Building 1.5 from ports ... is this normal?
Greg Lewis
glewis at eyesbeyond.com
Tue Nov 22 15:45:23 PST 2005
On Sun, Nov 20, 2005 at 04:51:22PM -0400, Marc G. Fournier wrote:
> Trying to build 1.5 from ports, on a 4-STABLE server, and it seems to hang
> forever at:
>
> Generating ../generated/jvmtifiles/jvmti.h
> /usr/local/linux-sun-jdk1.4.2/bin/java -classpath ../generated/jvmtifiles
> jvmtiGen -IN
> /vm/ports/usr/ports/java/jdk15/work/hotspot/src/share/vm/prims/jvmti.xml
> -XSL
> /vm/ports/usr/ports/java/jdk15/work/hotspot/src/share/vm/prims/jvmtiH.xsl
> -OUT ../generated/jvmtifiles/jvmti.h
> Info: jvmtiGen using javax.xml.transform.TransformerFactory =
> org.apache.xalan.processor.TransformerFactoryImpl
>
> Using ps, there seems to be alot of java processes started up related to
> this:
>
> # ps aux | grep java | grep jdk15
> root 29520 0.0 0.6 212008 22240 p9 S+ 4:21PM 0:00.77
> /usr/local/linux-sun-jdk1.4.2/bin/java -classpath ../generated/jvmtifiles
> jvmtiGen -IN
> /vm/ports/usr/ports/java/jdk15/work/hotspot/src/share/vm/prims/jvmti.xml
> -XSL /vm/ports/usr/ports/java/jdk15/work
> root 29519 0.0 0.6 212008 22240 p9 I+ 4:21PM 0:00.40
> /usr/local/linux-sun-jdk1.4.2/bin/java -classpath ../generated/jvmtifiles
> jvmtiGen -IN
> /vm/ports/usr/ports/java/jdk15/work/hotspot/src/share/vm/prims/jvmti.xml
> -XSL /vm/ports/usr/ports/java/jdk15/work
> root 29518 0.0 0.6 212008 22240 p9 I+ 4:21PM 0:00.00
> /usr/local/linux-sun-jdk1.4.2/bin/java -classpath ../generated/jvmtifiles
> jvmtiGen -IN
> /vm/ports/usr/ports/java/jdk15/work/hotspot/src/share/vm/prims/jvmti.xml
> -XSL /vm/ports/usr/ports/java/jdk15/work
> root 29517 0.0 0.6 212008 22240 p9 I+ 4:21PM 0:00.00
> /usr/local/linux-sun-jdk1.4.2/bin/java -classpath ../generated/jvmtifiles
> jvmtiGen -IN
> /vm/ports/usr/ports/java/jdk15/work/hotspot/src/share/vm/prims/jvmti.xml
> -XSL /vm/ports/usr/ports/java/jdk15/work
> root 29516 0.0 0.6 212008 22240 p9 I+ 4:21PM 0:00.01
> /usr/local/linux-sun-jdk1.4.2/bin/java -classpath ../generated/jvmtifiles
> jvmtiGen -IN
> /vm/ports/usr/ports/java/jdk15/work/hotspot/src/share/vm/prims/jvmti.xml
> -XSL /vm/ports/usr/ports/java/jdk15/work
> root 29515 0.0 0.6 212008 22240 p9 I+ 4:21PM 0:00.01
> /usr/local/linux-sun-jdk1.4.2/bin/java -classpath ../generated/jvmtifiles
> jvmtiGen -IN
> /vm/ports/usr/ports/java/jdk15/work/hotspot/src/share/vm/prims/jvmti.xml
> -XSL /vm/ports/usr/ports/java/jdk15/work
> root 29514 0.0 0.6 212008 22240 p9 S+ 4:21PM 0:00.27
> /usr/local/linux-sun-jdk1.4.2/bin/java -classpath ../generated/jvmtifiles
> jvmtiGen -IN
> /vm/ports/usr/ports/java/jdk15/work/hotspot/src/share/vm/prims/jvmti.xml
> -XSL /vm/ports/usr/ports/java/jdk15/work
> root 29513 0.0 0.6 212008 22240 p9 S+ 4:21PM 0:00.02
> /usr/local/linux-sun-jdk1.4.2/bin/java -classpath ../generated/jvmtifiles
> jvmtiGen -IN
> /vm/ports/usr/ports/java/jdk15/work/hotspot/src/share/vm/prims/jvmti.xml
