Community
Participate
Working Groups
http://download.eclipse.org/eclipse/downloads/drops4/I20150513-2000/buildlogs/comparatorlogs/buildtimeComparatorUnanticipated.log.txt Appears launcher was updated ... but, not qualifier? 1. rt.equinox.framework/bundles/org.eclipse.equinox.launcher.gtk.linux.ppc64le eclipse_1605.so: present in baseline only eclipse_1612.so: not present in baseline
An important question, in this case, is if only ppc64le effected? Or, did something go wrong with "the whole launcher build"? Just wondering if teams can "sign off" on the rest of the build ... or, do we need a rebuild first?
(In reply to David Williams from comment #1) > An important question, in this case, is if only ppc64le effected? Yes, only ppc64le launcher was rebuilt, so no changes for other platforms. (In reply to David Williams from comment #0) > > Appears launcher was updated ... but, not qualifier? Qualifier has in fact been updated, not sure whats wrong...
(In reply to Arun Thondapu from comment #2) > > Appears launcher was updated ... but, not qualifier? > > Qualifier has in fact been updated, not sure whats wrong... I think I know whats wrong, the ppc64le fragment was not touched during the qualifier update, looks like that was never working previously too but we haven't really done too many builds for ppc64le until now. David, I've already pushed a fix, should we schedule a rebuild for this? It may not be too critical to schedule an immediate rebuild if we expect other rebuild requests to happen as it affects only ppc64le (and does not really break anything), but I leave the decision to you. Thanks!
(In reply to Arun Thondapu from comment #3) > David, I've already pushed a fix, should we schedule a rebuild for this? It > may not be too critical to schedule an immediate rebuild if we expect other > rebuild requests to happen as it affects only ppc64le (and does not really > break anything), but I leave the decision to you. Thanks! Arun and I chatted over IM, and we know of no pressing need to do the rebuild immediately, so I'll do a test build to confirm it fixes things, and if we need a rebuild for other reasons we'll pick up this fix too. Otherwise, we saw no harm in waiting until RC2 to pick up the change.
(In reply to David Williams from comment #4) > (In reply to Arun Thondapu from comment #3) > > > David, I've already pushed a fix, should we schedule a rebuild for this? It > > may not be too critical to schedule an immediate rebuild if we expect other > > rebuild requests to happen as it affects only ppc64le (and does not really > > break anything), but I leave the decision to you. Thanks! > > Arun and I chatted over IM, and we know of no pressing need to do the > rebuild immediately, so I'll do a test build to confirm it fixes things, and > if we need a rebuild for other reasons we'll pick up this fix too. > Otherwise, we saw no harm in waiting until RC2 to pick up the change. +1
The test build did not actually complete. It *seemed* unrelated to this change, but was a pretty vague error message about "maven had unmet dependencies" so might be a network glitch, or might be related to bug 467268. (I never know how far "up" a feature tree to touch features when we have to for Orbit bundles). It failed while assembling Eclipse SDK, so I touched that, and will try test build again.
(In reply to David Williams from comment #6) > The test build did not actually complete. It *seemed* unrelated to this > change, but was a pretty vague error message about "maven had unmet > dependencies" so might be a network glitch, or might be related to bug > 467268. (I never know how far "up" a feature tree to touch features when we > have to for Orbit bundles). > It failed while assembling Eclipse SDK, so I touched that, and will try test > build again. FYI, the next test build completed normally, and there were no "comparator errors".
(In reply to David Williams from comment #7) > > FYI, the next test build completed normally, and there were no "comparator > errors". Thanks for the confirmation David! I'll go ahead and resolve the bug, can you please indicate your positive review on the bug?
