Note: This bug is displayed in read-only format because the product is no longer active in Red Hat Bugzilla.

Bug 1097350

Summary: [Rubygem-Staypuft]: Cannot Deploy LVM Block Storage using staypuft, the YAML is getting wrong parameters (Including wrong default IP: controller_priv_host: 172.16.0.1)
Product: Red Hat OpenStack Reporter: Omri Hochman <ohochman>
Component: rubygem-staypuftAssignee: Scott Seago <sseago>
Status: CLOSED WONTFIX QA Contact: Omri Hochman <ohochman>
Severity: high Docs Contact:
Priority: urgent    
Version: 4.0CC: aberezin, mburns, mlopes, morazi, ohochman, sclewis
Target Milestone: z4Keywords: TestOnly, ZStream
Target Release: 4.0   
Hardware: x86_64   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Previously, non-HA Block Storage (cinder) configuration reused the same 'cinder' service for the LVM and controller nodes. However, the code did not handle the duplicated/shared service as expected. In addition, certain parameters required for the LVM node were not listed. Consequently, multiple necessary configuration parameters for the LVM node were not being set. This update addresses this issue by creating separate service objects for the cinder node and controller; cinder on the HA controller was already handled separately. As a result, LVM storage nodes are now able to be correctly configured.
Story Points: ---
Clone Of: Environment:
Last Closed: 2014-07-15 16:59:36 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Bug Depends On:    
Bug Blocks: 1093126    

Description Omri Hochman 2014-05-13 15:36:04 UTC
[Rubygem-Staypuft]: Cannot Deploy LVM Block Storage using staypuft, the YAML is getting wrong parameters (Including wrong default IP: controller_priv_host: 172.16.0.1)


Environment  (puddle : OpenStack/Foreman/2014-05-06.6/) 
-------------------------------------------------------
ruby193-rubygem-staypuft-0.0.12-1.el6ost.noarch
foreman-installer-staypuft-0.0.10-1.el6ost.noarch
openstack-puppet-modules-2013.2-9.1.el6ost.noarch
puppet-server-3.3.2-2.el6.noarch
puppet-3.3.2-2.el6.noarch
foreman-1.5.0.22-1.el6sat.noarch


Steps: 
-------
(1) Create a new deployment --> set right all service configuration including Cinder's (Note: there are missing parameters at the cinder section). 
(2) Associate host with LVM Block Storage.
(3) Attempt to deploy that environment. 
(4) SSH to the HOST with the LVM Block Storage
(5) Check for: 'openstack-status' 

Results: 
---------

(1) All Cinder Services are Inactive :

[root@001a4a1696d5 ~]# openstack-status 
== Cinder services ==
openstack-cinder-api:                   inactive  (disabled on boot)
openstack-cinder-scheduler:             inactive  (disabled on boot)
openstack-cinder-volume:                dead
== Support services ==
messagebus:                             active
tgtd:                                   inactive  (disabled on boot)

(2) Browse the Staypuft GUI --> Edit the host -->  Check the YAML file
(there is a button on the left hand side that says yaml)
 
YAML View: 
------------
---
classes:
  quickstack::storage_backend::lvm_cinder:
    cinder_backend_gluster: 'false'
    cinder_backend_iscsi: 'false'
    cinder_db_password: de998d0a770404c13595a576dc8bf126
    cinder_gluster_peers: []
    cinder_gluster_volume: cinder
    cinder_iscsi_iface: eth1
    cinder_iscsi_network: ''
    controller_priv_host: 172.16.0.1
    mysql_ca: /etc/ipa/ca.crt
    mysql_host: 172.16.0.1
    qpid_host: 172.16.0.1
    qpid_password: bcdc6056ae5c808329118c8474743542
    qpid_username: openstack
    ssl: 'false'
    verbose: 'true'
parameters:
  puppetmaster: oh-havana-foreman.scl.lab.tlv.redhat.com
  domainname: Default domain used for provisioning
  hostgroup: base/omri-deployment6/LVM Block Storage
  root_pw: $1$WgX0moJ9$EQWos50SdNv1GkrOaQFvV0
  puppet_ca: oh-havana-foreman.scl.lab.tlv.redhat.com
  foreman_env: production
environment: production

Comment 5 Mike Orazi 2014-05-15 14:09:40 UTC
Can we get some additional info on whether this is HA or non-HA deployments (or both)

Comment 6 Scott Seago 2014-05-15 16:13:38 UTC
OK I"m slightly unclear on some of the details. Lets start with the default IP comment: "Including wrong default IP: controller_priv_host: 172.16.0.1" -- I think that is the default value, but I wouldn't call the default IP right or wrong -- the defaults always need to be customized to your specific environment.

