Closed
Bug 653971
Opened 14 years ago
Closed 14 years ago
Mac XULRunner packaging broken by bug 386760
Categories
(Firefox Build System :: General, defect)
Tracking
(firefox5+ fixed, firefox6 fixed)
RESOLVED
FIXED
mozilla5
People
(Reporter: philor, Assigned: robert.strong.bugs)
References
Details
(Keywords: regression)
Attachments
(1 file, 1 obsolete file)
601 bytes,
patch
|
nthomas
:
review+
khuey
:
review+
asa
:
approval-mozilla-aurora+
asa
:
approval-mozilla-beta+
|
Details | Diff | Splinter Review |
http://tinderbox.mozilla.org/showlog.cgi?log=XULRunner/1304157982.1304168790.8784.gz Compressing... cd ../../dist && (cd universal/xulrunner/xulrunner-6.0a1.en-US.mac-pkg.pkg/Contents/MacOS && /tools/buildbot/bin/python2.6 /builds/slave/cen-osx64-xr/build/config/createprecomplete.py) && /builds/slave/cen-osx64-xr/build/build/package/mac_osx/pkg-dmg --source "universal/xulrunner-pkg" --target "xulrunner-6.0a1.en-US.mac-pkg.dmg" --volname "XULRunner" /bin/sh: line 0: cd: universal/xulrunner/xulrunner-6.0a1.en-US.mac-pkg.pkg/Contents/MacOS: No such file or directory http://tinderbox.mozilla.org/showlog.cgi?log=XULRunner/1302516325.1302527024.4933.gz&fulltext=1 (the last one before bug 386760 landed) Compressing... cd ../../dist && /builds/slave/cen-osx64-xr/build/build/package/mac_osx/pkg-dmg --source "universal/xulrunner-pkg" --target "xulrunner-2.2a1pre.en-US.mac-pkg.dmg" --volname "XULRunner" mktemp -d pkg-dmg.71204.XXXXXXXX etc.
Comment 1•14 years ago
|
||
Can we prioritize this bug... It is preventing us as addon developers to be ready for the Firefox 5 launch on the Mac environment.
Assignee | ||
Comment 2•14 years ago
|
||
Nick, I'm going to take a look at what's going on here but I wondered if you might already know since I'm unfamiliar with Mac XULRunner?
Assignee | ||
Comment 3•14 years ago
|
||
Now that omnijar is on by default this is failing earlier on when trying to optimize the jar files. :(
Comment 4•14 years ago
|
||
That would be bug 659718, right?
Assignee | ||
Comment 5•14 years ago
|
||
I don't think so... I'm just compiling xulrunner and I suspect that fixes a similar issue when compiling xullrunner + app.
Assignee | ||
Comment 6•14 years ago
|
||
Errors with Mac xulrunner universal latest trunk Creating package directory... /opt/local/bin/python2.6 /moz/_1_mozilla-central/mozilla/config/optimizejars.py --optimize /moz/_1_mozilla-central/xr-universal/i386/xulrunner/installer/../../jarlog/ ../../dist/bin/chrome ../../dist/universal/xulrunner/xulrunner-7.0a1.en-US.mac-pkg.pkg/Contents/MacOS/chrome signing nss libraries moduleSpec configdir='' certPrefix='' keyPrefix='' secmod='' flags=noCertDB, noModDB Generate a DSA key pair ... ../../dist/universal/xulrunner/xulrunner-7.0a1.en-US.mac-pkg.pkg/Contents/MacOS/libsoftokn3.dylib: -5950: File not found moduleSpec configdir='' certPrefix='' keyPrefix='' secmod='' flags=noCertDB, noModDB Generate a DSA key pair ... ../../dist/universal/xulrunner/xulrunner-7.0a1.en-US.mac-pkg.pkg/Contents/MacOS/libnssdbm3.dylib: -5950: File not found Removing unpackaged files... cd ../../dist/universal/xulrunner/xulrunner-7.0a1.en-US.mac-pkg.pkg/Contents/MacOS; rm -rf xulrunner-config regchrome* regxpcom* core bsdecho gtscc js js-config jscpucfg nsinstall viewer TestGtkEmbed codesighs* elf-dynstr-gc mangle* maptsv* mfc* mkdepend* msdump* msmap* nm2tsv* nsinstall* res/samples