Linux machine goes for a hang upon running the lscm command while extracting the component
![]()
- Scenario
I have the machine (12 GB ram), allocated 4GB heap for RTC. when i start the extraction of workspace, i see that my 12 GB of free memory started getting used up, and at the end of source code extraction, i was left with almost 5GB only. When i checked the ps -eaf | grep -i java | grep -i rtc i saw that RTC java is running root 31180 31179 0 Apr25 ? 00:00:19 /usr/build/RTC502/jazz/scmtools/eclipse//../../client/eclipse/jdk/jre/bin/java -Xmx4096m -Xshareclasses:nonfatal -Xquickstart -Dosgi.requiredJavaVersion=1.6 -Xdump:system:events=systhrow,filter=java/lang/OutOfMemoryError,request=exclusive+prepwalk -jar /usr/build/RTC502/jazz/scmtools/eclipse//plugins/org.eclipse.equinox.launcher_1.1.1.R36x_v20101122_1400.jar -os linux -ws gtk -arch x86_64 -showsplash -launcher /usr/build/RTC502/jazz/scmtools/eclipse/scm -name Scm --launcher.library /usr/build/RTC502/jazz/scmtools/eclipse//plugins/org.eclipse.equinox.launcher.gtk.linux.x86_64_1.1.2.R36x_v20101019_1345/eclipse_1310.so -startup /usr/build/RTC502/jazz/scmtools/eclipse//plugins/org.eclipse.equinox.launcher_1.1.1.R36x_v20101122_1400.jar -exitdata 190032 -data @noDefault --config /root/.jazz-scm daemon start --connection-timeout 120000 --inactive-timeout 7200000 -vm /usr/build/RTC502/jazz/scmtools/eclipse//../../client/eclipse/jdk/jre/bin/java -vmargs -Xmx4096m -Xshareclasses:nonfatal -Xquickstart -Dosgi.requiredJavaVersion=1.6 -Xdump:system:events=systhrow,filter=java/lang/OutOfMemoryError,request=exclusive+prepwalk -jar /usr/build/RTC502/jazz/scmtools/eclipse//plugins/org.eclipse.equinox.launcher_1.1.1.R36x_v20101122_1400.jar when i have seen the ps -eaf | grep -i java | grep -i rtc | wc -l 24 i see 24 rtc java processes are running. When i started the source drop again, i have seen that i was left with only 100 Mb of memory and this time i have seen 48 java rtc process running. eventually when i start again, i see the machine will go to hang state and will never respond, only option i see is to reboot the system. My question here is why are the java processes not getting killed here after successful completion of source extraction, and what is the solution to this problem to resolve this hang issue I am using Red Hat Enterprise Linux Server release 6.7 (Santiago) x86_64 (64 bit machine) i am using the below script lscm login -r https://isljazzrtc.in.ibm.com:9550/ccm/ -n localwork -u ${1} -P ${2} rm -rf /source_drop/s810/$3 mkdir /source_drop/s810/$3 cd /source_drop/s810/$3 pres=`pwd` while read line do mkdir $line cd $line date lscm load ${3} $line -r localwork if [ $? -eq 0 ] then echo "READY extraction is success" date else echo " READY extraction failed" exit 1 fi #mv src $line cd $pres done < /usr/scripts/s810comlist // s810comlist is the component name list |
Accepted answer
![]()
Something is very wrong here. Since you are using lscm, it runs as a daemon, and will not quit unless you ask it to do so. But there should be only one lscm daemon running. For a quick check, simply change the "lscm" command to "scm" in your script. The "scm" process should quit once it finishes.
manobhiram juvvadi selected this answer as the correct answer
Comments Thanks Donald, looks like it is working fine for me now
You should get all the answers by reading the manuals carefully.
|
One other answer
![]() We had issues with too many daemons being created. For us it was because we were using a nfs mount. We had to change the config directory of the lscm command to a temp directory. i.e. lscm --config /tmp/rtcconfig$USER By doing this the daemon can correctly create the lock file it needs.
|
Your answer
Dashboards and work items are no longer publicly available, so some links may be invalid. We now provide similar information through other means. Learn more here.