trafodion
trafodion copied to clipboard
Add RH7 supported flag to build & test on additional OS
Testing jenkins processing, which keys off install_features file.
Check Test Started: https://jenkins.esgyn.com/job/Check-PR-master/2648/
Test Failed. https://jenkins.esgyn.com/job/Check-PR-master/2648/
jenkins, retest -- updated RH7 build machine image
New Check Test Started: https://jenkins.esgyn.com/job/Check-PR-master/2649/
Test Failed. https://jenkins.esgyn.com/job/Check-PR-master/2649/
jenkins, retest -- another image update
New Check Test Started: https://jenkins.esgyn.com/job/Check-PR-master/2652/
Test Failed. https://jenkins.esgyn.com/job/Check-PR-master/2652/
New Check Test Started: https://jenkins.esgyn.com/job/Check-PR-master/2653/
Test Failed. https://jenkins.esgyn.com/job/Check-PR-master/2653/
jenkins, retest
New Check Test Started: https://jenkins.esgyn.com/job/Check-PR-master/2655/
Test Failed. https://jenkins.esgyn.com/job/Check-PR-master/2655/
jenkins, retest
I merged PR 1569, it should fix the installer issue to install Trafodion on CentOS 7
New Check Test Started: https://jenkins.esgyn.com/job/Check-PR-master/2657/
Test Failed. https://jenkins.esgyn.com/job/Check-PR-master/2657/
It is mxosrvr not started, I can take a look at my CentOS 7 workstation. I didn't start the DCS in previous test. Will check this.
I have merged a temp fix to DCS start issue, let's try again
jenkins, retest
New Check Test Started: https://jenkins.esgyn.com/job/Check-PR-master/2660/
Test Failed. https://jenkins.esgyn.com/job/Check-PR-master/2660/
There are some regress tests failed. I will try to check them, but I also recommand that at this point, maybe we can disable those failed cases, and once we fix one, add one?
Not only some failed tests, but also core files.
And on the cloudera node on Centos7, the install is still not completing. It fails with: ***[ERROR]: Failed to copy files to remote nodes
Interesting since these are single-node clusters. -- I'll double check that the ssh is set up right in that image.
Follow-up. I've been debugging a bit on the cloudera node install issue. Does not seem to be a set-up or permissions issue. The discover step works fine, but then the install step fails. This scp command works on the first step, but then fails the second step.
['scp', '-oPasswordAuthentication=no', '-r', '/home/jenkins/workspace/jdbc_test-cdh/traf_inst/python-installer/configs', '/home/jenkins/workspace/jdbc_test-cdh/traf_inst/python-installer/scripts', u'slave7-cm57:/tmp/.trafodion_install_temp/']
So seems like some clean-up issue? Makes no sense to me why that would work differently on RH7 or on cloudera nodes than on other nodes.
I create a CentOS 7.2 dev enviorment, and run SQL regression with all hive tests pass... No idea why it failed in Jenkins. Must be some other reasons, but I used install_local_hadoop and not using python installer. So it is very hard for me to fix the hive regression, let me see next round of test result and see what I can do.
jenkins, retest
New Check Test Started: https://jenkins.esgyn.com/job/Check-PR-master/2678/
Test Failed. https://jenkins.esgyn.com/job/Check-PR-master/2678/
@svarnau, what's the state of this change? Should this pull request be closed or are you still working on it?