Re: [Call For Help] Clang + OpenJDK + head + amd64 == cocktail of death (for clusters)
On 25/07/2013 17:33, Baptiste Daroussin wrote: On Thu, Jul 25, 2013 at 01:13:02PM +0400, Dmitry Samersoff wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hi Baptiste, If I understand you correct, hotspot crashes because of out of memory or some other resource constraint? Could you send me whole hs_err_.log file (or better couple of it)? - -Dmitry FYI Dmitry fixed the problem thank you very much ! With the following patch: http://www.andric.com/freebsd/ports/java__openjdk6-deoptimization-crash.diff The bug is known upstream as: http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6636110 I have committed in the ports tree as it is necessary to finish properly building packages. regards, Bapt Thank you all. The new port version solved the issue I encountered with clang and java in 9.2-BETA1. regards Arnaud ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to "freebsd-ports-unsubscr...@freebsd.org"
Re: [Call For Help] Clang + OpenJDK + head + amd64 == cocktail of death (for clusters)
On Thu, Jul 25, 2013 at 01:13:02PM +0400, Dmitry Samersoff wrote: > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > > Hi Baptiste, > > If I understand you correct, hotspot crashes because of out of memory > or some other resource constraint? > > Could you send me whole hs_err_.log file (or better couple of it)? > > - -Dmitry > FYI Dmitry fixed the problem thank you very much ! With the following patch: http://www.andric.com/freebsd/ports/java__openjdk6-deoptimization-crash.diff The bug is known upstream as: http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6636110 I have committed in the ports tree as it is necessary to finish properly building packages. regards, Bapt pgpC0MMnoyo1o.pgp Description: PGP signature
Re: [Call For Help] Clang + OpenJDK + head + amd64 == cocktail of death (for clusters)
Trying to build openjdk6 dies on r253620 Tried setting MAKE_JOBS_UNSAFE=yes with not much success, happy to do further testing if anyone has any diffs or suggestions. Sevan / Venture37 ---> Reinstalling 'openjdk6-b27_5' (java/openjdk6) ---> Building '/usr/ports/java/openjdk6' ===> Cleaning for openjdk6-b27_5 ===> License GPLv2 accepted by the user ===> Found saved configuration for openjdk6-b27_5 ===> openjdk6-b27_5 depends on file: /usr/local/sbin/pkg - found ===> Fetching all distfiles required by openjdk6-b27_5 for building ===> Extracting for openjdk6-b27_5 => SHA256 Checksum OK for openjdk-6-src-b27-26_oct_2012.tar.gz. => SHA256 Checksum OK for jaxp144_05.zip. => SHA256 Checksum OK for jdk6-jaxws2_1_6-2011_06_13.zip. => SHA256 Checksum OK for jdk6-jaf-b20.zip. => SHA256 Checksum OK for apache-ant-1.8.4-bin.tar.bz2. ===> Patching for openjdk6-b27_5 ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/jaxp.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/jaxws.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20120830/7182135-impossible_to_use_some_editors_directly.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20130201/7201068.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20130201/6563318.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20130201/6664509.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20130201/6776941.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20130201/7141694.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20130201/7173145.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20130201/7186945.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20130201/7186948.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20130201/7186952.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20130201/7186954.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20130201/7192392.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20130201/7192393.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20130201/7192977.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20130201/7197546.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20130201/7200491.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20130201/7200500.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20130201/7201064.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20130201/7201066.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20130201/7201070.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20130201/7201071.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20130201/8000210.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20130201/8000537.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20130201/8000540.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20130201/8000631.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20130201/8001242.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20130201/8001307.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20130201/8001972.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20130201/8002325.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20130219/8006446.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20130219/8006777.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20130219/8007688.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20130304/8007014.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20130304/8007675.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/openjdk/7036559-concurrenthashmap_improvements.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20130416/8009063.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/openjdk/8004302-soap_test_failure.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20130416/6657673.patch ===> Applying extra patch /usr/ports/java/openjdk6/files/icedtea/security/20130416/6657673-fixup.patch ===> Applying extra patch
