On 17/10/2019 06:39, JamO Luhrsen wrote:

actually, it looks like build #60 just passed so we can check what CSIT
looks like from that. It'll
be this job:

https://jenkins.opendaylight.org/releng/job/integration-distribution-test-magnesium/53/
<https://jenkins.opendaylight.org/releng/job/integration-distribution-test-magnesium/52/>

I analyzed
https://jenkins.opendaylight.org/releng/job/netconf-csit-1node-userfeatures-all-magnesium/
and the failures look like:

1) something caused by Java 8 -> 11 transition

https://jenkins.opendaylight.org/releng/job/netconf-csit-1node-userfeatures-all-magnesium/33/
ran with Java 8 and succeeded

https://jenkins.opendaylight.org/releng/job/netconf-csit-1node-userfeatures-all-magnesium/34/
ran with Java 11 and failed

Both runs happened *before* the MRI patches went in.


2) they should be handled in CSIT infra

https://logs.opendaylight.org/releng/vex-yul-odl-jenkins-1/netconf-csit-1node-userfeatures-all-magnesium/34/robot-plugin/log.html.gz#s1-s3-s1-t3
is clearly a difference in whitespace, which is insignificant (and shows
Java 11 pretty-printing XML).

I know nothing about RF-fu, but the test does:

Perform_Test
[Arguments] ${name}
[Documentation] Load and send the request from the dataset and compare the returned reply to the one stored in the dataset.
${actual}= Load_And_Send_Message ${name}
${expected}= Load_Expected_Reply ${name}
${actual}= String.Remove String ${actual} ]]>]]>
${actual}= String.Strip String ${actual} mode=both
XML.Elements Should Be Equal ${actual} ${expected} normalize_whitespace=True
[Return] ${actual}

which I think should do the trick, based on
https://robotframework.org/robotframework/latest/libraries/XML.html#Elements%20Should%20Be%20Equal

Do we have bug in RF or am I missing something?

Regards,
Robert