Recently upgraded from 15.0.2 to 17.1.0. Clicking 'save' on a new scheduled outage or deleting an existing outage results in an error message in the UI and the zeroing out of collectd-configuration.xml, requiring the file to be restored before operations may continue.
web.log reports:
Environment
Centos 6.7, Postgres 9.1, Java 1.8.0_72
Acceptance / Success Criteria
None
Lucidchart Diagrams
Activity
Benjamin Reed June 29, 2016 at 5:37 PM
Haven't seen any update on this so I'm gonna close-as-Configuration.
Jeff Gehlbach March 8, 2016 at 12:28 PM
Correct. Unblockerized. We should resolve this one as Configuration, assuming confirms my suspicion, and replace it with a separate issue describing the problem with the XSDs.
Benjamin Reed March 8, 2016 at 12:03 PM
Can we assume this is not a blocker for 17.1.1 then? I'd love to do the release tomorrow if possible.
Jeff Gehlbach March 7, 2016 at 5:45 PM
Edited
I think the problem is that before the upgrade, you've broken the symbolic link for /opt/opennms/share to /var/opennms. As a result, the file referred to by the pathname /opt/opennms/share/xsds/collectd-configuration.xml is the version from Horizon 15. That version lacks the definition of the remote attribute on the <package> element.
Try copying all the files from /var/opennms/xsds to /opt/opennms/share/xsds and restart OpenNMS. The problem should go away if I'm right.
Palmer Sample March 7, 2016 at 1:37 PM
Thanks. In the meantime we have rolled back to 15.x and all is functional again. Will look for the 17.1.1 changes when released.
Recently upgraded from 15.0.2 to 17.1.0. Clicking 'save' on a new scheduled outage or deleting an existing outage results in an error message in the UI and the zeroing out of collectd-configuration.xml, requiring the file to be restored before operations may continue.
web.log reports: