[ https://issues.jboss.org/browse/WFWIP-218?page=com.atlassian.jira.plugin.... ] Ondrej Chaloupka edited comment on WFWIP-218 at 11/29/19 10:12 AM: ------------------------------------------------------------------- Just mentioning that I started to investigate on this. I agree with [~tomekadamski] that the WFTC-75 seems to be fixing the WFTC-77 as well. I created the reproducer https://gitlab.mw.lab.eng.bos.redhat.com/ochaloup/tests-transactions/comm... that does the same (or I assume it does the same) as the test in the eap qe openshift testsuite. On debugging and testing all works fine. But! The strange thing is that this issue WFWIP-218 is reproducible with the test {{EjbTxnRemotingScaleDownTest#testTxStatelessServerSecondCommitThrowRmFail}} and the scale down does clean the WFTC folder - the commit happens during the recovery, the Narayana object store is clean but the WFTC folder is not empty. I will be running more investigation on this. For sake of completeness I run this {code} mvn clean test -Pcd -Dtest=EjbTxnRemotingScaleDownTest#testTxStatelessServerSecondCommitThrowRmFail -Dconsole-log-level=DEBUG -Poperator {code} on local CRC instance (http://pastebin.test.redhat.com/815593) with {{test.properties}} {code} xtf.openshift.namespace=ochaloup-namespace xtf.bm.namespace=ochaloup-builds-namespace # this is the url of CRC, try `crc ip` xtf.openshift.url=https://192.168.130.11:6443 xtf.eap.properties.location=/opt/eap xtf.openshift.binary.path=/home/ochaloup/.crc/bin/oc # operator built by Jeff and published xtf.operator.wildfly.image=docker-registry.engineering.redhat.com/jbossqe... # for the latest EAP image to take xtf.eap.cd.image=registry-proxy.engineering.redhat.com/rh-osbs/jboss-eap-... xtf.eap.cd.properties.eap.runtime.image=registry-proxy.engineering.redhat... xtf.waiting.timeout=600000 test.script.debug=true # needed for yaml not being downloaded from quay repo or what xtf.operator.eap.resources.context= {code} was (Author: ochaloup): Just mentioning that I started to investigate on this. I agree with [~tomekadamski] that the WFTC-75 seems to be fixing the WFTC-77 as well. I created the reproducer https://gitlab.mw.lab.eng.bos.redhat.com/ochaloup/tests-transactions/comm... that does the same (or I assume it does the same) as the test in the eap qe openshift testsuite. On debugging and testing all works fine. But! The strange thing is that this issue WFWIP-218 is reproducible with the test `EjbTxnRemotingScaleDownTest#testTxStatelessServerSecondCommitThrowRmFail` and the scale down does clean the WFTC folder - the commit happens during the recovery, the Narayana object store is clean but the WFTC folder is not empty. I will be running more investigation on this. For sake of completeness I run this {code} mvn clean test -Pcd -Dtest=EjbTxnRemotingScaleDownTest#testTxStatelessServerSecondCommitThrowRmFail -Dconsole-log-level=DEBUG -Poperator {code} on local CRC instance (http://pastebin.test.redhat.com/815593) with {{test.properties}} {code} xtf.openshift.namespace=ochaloup-namespace xtf.bm.namespace=ochaloup-builds-namespace # this is the url of CRC, try `crc ip` xtf.openshift.url=https://192.168.130.11:6443 xtf.eap.properties.location=/opt/eap xtf.openshift.binary.path=/home/ochaloup/.crc/bin/oc # operator built by Jeff and published xtf.operator.wildfly.image=docker-registry.engineering.redhat.com/jbossqe... # for the latest EAP image to take xtf.eap.cd.image=registry-proxy.engineering.redhat.com/rh-osbs/jboss-eap-... xtf.eap.cd.properties.eap.runtime.image=registry-proxy.engineering.redhat... xtf.waiting.timeout=600000 test.script.debug=true # needed for yaml not being downloaded from quay repo or what xtf.operator.eap.resources.context= {code} -- This message was sent by Atlassian Jira (v7.13.8#713008)