[NETCONF-443] Netconf device configured in clustered topology may not get connected after restart

Mon, 21 Mar 2022 11:43:30 +0000

Not sure whether this is Netconf/netconf bug or Controller/clustering bug.

The "entity" suite has detected this as a failure [0]. This (without any failure in preceding tests) happened for the first time, but previous and next [1] run have no failures.

Comparing karaf logs I do not see a clear cause, but I see the test step is executed in different stage of ODL booting up.

In no-failure run I see [2]:

{netconf-test-device} : using the default directory cache/schema
2017-07-27 00:57:24,151 | INFO | lt-dispatcher-22 | NetconfNodeActor | 282 - org.opendaylight.netconf.topology-singleton - 1.2.2.SNAPSHOT | RemoteDevice{netconf-test-device}

: Schema context resolved: [ModuleEffectiveStatementImpl[name=ietf-inet-types, namespace=urn:ietf:params:xml:ns:yang:ietf-inet-types, revision=2010-09-24, prefix=inet, yangVersion=1], ModuleEffectiveStatementImpl[name=yang-ext, namespace=urn:opendaylight:yang:extension:yang-ext, revision=2013-07-09, prefix=ext, yangVersion=1], ModuleEffectiveStatementImpl[name=people, namespace=urn:opendaylight:params:xml:ns:yang:controller:config:sal-clustering-it:people, revision=2014-08-18, prefix=people, yangVersion=1], ModuleEffectiveStatementImpl[name=test, namespace=urn:opendaylight:test:netconf:crud, revision=2014-10-18, prefix=tt, yangVersion=1], ModuleEffectiveStatementImpl[name=ietf-yang-types, namespace=urn:ietf:params:xml:ns:yang:ietf-yang-types, revision=2013-07-15, prefix=yang, yangVersion=1], ModuleEffectiveStatementImpl[name=car, namespace=urn:opendaylight:params:xml:ns:yang:controller:config:sal-clustering-it:car, revision=2014-08-18, prefix=car, yangVersion=1], ModuleEffectiveStatementImpl[name=car-people, namespace=urn:opendaylight:params:xml:ns:yang:controller:config:sal-clustering-it:car-people, revision=2014-08-18, prefix=car, yangVersion=1], ModuleEffectiveStatementImpl[name=car-purchase, namespace=urn:opendaylight:params:xml:ns:yang:controller:config:sal-clustering-it:car-purchase, revision=2014-08-18, prefix=cp, yangVersion=1], ModuleEffectiveStatementImpl[name=ncmount-l2fib, namespace=org:opendaylight:coretutorials:ncmount:example:l2fib, revision=2016-03-07, prefix=l2fib, yangVersion=1], ModuleEffectiveStatementImpl[name=ietf-inet-types, namespace=urn:ietf:params:xml:ns:yang:ietf-inet-types, revision=2013-07-15, prefix=inet, yangVersion=1], ModuleEffectiveStatementImpl[name=ietf-netconf-monitoring, namespace=urn:ietf:params:xml:ns:yang:ietf-netconf-monitoring, revision=2010-10-04, prefix=ncm, yangVersion=1], ModuleEffectiveStatementImpl[name=ietf-netconf-monitoring-extension, namespace=urn:ietf:params:xml:ns:yang:ietf-netconf-monitoring-extension, revision=2013-12-10, prefix=ncme, yangVersion=1]]
2017-07-27 00:57:24,155 | INFO | lt-dispatcher-22 | SlaveSalFacade | 282 - org.opendaylight.netconf.topology-singleton - 1.2.2.SNAPSHOT | RemoteDevice

{netconf-test-device}: Slave mount point registered.
2017-07-27 00:57:24,941 | INFO | h for user karaf | command | 291 - org.apache.karaf.log.command - 3.0.8 | ROBOT MESSAGE: Starting test Modify_Device_Data_With_Original_Owner

In the failed run [3]:

2017-07-26 07:07:15,552 | INFO | on-dispatcher-59 | NetconfTopologyUtils | 282 - org.opendaylight.netconf.topology-singleton - 1.2.2.Carbon | RemoteDevice{netconf-test-device}

I do not think Robot connecting to karaf ssh port and logging the start of test step can make a difference.

I will try to reproduce on Sandbox with more verbose log.

[0] https://logs.opendaylight.org/releng/jenkins092/netconf-csit-3node-clustering-only-carbon/607/log.html.gz#s1-s5-t14-k2-k1-k2-k1-k4-k1
[1] https://logs.opendaylight.org/releng/jenkins092/netconf-csit-3node-clustering-only-carbon/608/log.html.gz#s1-s5-t14-k2-k2-k2-k1-k4-k1
[2] https://logs.opendaylight.org/releng/jenkins092/netconf-csit-3node-clustering-only-carbon/608/odl1_karaf.log.gz
[3] https://logs.opendaylight.org/releng/jenkins092/netconf-csit-3node-clustering-only-carbon/607/odl1_karaf.log.gz

OpenDaylight JIRA

[NETCONF-443] Netconf device configured in clustered topology may not get connected after restart