Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [eclipse-pmc] Regression in Luna SR1, due to "Platform Runtime Binary" regression, may be worthy of respin

For SR1, I think we should go with the rebuild. It looks good and worked fine in my testing.

I couldn't reproduce the gatekeeper issues. Looks like Apple caches some user responses there and keeps a hidden DB of earlier answers, so it's really hard to see what's going on. Trying to reproduce in a new install now.

For Mars M2, I don't think we need a rebuild. We should work on in master, and affected users can use I-builds.

Markus Keller




From:        Daniel Megert/Zurich/IBM@IBMCH
To:        eclipse-pmc@xxxxxxxxxxx
Cc:        Markus Keller/Zurich/IBM@IBMCH
Date:        2014-09-25 14:34
Subject:        Re: [eclipse-pmc] Regression in Luna SR1, due to "Platform Runtime Binary" regression, may be worthy of respin
Sent by:        eclipse-pmc-bounces@xxxxxxxxxxx




Not having a Mac and reading the description I'd say a rebuild is warranted, but for a final decision I want to hear what Markus has to say. He is familiar with Mac.

Dani




From:        
David M Williams <david_williams@xxxxxxxxxx>
To:        
eclipse-pmc@xxxxxxxxxxx
Date:        
25.09.2014 12:35
Subject:        
[eclipse-pmc] Regression in Luna SR1, due to "Platform Runtime Binary" regression, may be worthy of respin
Sent by:        
eclipse-pmc-bounces@xxxxxxxxxxx




Eclipse PMC,

There's been some late-breaking regressions found in EPP packages, which, at their root, are (partially) due to a regression in Eclipse Platform Runtime. (no regression with SDK, but even there, it will be effected by new Apple requirements for signing apps for 10.9.5).

This regression concerns Macs only and is related to the Info.plist file.

Plus, while investigating that, I've discovered there are new signing requirements for Mac OS X 10.9.5 and higher. While that OS version has only been out a week or two ... it has been no secret it's changing and we may look pretty poor if "Eclipse.app" suddenly stops working, after people upgrade to 10.9.5. (Though, would still work, from command line, which I suspect is why none of us noticed it, until now).

I have opened a "central" bug to discuss what options we have for Luna SR1 (
Bug 445064 ) as well as what to do for Mars M2, which has same issues (Bug 445062 ).

I have taken the liberty of starting a maintenance rebuild, which is nearly done building. And will soon start re-building the Sim. Release repo (where, only the Eclipse and Equinox contributions will change) ... but can always "throw them away" depending on if the PMC (or Planning Council) wants to take some other action. And/or "through them away" and do all over again, if the webmasters say they can fix the new signing requirements issue this morning. [The new signing requirements is a pretty sticky issue, since if I am reading it right, we should actually go "back in time" and re-sign existing, old releases as well, for them to work on 10.9.5! -- in other words, a lot of work -- for someone].

I'll be sending a note to Planning Council "in parallel", just for efficiency, but its my belief the "first decision" should come from the PMC.

Your advice welcome,

Thanks,

_______________________________________________
eclipse-pmc mailing list
eclipse-pmc@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit

https://dev.eclipse.org/mailman/listinfo/eclipse-pmc _______________________________________________
eclipse-pmc mailing list
eclipse-pmc@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
https://dev.eclipse.org/mailman/listinfo/eclipse-pmc

Back to the top