[05:27] morning [06:04] morning [07:55] PR snapd#11869 opened: overlord/hookstate/ctlcmd: fix timestamp coming out of sync in unit tests [08:26] how much a snap package is reproductible ? [08:27] said otherwise, is it safe if I rebuild from scracth one package to be able to extract its debug symbol and those will match the same package built by somebody else? [08:28] $SNAP is not defined at build time? [08:36] PR snapd#11844 closed: secboot, boot: support and use alternative PCR handles during factory reset [08:45] lissyx: there are a few variables, which you'll need to double check: [08:46] (it looks like https://github.com/snapcore/action-build only allows for one output, the snap package) [08:46] (so we cant generate debug symbols and extract them ?) [08:47] 1) if sources for the parts in the snap.yaml are defined as pointing to moving branches (like "master") instead of tags or commits, you'll need to update the yaml file to point exactly at the commits that were the latest ones at the time the snap was built [08:48] 2) the OS on which the snap was built also matters, if the snap was built locally (and not inside lxd or launchpad) [08:49] the snap one is firefox, it refers to specific mercurial branches [08:49] so I assume 1 is not an issue [08:49] lissyx: you could embed the debug symbols in the snap itself [08:49] for 2 does multipass counts? [08:49] mardy, we still need to extract them for socorro and others [08:49] lissyx: if multipass is used, #2 should not be an issue either [09:11] PR snapd#11865 closed: secboot: stage and transition encryption keys [09:33] mardy, ok, so while not optimal, if we had ubuntu's snap being built on our infra on the same mozilla-central commits, then we could at least extract those [09:42] mardy, is there no way to produce more outputs from `snap run snapcraft` than the snap file, [09:46] PR snapd#11830 closed: interfaces: update network-control interface with permissions required by resolvectl [09:46] PR snapd#11857 closed: gadget/install: do not assume dm device has same block size as disk [09:55] lissyx: no i don't think that's possible, but it's probably better to ask in the forum [11:01] PR snapd#11870 opened: o/devicestate: factory reset with encryption [11:36] PR snapd#11871 opened: gadget/install: do not assume dm device has same block size as disk (2.56) [12:08] PR snapcraft#3788 opened: parts: correct spelling of build-base in yaml_utils.load [12:11] PR snapd#11869 closed: overlord/hookstate/ctlcmd: fix timestamp coming out of sync in unit tests [13:17] PR snapd#11872 opened: boot: post factory reset cleanup [13:32] PR snapd#11873 opened: spread.yaml: add ubuntu-22.04-06 to qemu-nested [14:22] PR snapd#11860 closed: tests: add spread execution for fedora 36