Did you set this value to the proper (non-default) one for your env when you set up your deployment params?

Mike -- I think the LVM block storage node is the same in HA and non-HA.

Also, for this particular parameter, it looks like I've set it to nil. I think this is what either Mike (Burns) or Jay told me to set it to in this case.

Comment 7 Omri Hochman 2014-05-15 16:54:14 UTC
(In reply to Scott Seago from comment #6)
> OK I"m slightly unclear on some of the details. Lets start with the default
> IP comment: "Including wrong default IP: controller_priv_host: 172.16.0.1"
> -- I think that is the default value, but I wouldn't call the default IP
> right or wrong -- the defaults always need to be customized to your specific
> environment.
> 
> Did you set this value to the proper (non-default) one for your env when you
> set up your deployment params?
> 
> Mike -- I think the LVM block storage node is the same in HA and non-HA.
> 
> Also, for this particular parameter, it looks like I've set it to nil. I
> think this is what either Mike (Burns) or Jay told me to set it to in this
> case.


As explained over IRC : When using the following RPM's - there is no way to change the Cinder default - because they are missing from the Staypuft-GUI. According to that - there's no way to change the wrong default values that are getting in the YAML file.  

The source of the value - 'controller_priv_host: 172.16.0.1' is unknown to me (since I have change anywhere in the GUI parameters to valid controller IP) - It might be hard coded. 

ruby193-rubygem-staypuft-0.0.12-1.el6ost.noarch
foreman-installer-staypuft-0.0.10-1.el6ost.noarch
openstack-puppet-modules-2013.2-9.1.el6ost.noarch
puppet-server-3.3.2-2.el6.noarch
puppet-3.3.2-2.el6.noarch
foreman-1.5.0.22-1.el6sat.noarch

Comment 8 Scott Seago 2014-05-15 17:02:42 UTC
Actually, from the sound of it this is the fix from pchalupa that's already on master, but it's apparently not yet on the latest puddle.

Comment 9 Scott Seago 2014-05-16 14:18:06 UTC
This is fixed by the same change as https://bugzilla.redhat.com/show_bug.cgi?id=1095816

Comment 11 Omri Hochman 2014-05-27 21:32:28 UTC
Verified ruby193-rubygem-staypuft-0.0.16-1.el6ost.noarch :

---
classes:
  quickstack::storage_backend::lvm_cinder:
    cinder_backend_gluster: 'false'
    cinder_backend_iscsi: 'false'
    cinder_db_password: 0b5fe6a3f4f9b0630f8cf49799fa298c
    cinder_gluster_peers: []
    cinder_gluster_volume: cinder
    cinder_iscsi_iface: eth1
    cinder_iscsi_network: ''
    controller_priv_host: 192.168.0.5
    mysql_ca: /etc/ipa/ca.crt
    mysql_host: 192.168.0.5
    qpid_host: 192.168.0.5
    qpid_password: dbedd6b61a81d2b01811e41d513ae9a6
    qpid_username: openstack
    ssl: 'false'
    verbose: 'true'
parameters:
  puppetmaster: oh-havana-foreman.scl.lab.tlv.redhat.com
  domainname: Default domain used for provisioning
  hostgroup: base/omri-deployment2/LVM Block Storage
  root_pw: $1$Nf3Ylc+b$mp3FYWDM1KabZZwxc9qyG/
  puppet_ca: oh-havana-foreman.scl.lab.tlv.redhat.com
  foreman_env: production
  owner_name: Admin User
  owner_email: root.tlv.redhat.com
environment: production