** Changed in: tripleo
Status: Incomplete => Triaged
** Summary changed:
- Overcloud deployment fails when node introspection is enabled
+ Errors when starting introspection are silently ignored
** Changed in: tripleo
Importance: Undecided => High
** Changed in: tripleo
Assignee:
After some additional testing, i'd say that this issue may be related to
error handling when running introspection bulk and one of the
introspection fails.
In this case one of them is failing because of
https://bugs.launchpad.net/tripleo/+bug/1651719 but "openstack baremetal
introspection bulk sta
I'm following the tripleo-quickstart workflow.
1. Create VMs in virthost, one for undercloud, one for controller and one for
compute
2. I deployed undercloud
3. Run following commands:
openstack baremetal import --json instackenv.json
openstack baremetal configure boot
openstack baremetal intros
This seems to be a problem:
| provision_state| manageable
I wonder if something is wrong with the workflows. Which command did you
use for introspection?
** Changed in: nova
Status: New => Invalid
** Changed in: ironic-inspector (Ubuntu)
Status: New => Invalid
** Changed
I've reproduced this locally running RDO from master. I have a ironic
installation (a tripleo undercloud), and i've registered a node and
instrospect it:
This is how the node looks like from ironic perspective:
$ openstack baremetal show e14be55d-8dd9-4b08-aa5d-efab2b5a5c01
This command is deprec
Checking nova-compute.log it seems it's not a matter of waiting, see:
2016-12-21 06:38:28.987 23446 INFO nova.compute.resource_tracker
[req-4ec06e18-d31c-4254-8bc1-9cad7fb8b5e8 - - - - -] Final resource view:
name=c6f8f4ba-9c7c-4c87-b95a-67a5861b7bec phys_ram=0MB used_ram=0MB
phys_disk=0GB used
So, if we focus in the node where introspection run properly (c6f8f4ba-
9c7c-4c87-b95a-67a5861b7bec), we can see following info:
1. Introspection finishes successfully with right information according to
inspector logs:
2016-12-21 06:41:24.030 24006 INFO ironic_inspector.plugins.standard [-] [n
I had reported the db integrity error in a separated LP
https://bugs.launchpad.net/tripleo/+bug/1651719 .
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1651704
Title:
Overcloud deployment fails when
I'm also marking ironic-inspector as affected here because in the logs I
see (thanks to bfournier for pointing it out):
2016-12-21 06:39:16.254 24006 ERROR ironic_inspector.node_cache [-] [node:
f06c26d1-8008-47a9-a687-d5ef05083e41 state None] Database integrity error
(pymysql.err.IntegrityError