res/throbber shlibsign* ssltunnel* certutil* pk12util* winEmbed.exe chrome/chrome.rdf chrome/app-chrome.manifest chrome/overlayinfo components/compreg.dat components/xpti.dat content_unit_tests necko_unit_tests *.dSYM /moz/_1_mozilla-central/xr-universal/i386/config/nsinstall -D ../../dist/ Compressing... cd ../../dist && (cd universal/xulrunner/xulrunner-7.0a1.en-US.mac-pkg.pkg/Contents/MacOS && rm -f omni.jar components/binary.manifest && grep -h '^binary-component' components/*.manifest > binary.manifest ; sed -e 's/^binary-component/#binary-component/' components/components.manifest > components.manifest && mv components.manifest components && zip -r9m omni.jar chrome chrome.manifest components/*.js components/*.xpt components/*.manifest modules res defaults greprefs.js jsloader -x chrome/icons/\* defaults/pref/channel-prefs.js res/cursors/\* res/MainMenu.nib/\* && true && /opt/local/bin/python2.6 /moz/_1_mozilla-central/mozilla/config/optimizejars.py --optimize /moz/_1_mozilla-central/xr-universal/i386/xulrunner/installer/../../jarlog/ ./ ./ && mv binary.manifest components && printf "manifest components/binary.manifest\n" > chrome.manifest) && (cd universal/xulrunner/xulrunner-7.0a1.en-US.mac-pkg.pkg/Contents/MacOS && /opt/local/bin/python2.6 /moz/_1_mozilla-central/mozilla/config/createprecomplete.py) && /moz/_1_mozilla-central/mozilla/build/package/mac_osx/pkg-dmg --source "universal/xulrunner-pkg" --target "xulrunner-7.0a1.en-US.mac-pkg.dmg" --volname "XULRunner" grep: components/*.manifest: No such file or directory sed: components/components.manifest: No such file or directory make[2]: *** [make-package] Error 1 make[1]: *** [default] Error 2 make: *** [package] Error 2
tracking-firefox5:
--- → ?
Assignee | ||
Comment 8•14 years ago
|
||
This fixes building with omnijar which I believe is the default now and the generation of the precomplete file. Still need to check building with omnijar disabled but I'd like to get your take on this since I'm not very familiar with universal builds.
Assignee: nobody → robert.bugzilla
Status: NEW → ASSIGNED
Attachment #536993 -
Flags: review?(ted.mielczarek)
Assignee | ||
Comment 9•14 years ago
|
||
Comment on attachment 536993 [details] [diff] [review] patch rev1 I think that the changes to config/createprecomplete.py aren't needed though I am not positive.
Assignee | ||
Comment 10•14 years ago
|
||
It also appears that omnijar being turned on by default broke the generation of the binary.manifest from the components.manifest sed: components/components.manifest: No such file or directory cc'ing glandium Mike, the Mac xulrunner nightly builds are broken in part due to bug 620931. I believe the attached patch minus the changes to createprecomplete.py are all that is needed to fix this bug for both MOZ_CHROME_FILE_FORMAT=omni and MOZ_CHROME_FILE_FORMAT=jar builds but I haven't looked into why it is having problem with the components.manifest with just MOZ_CHROME_FILE_FORMAT=omni.
Assignee | ||
Comment 11•14 years ago
|
||
Attachment #536993 -
Attachment is obsolete: true
Attachment #536993 -
Flags: review?(ted.mielczarek)
Attachment #537053 -
Flags: review?(ted.mielczarek)
Assignee | ||
Comment 12•14 years ago
|
||
Comment on attachment 537053 [details] [diff] [review] patch rev2 Nick, is there any way you can check to see if this fixes the xulrunner build and mar generation.