> -XSL /vm/ports/usr/ports/java/jdk15/work
> root 29512 0.0 0.6 212008 22240 p9 I+ 4:21PM 0:01.11
> /usr/local/linux-sun-jdk1.4.2/bin/java -classpath ../generated/jvmtifiles
> jvmtiGen -IN
> /vm/ports/usr/ports/java/jdk15/work/hotspot/src/share/vm/prims/jvmti.xml
> -XSL /vm/ports/usr/ports/java/jdk15/work
> root 29059 0.0 0.0 2080 844 p9 I+ 4:21PM 0:00.01 gmake -f
> /vm/ports/usr/ports/java/jdk15/work/hotspot/build/bsd/Makefile product
> HOTSPOT_BUILD_VERSION=1.5.0-p2-root_20_nov_2005_16_21
> GAMMADIR=/vm/ports/usr/ports/java/jdk15/work/hotspot
> root 29058 0.0 0.0 628 352 p9 I+ 4:21PM 0:00.00 /bin/sh -c
> (cd ./../build/bsd-i586/hotspot-i586/tmp; gmake -f
> /vm/ports/usr/ports/java/jdk15/work/hotspot/build/bsd/Makefile product
> HOTSPOT_BUILD_VERSION=1.5.0-p2-root_20_nov_2005_16_21 GAMMADIR=
> root 28650 0.0 0.0 632 352 p9 I+ 4:20PM 0:00.00 /bin/sh -ec
> (cd /vm/ports/usr/ports/java/jdk15/work/control/make; /usr/bin/env
> DONT_BUILD_DEPLOY="YES" ALT_BOOTDIR="/usr/local/linux-sun-jdk1.4.2"
> ALT_MOTIF_DIR="/usr/X11R6" SYS_CFLAGS="-O -mpentium
> uranus#
>
> The top one's CPU time is slowly increasing, but the rest just seem to be
> sitting there doing nothing ...
>
> The server isn't doing much in the way of swapping:
>
> # pstat -s
> Device 1K-blocks Used Avail Capacity Type
> /dev/da0s1b 6291328 2584 6288744 0% Interleaved
>
> And top doesn't show any 'cpu hog' processes or anything:
>
> last pid: 31921; load averages: 0.37, 0.42, 0.61
> up 6+00:07:14 16:45:45
> 473 processes: 2 running, 470 sleeping, 1 zombie
> CPU states: 0.0% user, 0.0% nice, 0.0% system, 0.0% interrupt, 100%
> idle
> Mem: 2290M Active, 648M Inact, 395M Wired, 95M Cache, 199M Buf, 93M Free
> Swap: 6144M Total, 2584K Used, 6141M Free
>
> PID USERNAME PRI NICE SIZE RES STATE C TIME WCPU CPU COMMAND
> 63760 88 2 0 81516K 58392K poll 2 405:56 9.52% 9.52% mysqld
> 31920 root 28 0 2476K 1144K CPU2 2 0:01 4.04% 1.46% top
> 30217 106 18 0 46184K 40244K lockf 0 0:01 0.88% 0.88% perl5.8.7
> 13980 nobody 2 0 452M 356M poll 0 16:48 0.00% 0.00% nsd
> 38721 106 2 0 14668K 13596K poll 0 14:31 0.00% 0.00% clamd
> 4259 nobody 2 0 152M 126M poll 2 9:26 0.00% 0.00% nsd
> 70523 www 18 0 12276K 9844K lockf 2 7:38 0.00% 0.00% httpd
> 70528 www 2 0 12340K 9924K select 0 6:22 0.00% 0.00% httpd
> 80397 www 2 0 227M 40704K poll 0 5:53 0.00% 0.00% java
>
> Is it normal for this to take a long time to build? Those processes have
> been running for >20 minutes right now, with negligible CPU time :(
Thats not normal. I've built 1.5 many times on my 4.11 box without it
taking anywhere near that long at that step.
Do the processes ever complete or are they essentially wedged?
--
Greg Lewis Email : glewis at eyesbeyond.com
Eyes Beyond Web : http://www.eyesbeyond.com
Information Technology FreeBSD : glewis at FreeBSD.org
More information about the freebsd-java
mailing list