16166 2026-02-09 12:10:02 +0000 AB-INT: mdadm reproducibility issues 2026-04-02 15:02:59 +0000 1 1 10 QA/Testing Functional (self) Testing oe-selftest unspecified x86 Multiple RESOLVED WORKSFORME AB-INT Medium normal 6.0 1 mathieu.dubois-briand ankur.tyagi85 randy.macleod ross.burton yoann.congal --- --- No (bug/feature does not impact docs) oldest_to_newest 104200 0 mathieu.dubois-briand 2026-02-09 12:10:02 +0000 AssertionError: The following deb packages are different and not in exclusion list: /srv/pokybuild/yocto-worker/reproducible/build/build-st/reproducibleB-extended/tmp/deploy/deb/./x86-64-v3/mdadm_4.5-r0_amd64.deb The following ipk packages are different and not in exclusion list: /srv/pokybuild/yocto-worker/reproducible/build/build-st/reproducibleB-extended/tmp/deploy/ipk/./x86-64-v3/mdadm_4.5-r0_x86-64-v3.ipk The following rpm packages are different and not in exclusion list: /srv/pokybuild/yocto-worker/reproducible/build/build-st/reproducibleB-extended/tmp/deploy/rpm/./x86_64_v3/mdadm-4.5-r0.x86_64_v3.rpm 104201 1 mathieu.dubois-briand 2026-02-09 12:13:49 +0000 Sorry, clicked too quickly on the send button... So we can see some path in mdcheck script is sometimes "BINDIR/mdadm", sometimes "/sbin/mdadm". This is probably the source issue, which then change data all around package metadata. https://valkyrie.yocto.io/pub/repro-fail/oe-reproducible-20260207-1x39j7rf/packages/diff-html/ 104202 2 mathieu.dubois-briand 2026-02-09 12:14:28 +0000 reproducible debian12-vk-1 master completed at 2026-02-07 02:39:58+00:00 https://valkyrie.yocto.io/pub/repro-fail/oe-reproducible-20260207-1x39j7rf https://autobuilder.yoctoproject.org/valkyrie/#/builders/37/builds/3310/steps/13/logs/stdio reproducible fedora41-vk-1 master completed at 2026-02-08 02:28:01+00:00 https://valkyrie.yocto.io/pub/repro-fail/oe-reproducible-20260208-00rfa3je https://autobuilder.yoctoproject.org/valkyrie/#/builders/37/builds/3313/steps/13/logs/stdio reproducible opensuse156-vk-1 master completed at 2026-02-09 02:31:23+00:00 https://valkyrie.yocto.io/pub/repro-fail/oe-reproducible-20260209-a0dgcw5d https://autobuilder.yoctoproject.org/valkyrie/#/builders/37/builds/3317/steps/14/logs/stdio 104208 3 mathieu.dubois-briand 2026-02-10 10:05:49 +0000 Revert has been sent by Alex: https://lists.openembedded.org/g/openembedded-core/message/230798 104209 4 mathieu.dubois-briand 2026-02-10 10:19:50 +0000 reproducible ubuntu2204-vk-1 master completed at 2026-02-10 02:45:10+00:00 https://valkyrie.yocto.io/pub/repro-fail/oe-reproducible-20260210-mvxmxr1o https://autobuilder.yoctoproject.org/valkyrie/#/builders/37/builds/3320/steps/13/logs/stdio 104232 5 mathieu.dubois-briand 2026-02-11 15:45:12 +0000 reproducible ubuntu2204-vk-4 master completed at 2026-02-11 02:25:44+00:00 https://valkyrie.yocto.io/pub/repro-fail/oe-reproducible-20260211-7w04g33o https://autobuilder.yoctoproject.org/valkyrie/#/builders/37/builds/3328/steps/13/logs/stdio 104287 6 ankur.tyagi85 2026-02-13 01:39:15 +0000 I would also like to rule out a possible race condition during install. Will add check to fail build if sed substitution still fails when installing files serially. 104568 7 ross.burton 2026-03-09 14:12:38 +0000 I glared at this recently and couldn't see how this can happen. The sed happens in a single target that doesn't get called repeatedly. I ran the install in a loop with lots of parallel and make shuffling enabled, and couldn't replicate it. 104935 8 randy.macleod 2026-04-02 15:02:59 +0000 We've upgraded to 4.6 and haven't seen this issue so far.