[INTTEST-99] robot crashing with heap dump running controller longevity job

Thu, 12 Aug 2021 09:43:52 +0000

controller-csit-1node-notifications-longevity-only-magnesium runs and passes in
magneisum (fails in aluminium, but that has it's own jira to track).

console log

looks like robot is killed by the host:

05:25:07 /tmp/jenkins1897974077684561086.sh: line 87: 12020 Killed                  robot -N "${TESTPLAN}" --removekeywords wuks -c critical -e exclude -e "skip_if_${DISTROSTREAM}" -v BUNDLEFOLDER:"${BUNDLEFOLDER}" -v BUNDLE_URL:"${ACTUAL_BUNDLE_URL}" -v CONTROLLER:"${ODL_SYSTEM_IP}" -v CONTROLLER_USER:"${USER}" -v GERRIT_BRANCH:"${GERRIT_BRANCH}" -v GERRIT_PROJECT:"${GERRIT_PROJECT}" -v GERRIT_REFSPEC:"${GERRIT_REFSPEC}" -v JAVA_HOME:"${JAVA_HOME}" -v JDKVERSION:"${JDKVERSION}" -v JENKINS_WORKSPACE:"${WORKSPACE}" -v MININET1:"${TOOLS_SYSTEM_2_IP}" -v MININET2:"${TOOLS_SYSTEM_3_IP}" -v MININET3:"${TOOLS_SYSTEM_4_IP}" -v MININET4:"${TOOLS_SYSTEM_5_IP}" -v MININET5:"${TOOLS_SYSTEM_6_IP}" -v MININET:"${TOOLS_SYSTEM_IP}" -v MININET_USER:"${USER}" -v NEXUSURL_PREFIX:"${NEXUSURL_PREFIX}" -v NUM_ODL_SYSTEM:"${NUM_ODL_SYSTEM}" -v NUM_TOOLS_SYSTEM:"${NUM_TOOLS_SYSTEM}" -v ODL_STREAM:"${DISTROSTREAM}" -v ODL_SYSTEM_1_IP:"${ODL_SYSTEM_IP}" -v ODL_SYSTEM_IP:"${ODL_SYSTEM_IP}" -v ODL_SYSTEM_USER:"${USER}" -v SUITES:"${SUITES}" -v TOOLS_SYSTEM_IP:"${TOOLS_SYSTEM_IP}" ${tools_variables} -v TOOLS_SYSTEM_USER:"${USER}" -v USER_HOME:"${HOME}" -v WORKSPACE:/tmp ${TESTOPTIONS} ${SUITES}

Probably the job needs to run on a host with more memory. Possibly the robot
logs are too verbose as well which is usually what eats up lots of memory when
it is post processing the output.xml file

OpenDaylight JIRA

[INTTEST-99] robot crashing with heap dump running controller longevity job