Re: [Call For Help] Clang + OpenJDK + head + amd64 == cocktail of death (for clusters)
On Thu, 25 Jul 2013 10:39:17 +0200 Baptiste Daroussin articulated: > Otherwise I'll have to mark all failing packages as broken > on head amd64 in the portstree :( Ouch :poop: -- Jerry ♔ Disclaimer: off-list followups get on-list replies or get ignored. Please do not ignore the Reply-To header. __ signature.asc Description: PGP signature
Re: [Call For Help] Clang + OpenJDK + head + amd64 == cocktail of death (for clusters)
On Thu, Jul 25, 2013 at 01:13:02PM +0400, Dmitry Samersoff wrote: > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > > Hi Baptiste, > > If I understand you correct, hotspot crashes because of out of memory > or some other resource constraint? > > Could you send me whole hs_err_.log file (or better couple of it)? > > - -Dmitry I would love to but this is done on the cluster where the jail is wiped out once it finished (failure or not) Best would be probably to ulimit -m 4194304 before building let say textproc/fop from the ports tree. you should hit it pretty quickly. regards, Bapt pgpLzQNmMpX7K.pgp Description: PGP signature
Re: [Call For Help] Clang + OpenJDK + head + amd64 == cocktail of death (for clusters)
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hi Baptiste, If I understand you correct, hotspot crashes because of out of memory or some other resource constraint? Could you send me whole hs_err_.log file (or better couple of it)? - -Dmitry On 2013-07-25 12:39, Baptiste Daroussin wrote: > Hi all, > > It's been a while since we are experiencing major problem on the > package buidling cluster: it dies when building package for head > amd64. (the buidling jail being the snapshort from July 7th iirc.) > > After some investigation we discover that blacklisting openjdk6 > allows the building process to go to completion again. > > But not providing java packages is not a good solution. > > We have managed to "fix" the build by limiting the allowed memory > usage on each building jail to 8G, the result is a huge load of > java packages just fail to build because it eats all the memory and > then segfault, but at least they do not kill the box again. Have a > look at the pkg-fallout mailing list you should be able to see lots > of java port failed because of that. > > It seems to happen only on head amd64, so far we think it is only > happening when jdk is built with clang. > > I have no time, neither skill to investigate that, so please if you > are interested in having java packages available have a look at it, > fix the problem if any may that be clang, openjdk, the ports tree > framework for java building, or anything related. Otherwise I'll > have to mark all failing packages as broken on head amd64 in the > portstree :( > > regards, Bapt > - -- Dmitry Samersoff Saint Petersburg, Russia, http://devnull.samersoff.net * There will come soft rains ... -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.19 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQEcBAEBAgAGBQJR8OweAAoJEHEy08c4gIABwsAH/jdIMktpssHgyhaE8GH2uEme /8S5n8tXd2mCRszKsPde0FPwBsz1XfyeM/6sD1o6vBeLMCEwCKxyUHyRxL3cwZ0E XYosan6id1v8vTGzHid3qGxJ68MHsGNbTO4uJ2mSh0sO39frofiL11kVyzMel31/ NhUElw/aZcWWVSCM3rVXx+QuzzF9tHn2WhYG9AQLYJcMuqerjvgniIYKB2C/TKo8 eM0qxhsR2JfI08TIruDmo/2AnDQICMI9oObdO1ewvuqOOP1ydndVMUj0ZR8NQgEx eUYIcoWhpo6o2F9j1QXdjNIOfRod02MbCrtAimAXXpXlwgFMrVujowBB5P01ErY= =svP7 -END PGP SIGNATURE- ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to "freebsd-ports-unsubscr...@freebsd.org"
Re: [Call For Help] Clang + OpenJDK + head + amd64 == cocktail of death (for clusters)
On 2013-Jul-25 10:39:17 +0200, Baptiste Daroussin wrote: >After some investigation we discover that blacklisting openjdk6 allows the >building process to go to completion again. ... >It seems to happen only on head amd64, so far we think it is only >happening when jdk is built with clang. This mail arrives at an opportune time. I've just discovered that if I build openjdk6 with clang (on head/amd64), the resultant jdk SEGV's if I again try to build openjdk6. If I build it with "USE_GCC=any" then the problem goes away. >I have no time, neither skill to investigate that, I don't have the time to investigate further but forcing the use of gcc instead of clang is at least a workaround. -- Peter Jeremy pgprMpeYl514w.pgp Description: PGP signature
[Call For Help] Clang + OpenJDK + head + amd64 == cocktail of death (for clusters)
Hi all, It's been a while since we are experiencing major problem on the package buidling cluster: it dies when building package for head amd64. (the buidling jail being the snapshort from July 7th iirc.) After some investigation we discover that blacklisting openjdk6 allows the building process to go to completion again. But not providing java packages is not a good solution. We have managed to "fix" the build by limiting the allowed memory usage on each building jail to 8G, the result is a huge load of java packages just fail to build because it eats all the memory and then segfault, but at least they do not kill the box again. Have a look at the pkg-fallout mailing list you should be able to see lots of java port failed because of that. It seems to happen only on head amd64, so far we think it is only happening when jdk is built with clang. I have no time, neither skill to investigate that, so please if you are interested in having java packages available have a look at it, fix the problem if any may that be clang, openjdk, the ports tree framework for java building, or anything related. Otherwise I'll have to mark all failing packages as broken on head amd64 in the portstree :( regards, Bapt pgpTfK1ILvqc3.pgp Description: PGP signature