Ask Your Question

agti's profile - activity

2020-05-03 12:26:03 -0500 received badge  Famous Question (source)
2020-05-03 12:26:03 -0500 received badge  Notable Question (source)
2017-10-13 07:30:35 -0500 marked best answer Heat resource OS::Heat::SoftwareDeployment - create in progress

Hi, I'm trying to build a heat template with SoftwareConfig and SoftwareDeployment resources; it correctly creates all resources but SoftwareDeployment stays on a 'Create in progress' status; I could not see any relevant error on heat logs on controller node.

Here a sample heat template illustrating my case:

    type: OS::Nova::Server
      name: dns
      image: vm-base
      flavor: m1.small
      key_name: { get_param: KeyName }
        - port: { get_resource: dns_port }
    type: OS::Neutron::Port
      network_id: { get_param: PrivateNetworkId }
      security_groups: [get_resource: total_sg]
        - subnet_id: { get_param: PrivateSubNetworkId }
    type: OS::Nova::Server
      name: vm
      image: cw-base
      flavor: m1.small
      key_name: { get_param: KeyName }
      user_data_format: SOFTWARE_CONFIG
        - port: { get_resource: vm_port }
    type: OS::Neutron::Port
      network_id: { get_param: PrivateNetworkId }
      security_groups: [get_resource: total_sg]
        - subnet_id: { get_param: PrivateSubNetworkId }
    type: OS::Heat::SoftwareConfig
      - name: previous
        default: 'NONE'
      - name: dns_ip
        default: {get_attr: [dns, first_address]}
      group: script
      config: |
        echo "DNS_IP = $dns_ip" >> /home/ubuntu/dns_ip_software_config
        echo "Previous: $previous"
        echo "${deploy_resource_name} is running on $(hostname) at $(date)"
    type: OS::Heat::SoftwareDeployment
        get_resource: vm_config
        get_resource: vm

Where could I find any interesting log showing me why the process does not complete? Am I wrong in template creation?


No way to let it work... is it available a reference guide on how to use it? Could it be a problem on my infrastructure? Do I have to use a particular cloud image?


I'm trying with a new stack, directly cloned from Github. I have the same problem with this stack, so I imagine I am missing something. Is there any special configuration of heat I need to switch on? Does the image I use need a special package/heat-handler/whatsoever ?

Thanks in advance!

2017-09-01 16:27:10 -0500 received badge  Famous Question (source)
2017-09-01 16:27:10 -0500 received badge  Notable Question (source)
2017-07-27 06:56:40 -0500 received badge  Enlightened (source)
2017-07-27 06:56:40 -0500 received badge  Good Answer (source)
2017-03-20 04:12:12 -0500 marked best answer RDO - Nagios GUI not available

Hi, I recently installed RDO juno in 3 node setup (CentOS7) with CONFIG_NAGIOS_INSTALL=y correctly set in answer-file; packstack process ended correctly showing horizon GUI and nagios GUI addresses. The former works correctly, allowing me to spawn VMs and to work with Openstack; the latter is not available, showing me the following error:

The requested URL /nagios was not found on this server.

I do not know why packstack ended correctly, even if nagios GUI seems KO. In the controller node nagios is up and running:

service nagios status
nagios (pid 25280) is running...

Is anyone experiencing the same problem or has any hint on how to resolve it? Thanks


No way to let it work, I do not know how to access Nagios GUI

2016-12-20 07:21:39 -0500 received badge  Famous Question (source)
2016-12-20 07:19:53 -0500 received badge  Great Question (source)
2016-09-02 01:54:32 -0500 received badge  Popular Question (source)
2016-08-31 16:20:16 -0500 received badge  Notable Question (source)
2016-08-30 08:42:43 -0500 commented question Cloudkitty CRITICAL Error: cloudkitty [-] MissingAuthPlugin: An auth plugin is required to determine endpoint URL

Here the bug referring to the problem

2016-08-30 08:15:48 -0500 commented question Cloudkitty CRITICAL Error: cloudkitty [-] MissingAuthPlugin: An auth plugin is required to determine endpoint URL

Seems there is a bug fix request open. Ceilometer client can't find the keystone auth plugin.

2016-08-30 01:54:29 -0500 received badge  Popular Question (source)
2016-08-30 01:53:55 -0500 commented question Cloudkitty launch error on Mitaka

I have the same error (MissingAuthPlugin: An auth plugin is required to determine endpoint URL). [keystone_fetcher] url and [keystone_authtoken] auth_uri point to public keysone endpoint.

