#fedora-meeting: F39 Beta Go/No-Go meeting

Meeting summary

1. Roll Call (adamw, 17:00:55)
   1. nirik will represent release engineering (adamw, 17:05:34)
   2. adamw will represent QA (adamw, 17:05:39)
   3. sgallagh, decathorpe and/or dcantrell will represent development (buy your tickets now for the last-man-standing battle) (adamw, 17:06:18)
   
   
2. Purpose of this meeting (adamw, 17:07:24)
   1. Purpose of this meeting is to check whether or not F39 Beta is ready for shipment, according to the release criteria. (adamw, 17:07:24)
   2. This is determined in a few ways: (adamw, 17:07:24)
   3. 1. No remaining blocker bugs (adamw, 17:07:24)
   4. 2. Release candidate compose is available (adamw, 17:07:24)
   5. 3. Test matrices are fully completed (adamw, 17:07:26)
   6. 4. Fedora CoreOS and Fedora IoT are prepared (adamw, 17:07:28)
   
   
3. Current status - blockers (adamw, 17:07:59)
   1. https://qa.fedoraproject.org/blockerbugs/milestone/39/beta/buglist (adamw, 17:08:00)
   
   
4. (2232838) Some NVME controllers fail to initialize with kernel 6.4.11 or later (nvme0: controller is down; will reset: CSTS:0xffffffff, PCI_STATUS=0xffff) (adamw, 17:08:27)
   1. https://bugzilla.redhat.com/show_bug.cgi?id=2232838 (adamw, 17:08:27)
   2. https://pagure.io/fedora-qa/blocker-review/issue/1291 (adamw, 17:08:27)
   3. Accepted Blocker, kernel, ON_QA (adamw, 17:08:27)
   4. https://bodhi.fedoraproject.org/updates/FEDORA-2023-ac464517f7#comment-3195672 (adamw, 17:09:49)
   5. https://bugzilla.redhat.com/show_bug.cgi?id=2232838#c23 (adamw, 17:11:04)
   6. https://kojipkgs.fedoraproject.org/compose/branched/Fedora-39-20230911.n.0/ (adamw, 17:18:34)
   7. https://kojipkgs.fedoraproject.org/compose/branched/Fedora-39-20230910.n.0/ then (adamw, 17:18:52)
   8. folks are testing to see if they can reproduce this and verify the fix at present, we will circle back (adamw, 17:20:56)
   
   
5. (2113005) Live image made with BOOTX64.EFI from latest shim-x64-15.6-2 fails to boot on some boards (adamw, 17:21:11)
   1. https://bugzilla.redhat.com/show_bug.cgi?id=2113005 (adamw, 17:21:11)
   2. https://pagure.io/fedora-qa/blocker-review/issue/1154 (adamw, 17:21:11)
   3. Accepted Blocker, shim, NEW (adamw, 17:21:11)
   4. this is still stuck in the same pattern it's been in for several releases: we have a fix, but we cannot get a new shim build signed by microsoft until the kernel supports nx, and those patches are still being reviewed (adamw, 17:22:39)
   5. AGREED: 2113005 - waived to Fedora 39 Final - this blocker is waived to Fedora 39 Final as a "Difficult to fix blocker bug": we cannot practically fix it at the moment while the nx patches are still under discussion (adamw, 17:27:06)
   
   
6. Current status - test matrices (adamw, 17:28:01)
   1. https://fedoraproject.org/wiki/Category:Fedora_39_Test_Results (adamw, 17:28:01)
   2. test matrix coverage is good except for aarch64 desktop testing, sgallagh is going to try that out quickly and report back any issues he sees (adamw, 17:45:14)
   
   
7. Current status - RC (adamw, 17:46:11)
   1. we have a candidate compose, https://dl.fedoraproject.org/pub/alt/stage/39_Beta-1.1/ , nobody has said there's anything wildly wrong with it AFAIK (adamw, 17:46:48)
   
   
8. Fedora CoreOS and IoT check-in (adamw, 17:53:43)
   1. Fedora CoreOS is ready to go according to dustymabe, the plan is to update the "next" stream to Fedora 39 on the same day as the Beta release (adamw, 17:56:28)
   2. IoT matrix - https://fedoraproject.org/wiki/Test_Results:Fedora-IoT_39_RC_20230912.0_General (adamw, 17:57:14)
   3. we will nominate an IoT compose for release after the packages from the 1.1 candidate are pushed stable, but don't anticipate any problems (adamw, 17:57:39)
   
   
9. (2232838) Some NVME controllers fail to initialize with kernel 6.4.11 or later (nvme0: controller is down; will reset: CSTS:0xffffffff, PCI_STATUS=0xffff) (adamw, 18:05:17)
   1. AGREED: 2232838 - treat as VERIFIED - it's a shame that the fix for this has not been formally verified by someone affected, but we have checked that the commit identified upstream as the culprit has indeed been reverted, so we think it's reasonable to consider this addressed (adamw, 18:15:27)
   
   
10. Go/No-Go decision (adamw, 18:16:18)
    1. AGREED: Fedora Linux 39 Beta is GO (adamw, 18:18:18)
    2. Fedora Linux 39 Beta will release on the target date #1 (2023-09-19) (adamw, 18:19:06)
    3. ACTION: adamw or amoloney to announce decision (adamw, 18:19:49)
    
    
11. Open floor (adamw, 18:20:02)

Action items

1. adamw or amoloney to announce decision

Action items, by person

1. adamw
   1. adamw or amoloney to announce decision

People present (lines said)

1. adamw (154)
2. Son_Goku (45)
3. geraldosimiao (29)
4. zodbot (23)
5. michel-slm (23)
6. sgallagh (22)
7. decathorpe (18)
8. nirik (13)
9. thebeanogamer (13)
10. dcantrell (12)
11. dustymabe (7)
12. aleasto (7)
13. lruzicka (6)
14. frantisekz (5)
15. Southern_Gentlem (5)
16. coremodule (4)
17. dtometzki (4)
18. pjones (2)
19. zlopez (2)
20. tflink (1)
21. universal-blue-f (1)