Make a custom release of sbt-osgi 0.9.4 + Invalid CEN extra data field patch about sbt-osgi

[eed3si9n]

I'm ok with whatever approach that unblocks you folks, but I guess a better approach would be to bisect to identify the regression and revert it.

[mdedetrich]

I'm ok with whatever approach that unblocks you folks, but I guess a better approach would be to bisect to identify the regression and revert it.

I would normally do this but because of the pressure that Pekko has to unbreak the branches and/or whether reverting that commit won't cause more problems (I assume it was done for a reason).

In any case I will go ahead and make the branch

[lefou]

• Probably related: #75

My hope was, that PR #79 might fix it. @mdedetrich could you please test this PR?

[lefou]

In case, #75 is related, an potential workaround is described here: akka/akka#31437 (comment)

[mdedetrich]

@lefou Sure, I just initiated a release of v0.9.4-INVALID-CEN-JAR-PATCH just to unblock pekko but I can look at those PR's to see if it fixes these issues.

[mdedetrich]

Closing as completed, can confirm that v0.9.4-INVALID-CEN-JAR-PATCH works and both branches of pekko are now unblocked. Will now continue to proceed to properly fix the regression.

[mdedetrich]

• Probably related: Race condition in the dependencies of the osgiBundle task #75

My hope was, that PR #79 might fix it. @mdedetrich could you please test this PR?

@lefou I can confirm that this didn't fix the underlying issue, at least with Pekko

[lefou]

• Probably related: Race condition in the dependencies of the osgiBundle task #75

My hope was, that PR #79 might fix it. @mdedetrich could you please test this PR?

@lefou I can confirm that this didn't fix the underlying issue, at least with Pekko

Too bad. Thanks for checking.