Description of problem: Attempted to boot using a ks stream promoted to dev and aeolus would fail. Ian McCleod debugged: 14:05:23) imcleod: sreichar: So, here is what is happening. Oz expects to find a ".treeinfo" file in the root of your install URL. This is present on the DVD and on local mirrors of the DVD. (14:05:51) imcleod: sreichar: It is not present in your repo. As a result, Oz falls back to assuming that the initrd and kernel are in "images/pxeboot", which they are not. (14:06:10) imcleod: sreichar: So, ideally, we need Katello to produce a ks tree that more closely matches the content of the DVD. When I switched to use the Locker ks tree - there were no issues. Version-Release number of selected component (if applicable): [root@cf-se1 Red_Hat_Enterprise_Linux_6_Server_RPMs_61_x86_64]# /pub/scripts/post_install_configuration_scripts/cf-se-versions Red Hat Enterprise Linux Server release 6.1 (Santiago) Linux cf-se1.cloud.lab.eng.bos.redhat.com 2.6.32-131.17.1.el6.x86_64 #1 SMP Thu Sep 29 10:24:25 EDT 2011 x86_64 x86_64 x86_64 GNU/Linux PyYAML-3.09-14.el6_1.x86_64 package epel-release is not installed facter-1.5.9-1.el6.noarch js-1.70-12.el6_0.x86_64 mongodb-1.6.4-3.el6_0.x86_64 mongodb-server-1.6.4-3.el6_0.x86_64 puppet-2.6.11-1.el6_1.noarch pymongo-1.9-8.el6_1.x86_64 katello-0.1.97-1.el6.x86_64 katello-all-0.1.97-1.el6.x86_64 katello-cli-0.1.10-1.el6.noarch katello-configure-0.1.8-1.el6.noarch package katello-repos is not installed [root@cf-se1 Red_Hat_Enterprise_Linux_6_Server_RPMs_61_x86_64]# How reproducible: yes - repeatable Steps to Reproduce: 1. 2. 3. Actual results: Expected results: Additional info:
The issue here is not treeinfo, pulp sets the treeinfo file at the top level url. The issue is how the repo associate was doing symlinks. There is a bug in pulp i noticed and fixed today where the symlink logic was missing the sub directory and since you're doing promote in katello it uses the same logic from pulp. This should make it into next pulp qe build. I'll update this bug with the fixed version of pulp and you can validate it.
Putting this ON_QA. Note: Please test it against next pulp qa build or CR. ^^^
Prad, can you tell us how to test this - I am guessing checking for some symlinks somewhere in pulp after a promotion of something?
This BZ is a disaster. There is no git commit hash,no indication of what version of Pulp the fix lands in, and no information on how to verify the fix. Pulp dev needs to provide this information. I'm moving it back to ASSIGNED and resetting to beta 6.
Hmm... dev should not be moving to on_qa at this point in the workflow, only the errata tool should do that. I'm going to move this to modified, please make sure the issue is attached to the correct advisory and the BZ state will be handled automatically.
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. http://rhn.redhat.com/errata/RHEA-2012-0666.html