 | +- Andrew Lukoshko |
| 9 | +- Elkhan Mammadli |
| 10 | +- Jonathan Wright |
| 11 | +- Neal Gompa |
| 12 | +- Ben Thomas |
| 13 | + |
| 14 | +### Unable to attend |
| 15 | +- Cody Robertson |
| 16 | + |
| 17 | +### Board Attendees |
| 18 | +- benny Vasquez |
| 19 | + |
| 20 | +## Community Attendees |
| 21 | + |
| 22 | +## Decisions Adopted |
| 23 | + - Merged updated ALESCo Policy/Workflow for RFCs |
| 24 | + - https://github.com/AlmaLinux/ALESCo/pull/5 |
| 25 | + |
| 26 | +## Minutes |
| 27 | + |
| 28 | +### RFC: Enable KVM on AlmaLinux 9 on ppc64le |
| 29 | +https://github.com/AlmaLinux/ALESCo/pull/3 |
| 30 | +- Neal: what if we ship hyperscale kernel as an alternative kernel and turn it on there |
| 31 | + - Lance: That would work |
| 32 | + - Jonathan: That changes the RFC pretty significantly |
| 33 | +- Andrew: We can ship 2 AlmaLinux kernels, one with KVM/PPC enabled similar to something we do for ARM kernels already |
| 34 | +- Andrew/Neal: Userspace supports clean failure/fallback if kernel support is missing. |
| 35 | +- Neal: Default on PPC is PowerKVM anyway. |
| 36 | +- Jonathan: Hyperscale kernel is only supported 5 years |
| 37 | +- Neal: Hyperscale can look at supporting it for 10. |
| 38 | +- Jonathan: How many alt kernels is too many? |
| 39 | +- Jonathan to Lance: can you skip 9 and go to 10, where we're already shipping KVM/PPC support? |
| 40 | +- Lance: Don't know yet due to OpenStack. |
| 41 | +- Jonathan: We need Andrew's input since the workload of build and maintaining this will fall on his team. |
| 42 | +- Andrew: Will prepare POC of KVM/PPC on stock AlmaLinux kernel |
| 43 | +- Lance: Can test artifacts and provide feedback. |
| 44 | +- All: Will revisit after receiving Lance's feedback. |
| 45 | + |
| 46 | +### RFC: Introduce x86_64-v1 compatibility for AlmaLinux 10 |
| 47 | +https://github.com/AlmaLinux/ALESCo/pull/4 |
| 48 | +- benny: Until the work can be justified we can't do anything with this yet. |
| 49 | +- Jonathan: Premature to vote right now, proposer needs more time to reply to comments. |
| 50 | +- Neal: Sympathetic to his argument biggest problem is virtualization stack which straight up will not work on v1. |
| 51 | +- Jonathan: Even Fedora will probably move on from v1 in the next several years - conversations about this are coming up more and more. That would put us in a worse position supporting v1. |
| 52 | +- Neal: We're constrained by the software we ship, of which a lot is skewing towards requiring v2 and higher. |
| 53 | +- Jonathan: Tabling for vote at next meeting. |
| 54 | + |
| 55 | +### ALESCo Process Updates |
| 56 | +https://github.com/AlmaLinux/ALESCo/pull/5 |
| 57 | +This was heavily discussed in person at our leadership summit in Brussels, Belgium in February of 2025. |
| 58 | + |
| 59 | +- Neal: Let's add a chunk to make PRs as draft so they don't get accidentally merged. |
| 60 | +- benny: Haven't seen anything accidentally merged. |
| 61 | +- Neal: I almost did one this morning. |
| 62 | +- Ben: Does this stop alerts? |
| 63 | +- benny: No, alerts still go out. |
| 64 | +- Neal: Draft PRs don't trigger actions and stops from merging until it's not a draft. That's the only implications of a draft. |
| 65 | +- benny: I'm fine with drafts. |
| 66 | +- Neal: Define GH action to convert it to a draft? |
| 67 | +- Ben: What about a branch rule? |
| 68 | +- Neal: Complex to make rules like that. |
| 69 | +- All: Good with it, pending Jonathan's more detailed review. |
| 70 | +- benny: Make GH account using board email, add as reviewer to new PRs. |
0 commit comments