|
||||||||
This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators. For more information on JIRA, see: http://www.atlassian.com/software/jira |
[JIRA] (JENKINS-14109) Passing ${NODE_NAME} as Node Parameter to downstream job not possible on master
kuypers.d...@googlemail.com (JIRA) Mon, 25 Jun 2012 01:31:26 -0700
- [JIRA] (JENKINS-14109) Passing ${NODE_N... kuypers.d...@googlemail.com (JIRA)
- [JIRA] (JENKINS-14109) Passing ${N... d...@fortysix.ch (JIRA)
- [JIRA] (JENKINS-14109) Passing ${N... kuypers.d...@googlemail.com (JIRA)
- [JIRA] (JENKINS-14109) Passing ${N... scm_issue_l...@java.net (JIRA)
- [JIRA] (JENKINS-14109) Passing ${N... scm_issue_l...@java.net (JIRA)
I have a label for the starting job named "clearcase" because we use clearcase and it is only available on some of my slaves with most CPU/disk power and on the master. So it is possible for the job to start on 3 out of 8 slaves and the master. After updating the sources and compiling everything two other jobs are started: One to build the installer and another one to apply a label for the successful build to all files used for the build. So I need the work space of the successful build for those two depending jobs. I pass the name of the host where the job ran via NodeLabel parameter to the downstream jobs: ${HOSTNAME}=${NODE_NAME}. ${NODE_NAME} is empty on the master which leads to downstream jobs waiting forever because they cannot find a node named "".
I tried to mess around a bit with groovy and set ${NODE_NAME} environment variable as post-build if empty, but did not succeed.:-/