Attachment #537053 -
Flags: review?(nrthomas)
Comment 13•14 years ago
|
||
(In reply to comment #10) > Mike, the Mac xulrunner nightly builds are broken in part due to bug 620931. > I believe the attached patch minus the changes to createprecomplete.py are > all that is needed to fix this bug for both MOZ_CHROME_FILE_FORMAT=omni and > MOZ_CHROME_FILE_FORMAT=jar builds but I haven't looked into why it is having > problem with the components.manifest with just MOZ_CHROME_FILE_FORMAT=omni. Probably due to how PACK_OMNIJAR is defined in toolkit/mozapps/installer/packager.mk (it should be using $(ZIP) and $(UNZIP), btw)
Comment 14•14 years ago
|
||
I can certainly try a xulrunner build to get a dmg + SDKs with this patch, but I'm not sure what you mean about mar generation. We don't create those for XR.
Assignee | ||
Comment 15•14 years ago
|
||
(In reply to comment #14) > I can certainly try a xulrunner build to get a dmg + SDKs with this patch, > but I'm not sure what you mean about mar generation. We don't create those > for XR. Thanks Nick... since mars aren't generated I'm tempted not to create the precomplete file. What do you think?
Assignee | ||
Comment 16•14 years ago
|
||
btw: no need to try out the patch since mars aren't generated.
Assignee | ||
Comment 17•14 years ago
|
||
I've verified that both MOZ_CHROME_FILE_FORMAT=jar and MOZ_CHROME_FILE_FORMAT=omni build and package correctly with the patch though the second time packaging with MOZ_CHROME_FILE_FORMAT=omni fails. This is due to it modifying the files in dist/universal/xulrunner/XUL.framwork. I'll take a look at what we do for app omnijar packaging to see if there is an obvious fix.
Comment 18•14 years ago
|
||
Comment on attachment 537053 [details] [diff] [review] patch rev2 I took this patch for a spin in a staging instance of the buildbot, on top of m-c rev a52b04dc29ea. It completed make package, and the following step 'make -f client.mk sdk', without errors. It does expose a problem with the buildbot setup which is filed as bug 661792, FWIW, I think we're also hitting the manifest error sed: can't read components/components.manifest: No such file or directory on windows m-c xulrunner, but not on linux.
Attachment #537053 -
Flags: review?(nrthomas) → review+
Comment 19•14 years ago
|
||
(In reply to comment #17) > I've verified that both MOZ_CHROME_FILE_FORMAT=jar and > MOZ_CHROME_FILE_FORMAT=omni build and package correctly with the patch > though the second time packaging with MOZ_CHROME_FILE_FORMAT=omni fails. > This is due to it modifying the files in > dist/universal/xulrunner/XUL.framwork. I'll take a look at what we do for > app omnijar packaging to see if there is an obvious fix. Bug 600832 has a work around for this - making the postflight_all target again before calling package.
Assignee | ||
Comment 20•14 years ago
|
||
Thanks a lot for doing all of this Nick!
Assignee | ||
Updated•14 years ago
|
Attachment #537053 -
Flags: review?(khuey)
Comment on attachment 537053 [details] [diff] [review] patch rev2 Hardcoding it might be a little ugly but we need to fix this for 5 and this is good enough IMO.
Attachment #537053 -
Flags: review?(khuey) → review+
Assignee | ||
Comment 22•14 years ago
|
||
Pushed to mozilla-central http://hg.mozilla.org/mozilla-central/rev/6ee7ab2989ba note: it is hardcoded here as well http://mxr.mozilla.org/mozilla-central/source/xulrunner/installer/mac/Makefile.in#67 I also pushed to try to check if this would have any adverse affect on mobile packaging (didn't think it would... just wanted to be safe) http://tbpl.mozilla.org/?tree=Try&rev=2b0f7c6aaf11
Status: ASSIGNED → RESOLVED
Closed: 14 years ago
Resolution: --- → FIXED
Assignee | ||
Comment 23•14 years ago
|
||
Not related to this bug but bug 661910 also has a problem with reading the components.manifest except it is on windows where xulrunner doesn't have any binary components.
Assignee | ||
Comment 24•14 years ago
|
||
Comment on attachment 537053 [details] [diff] [review] patch rev2 Drivers, this is a xulrunner only one line change and is needed so the sdk is built.
Attachment #537053 -
Flags: review?(ted.mielczarek)
Attachment #537053 -
Flags: approval-mozilla-beta?
Attachment #537053 -
Flags: approval-mozilla-aurora?
Comment 25•14 years ago
|
||
Comment on attachment 537053 [details] [diff] [review] patch rev2 Yes, we want this for [5] Beta and [6] Aurora.
Attachment #537053 -
Flags: approval-mozilla-beta?
Attachment #537053 -
Flags: approval-mozilla-beta+
Attachment #537053 -
Flags: approval-mozilla-aurora?
Attachment #537053 -
Flags: approval-mozilla-aurora+
Updated•14 years ago
|
status-firefox5:
--- → affected
status-firefox6:
--- → affected
Assignee | ||
Comment 26•14 years ago
|
||
Pushed to mozilla-beta http://hg.mozilla.org/releases/mozilla-beta/rev/9171cee0df6e Pushed to mozilla-aurora http://hg.mozilla.org/releases/mozilla-aurora/rev/39a95a10faf0
Assignee | ||
Comment 27•14 years ago
|
||
Before the patch landed cd ../../dist && (cd universal/xulrunner/xulrunner-6.0a2.en-US.mac-pkg.pkg/Contents/MacOS && rm -f omni.jar components/binary.manifest && grep -h '^binary-component' components/*.manifest > binary.manifest ; sed -e 's/^binary-component/#binary-component/' components/components.manifest > components.manifest && mv components.manifest components && zip -r9m omni.jar chrome chrome.manifest components/*.js components/*.xpt components/*.manifest modules res defaults greprefs.js jsloader -x chrome/icons/\* defaults/pref/channel-prefs.js res/cursors/\* res/MainMenu.nib/\* && true && /tools/buildbot/bin/python2.6 /builds/slave/aurora-osx64-xr/build/config/optimizejars.py --optimize /builds/slave/aurora-osx64-xr/build/obj-firefox/i386/xulrunner/installer/../../jarlog/ ./ ./ && mv binary.manifest components && printf "manifest components/binary.manifest\n" > chrome.manifest) && (cd universal/xulrunner/xulrunner-6.0a2.en-US.mac-pkg.pkg/Contents/MacOS && /tools/buildbot/bin/python2.6 /builds/slave/aurora-osx64-xr/build/config/createprecomplete.py) && /builds/slave/aurora-osx64-xr/build/build/package/mac_osx/pkg-dmg --source "universal/xulrunner-pkg" --target "xulrunner-6.0a2.en-US.mac-pkg.dmg" --volname "XULRunner" /bin/sh: line 0: cd: universal/xulrunner/xulrunner-6.0a2.en-US.mac-pkg.pkg/Contents/MacOS: No such file or directory sed: components/components.manifest: No such file or directory <a name='err1'></a><font color='000080'>make[2]: *** [make-package] Error 1 </font><font color='000080'>make[1]: *** [default] Error 2 </font><font color='000080'>make: *** [package] Error 2 </font>program finished with exit code 2 After the patch landed cd ../../dist/universal/xulrunner/XUL.framework/Versions/Current; rm -rf xulrunner-config regchrome* regxpcom* core bsdecho gtscc js js-config jscpucfg nsinstall viewer TestGtkEmbed codesighs* elf-dynstr-gc mangle* maptsv* mfc* mkdepend* msdump* msmap* nm2tsv* nsinstall* res/samples res/throbber shlibsign* ssltunnel* certutil* pk12util* winEmbed.exe chrome/chrome.rdf chrome/app-chrome.manifest