On Wed, Oct 25, 2017 at 10:37 AM, James LaBarre <jlabarre@redhat.com> wrote:
Would I need to completely remove the existing .quickstart directory in order to run this, or can I update an existing install? (it didn't seen to work with an existing config, but I might have done it wrong)
If you want to update your .quickstart directory vs blow it away you can source ~/.quickstart/bin/activate and execute python setup.py install from tq and tqe to get the latest changes into the right locations.
I've been trying this, but the re-run of quickstart.sh fails a short way into the process. Once O get to a fetch-images task, it quits the process.
===================================
TASK [fetch-images : Get image expected checksum]
******************************
task path:
/root/tripleo-quickstart/roles/fetch-images/tasks/fetch.yml:66
Monday 30 October 2017 10:28:18 -0400 (0:00:00.053)
0:01:01.036 ********
fatal: [127.0.0.2]: FAILED! => {"changed": true, "cmd":
["curl", "-sfL",
"http://images.rdoproject.org/pike/rdo_trunk/current-tripleo/stable/undercloud.qcow2.md5"],
"delta": "0:00:00.338819", "end": "2017-10-30 10:28:19.249427",
"failed": true, "rc": 22, "start": "2017-10-30 10:28:18.910608",
"stderr": "", "stdout": "", "stdout_lines": [], "warnings":
["Consider using get_url or uri module rather than running curl"]}
PLAY RECAP
*********************************************************************
127.0.0.2 : ok=84 changed=25 unreachable=0
failed=1
localhost : ok=11 changed=4 unreachable=0
failed=0
===================================
At this point the only thing I can do is reprovision the system
(load a fresh OS image and start the whole install again). Have
been trying to find a way of avoiding that, since it takes a
couple hours to run quickstart again (on top of bringing the OS to
a ready state first).
As it is, have yet to make Horizon accessible. I expect it's some simple setup switch I'm missing (it'll be a "duh" moment when I figure that one out).