Description of problem: ----------------------- There is no way to stop the gdeploy configuration, when there is a failure encountered. This leads to bad state of the machine, where lvm commands hang without disabling multipath The issue will be solved with bug[1] which provides an option for each section to abort on failure. But because that change take a time, the temporary solution is to provide 2 gdeploy config files. One containing all the installation of rpms from the channels. Only when the first step succeeds, user is asked to run the second gdeploy conf file. Note this will be the temporary solution for the limited availability release [1] - https://bugzilla.redhat.com/show_bug.cgi?id=1346253 Version-Release number of selected component (if applicable): ------------------------------------------------------------- gdeploy-2.0-16 How reproducible: ----------------- NA Steps to Reproduce: ------------------- NA Actual results: --------------- NA Expected results: ----------------- provide 2 gdeploy config files, with one conf file only installing the required rpms, and other doing all the configurations
The solution to provide 2 gdeploy config files is a workaround to provide a way to stop the configuration, when encountered with a failure. The actual issue is filed in the bz 1341484
With the new hc requirements, we no longer do a yum update or install. Layered install will not be done in future. The requirement of two configuration files is no longer needed. Will close once I get confirmation from sahina@
I think if we have Bug 1341484 implemented - we don't need to take this bug in, but can close this.
Yes we can close this, this is a duplicate of 1341484. *** This bug has been marked as a duplicate of bug 1341484 ***
Since this bug is close removing all the acks and keywords