<slangasek> slashd: interesting that this is a behavior difference; we actually *would* want it to pull in openjdk-8-jre-headless by default instead of -9- in xenial, since 9 is in universe and 8 is in main. It is a behavior change that we should call out as part of the SRU (i.e. "regression potential"), but I don't consider it a blocker for the SRU.
<slangasek> The debdiff for xenial looks fine, the question is the test case. Did you verify already that having this package available in apt sources fixes the upgrade failure?
# Discussion with foundation team.
<slangasek> slashd: interesting that this is a behavior difference; we actually *would* want it to pull in openjdk- 8-jre-headless by default instead of -9- in xenial, since 9 is in universe and 8 is in main. It is a behavior change that we should call out as part of the SRU (i.e. "regression potential"), but I don't consider it a blocker for the SRU.
<slangasek> The debdiff for xenial looks fine, the question is the test case. Did you verify already that having this package available in apt sources fixes the upgrade failure?
<slangasek> then +1 from me