| Summary: | something is amiss with org.eclipse.equinox.executable.feature.group | ||
|---|---|---|---|
| Product: | [Eclipse Project] Platform | Reporter: | David Williams <david_williams> |
| Component: | Releng | Assignee: | Platform-Releng-Inbox <platform-releng-inbox> |
| Status: | RESOLVED FIXED | QA Contact: | |
| Severity: | blocker | ||
| Priority: | P3 | CC: | kim.moir, remy.suen, thomas |
| Version: | 3.7 | ||
| Target Milestone: | --- | ||
| Hardware: | PC | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Bug Depends on: | 337334 | ||
| Bug Blocks: | |||
|
Description
David Williams
I think this is due to bug 337334 We don't build all the launcher fragments but Andrew still lists them in the executable feature. Can you exclude this fragment somehow or skip it? I thought there was an option to do that. As an aside, I think this bug will go away when I switch my Indigo contribution back to point to the milestones repo. That fragment still exists in the repo. It wasn't removed until 3.7M4. Yes, does sound like an implication of bug 337334, so I'll list is as "depends on". I am raising the severity to "blocker", though, since we can't aggregate the common repository until this is resolved. There are ways to disable specific features, if they are listed as contributions in the aggregation file; but that would mean disabling 'org.eclipse.equinox.sdk' and suspect that's not desired? It might work if we list an "old version" in an "old repository" but I'll need to re-read bug 337334 a few times to know if that's a "fix" or a temporary workaround. (In reply to comment #0) > We do not have any "motif.linux.x86" configured in Indigo.b3aggr file (just aix > and hpux): I beg to differ. I'm looking at indigo.b3aggr version 1.59. It contains this entry: <configurations operatingSystem="linux" windowSystem="motif"/> (the x86 is default and hence, not explicitly stated) (In reply to comment #3) > (In reply to comment #0) > > We do not have any "motif.linux.x86" configured in Indigo.b3aggr file (just aix > > and hpux): > > I beg to differ. I'm looking at indigo.b3aggr version 1.59. It contains this > entry: > > <configurations operatingSystem="linux" windowSystem="motif"/> > > (the x86 is default and hence, not explicitly stated) Oh ... thanks! I'd missed that. So, the "fix" here might be pretty simple .. just remove that configuration since "no longer" valid? (In reply to comment #4) > So, the "fix" here might be pretty simple .. just remove that configuration > since "no longer" valid? I think the rule of thumb should be that the release train should support the same configurations as the Eclipse IDE. Working "locally" I've found I can remove 4 configurations, and the local build verifies correctly. I've check in that change but expect it to get past the verification phase, so will close this as fixed. <configurations operatingSystem="hpux" windowSystem="motif" architecture="ia64_32"/> <configurations operatingSystem="aix" windowSystem="motif" architecture="ppc"/> <configurations operatingSystem="linux" windowSystem="gtk" architecture="ppc"/> <configurations operatingSystem="linux" windowSystem="motif"/> (In reply to comment #5) > I think the rule of thumb should be that the release train should support the > same configurations as the Eclipse IDE. Hmm, ok ... and what are those? I suggest the Platform team take "ownership" of the issue, and they add back configurations they support, implying, of course, they provide the fragments in a repository. David, Yes, you can remove those platforms. See bug 317309 for the list of platforms we removed in 3.7 |