chrome/overlayinfo components/compreg.dat components/xpti.dat content_unit_tests necko_unit_tests *.dSYM /builds/slave/aurora-osx64-xr/build/obj-firefox/i386/config/nsinstall -D ../../dist/ Compressing... cd ../../dist && (cd universal/xulrunner/XUL.framework/Versions/Current && rm -f omni.jar components/binary.manifest && grep -h '^binary-component' components/*.manifest > binary.manifest ; sed -e 's/^binary-component/#binary-component/' components/components.manifest > components.manifest && mv components.manifest components && zip -r9m omni.jar chrome chrome.manifest components/*.js components/*.xpt components/*.manifest modules res defaults greprefs.js jsloader -x chrome/icons/\* defaults/pref/channel-prefs.js res/cursors/\* res/MainMenu.nib/\* && true && /tools/buildbot/bin/python2.6 /builds/slave/aurora-osx64-xr/build/config/optimizejars.py --optimize /builds/slave/aurora-osx64-xr/build/obj-firefox/i386/xulrunner/installer/../../jarlog/ ./ ./ && mv binary.manifest components && printf "manifest components/binary.manifest\n" > chrome.manifest) && (cd universal/xulrunner/XUL.framework/Versions/Current && /tools/buildbot/bin/python2.6 /builds/slave/aurora-osx64-xr/build/config/createprecomplete.py) && /builds/slave/aurora-osx64-xr/build/build/package/mac_osx/pkg-dmg --source "universal/xulrunner-pkg" --target "xulrunner-6.0a2.en-US.mac-pkg.dmg" --volname "XULRunner" etc... There is also the following that appears to be breaking uploading the build which is not related to this bug or bug 386760 that also affects 1.9.1, 1.9.2, and 2.0 where bug 386760 did not land. ======== BuildStep started ======== find filepath failed === Output === bash -c 'find build/obj-firefox/i386/dist -maxdepth 1 -type f -name *.dmg' in dir /builds/slave/aurora-osx64-xr/. (timeout 1200 secs) watching logfiles {} argv: ['bash', '-c', 'find build/obj-firefox/i386/dist -maxdepth 1 -type f -name *.dmg'] environment: Apple_PubSub_Socket_Render=/tmp/launch-u6x8Ty/Render CVS_RSH=ssh DISPLAY=/tmp/launch-C7ZmUA/:0 HOME=/Users/cltbld LOGNAME=cltbld PATH=/tools/buildbot/bin:/tools/python/bin:/opt/local/bin:/usr/bin:/bin:/usr/sbin:/sbin:/usr/local/bin:/usr/X11/bin PWD=/builds/slave/aurora-osx64-xr PYTHONPATH=/tools/buildbot/lib/python2.6/site-packages:/tools/twisted/lib/python2.6/site-packages/:/tools/twisted-core/lib/python2.6/site-packages:/tools/zope-interface/lib/python2.6/site-packages SHELL=/bin/bash SSH_AUTH_SOCK=/tmp/launch-3kaUB2/Listeners TMPDIR=/var/folders/7I/7I253dv+HLesSBUBGCX08E+++TM/-Tmp-/ USER=cltbld VERSIONER_PYTHON_PREFER_32_BIT=no VERSIONER_PYTHON_VERSION=2.6 __CF_USER_TEXT_ENCODING=0x1F6:0:0 using PTY: False program finished with exit code 0 elapsedTime=0.284242 === Output ended === ======== BuildStep ended ======== ======== BuildStep started ======== maybe rebooting slave lost === Output === === Output ended === ======== BuildStep ended ========
Comment 28•14 years ago
|
||
The filepath failure is bug 661792, which should be fixed in tomorrows builds.
Updated•7 years ago
|
Product: Core → Firefox Build System
You need to log in
before you can comment on or make changes to this bug.
Description
•