2016-08-30 01:31:43 -0500 received badge  Popular Question (source)
2016-08-29 10:56:01 -0500 asked a question Cloudkitty CRITICAL Error: cloudkitty [-] MissingAuthPlugin: An auth plugin is required to determine endpoint URL

Hi, I'm getting the following error after executing cloudkitty-api following the official guide (I'm working on Ubuntu 16.04 and Mitaka rel).

root@ubuntu-VM:~# cloudkitty-api --config-file /etc/cloudkitty/cloudkitty.conf
2016-08-29 17:48:41.334 4679 INFO [-] Starting server in PID 4679
2016-08-29 17:48:41.335 4679 INFO [-] Configuration:
2016-08-29 17:48:41.335 4679 INFO [-] ********************************************************************************
2016-08-29 17:48:41.336 4679 INFO [-] Configuration options gathered from:
2016-08-29 17:48:41.336 4679 INFO [-] command line args: ['--config-file', '/etc/cloudkitty/cloudkitty.conf']
2016-08-29 17:48:41.336 4679 INFO [-] config files: ['/etc/cloudkitty/cloudkitty.conf']
2016-08-29 17:48:41.336 4679 INFO [-] ================================================================================
2016-08-29 17:48:41.336 4679 INFO [-] api_paste_config               = api_paste.ini
2016-08-29 17:48:41.337 4679 INFO [-] auth_strategy                  = keystone
2016-08-29 17:48:41.337 4679 INFO [-] config_dir                     = None
2016-08-29 17:48:41.337 4679 INFO [-] config_file                    = ['/etc/cloudkitty/cloudkitty.conf']
2016-08-29 17:48:41.337 4679 INFO [-] control_exchange               = openstack
2016-08-29 17:48:41.337 4679 INFO [-] debug                          = False
2016-08-29 17:48:41.337 4679 INFO [-] default_log_levels             = ['amqp=WARN', 'amqplib=WARN', 'boto=WARN', 'qpid=WARN', 'sqlalchemy=WARN', 'suds=INFO', 'oslo.messaging=INFO', 'iso8601=WARN', 'requests.packages.urllib3.connectionpool=WARN', 'urllib3.connectionpool=WARN', 'websocket=WARN', 'requests.packages.urllib3.util.retry=WARN', 'urllib3.util.retry=WARN', 'keystonemiddleware=WARN', 'routes.middleware=WARN', 'stevedore=WARN', 'taskflow=WARN', 'keystoneauth=WARN', 'oslo.cache=INFO', 'dogpile.core.dogpile=INFO']
2016-08-29 17:48:41.338 4679 INFO [-] host                           = ubuntu-VM
2016-08-29 17:48:41.338 4679 INFO [-] instance_format                = [instance: %(uuid)s] 
2016-08-29 17:48:41.338 4679 INFO [-] instance_uuid_format           = [instance: %(uuid)s] 
2016-08-29 17:48:41.338 4679 INFO [-] log_config_append              = None
2016-08-29 17:48:41.338 4679 INFO [-] log_date_format                = %Y-%m-%d %H:%M:%S
2016-08-29 17:48:41.338 4679 INFO [-] log_dir                        = None
2016-08-29 17:48:41.338 4679 INFO [-] log_file                       = None
2016-08-29 17:48:41.339 4679 INFO [-] logging_context_format_string  = %(asctime)s.%(msecs)03d %(process)d %(levelname)s %(name)s [%(request_id)s %(user_identity)s] %(instance)s%(message)s
2016-08-29 17:48:41.339 4679 INFO [-] logging_debug_format_suffix    = %(funcName)s %(pathname)s:%(lineno)d
2016-08-29 17:48:41.339 4679 INFO [-] logging_default_format_string  = %(asctime)s.%(msecs)03d %(process)d %(levelname)s %(name)s [-] %(instance)s%(message)s
2016-08-29 17:48:41.339 4679 INFO [-] logging_exception_prefix       = %(asctime)s.%(msecs)03d %(process)d ERROR %(name)s %(instance)s
2016-08-29 17:48:41.339 4679 INFO [-] logging_user_identity_format   = %(user)s %(tenant)s %(domain)s %(user_domain)s %(project_domain)s
2016-08-29 17:48:41.339 4679 INFO [-] publish_errors                 = False
2016-08-29 17:48:41.339 4679 INFO [-] rpc_backend                    = rabbit
2016-08-29 ...
2016-08-26 07:22:25 -0500 asked a question OVS linux-br switch on default installation guide Ubuntu

I was wondering if there is any specific reason behind the switch from OVS to linux-br on default OpenStack installation guide from Kilo to Liberty (and then Mitaka).

Thanks for any suggestion.

2016-08-25 06:30:35 -0500 asked a question ERROR (AttributeError): 'Namespace' object has no attribute 'os_project_domain_id'

Hi, I'm using Devstack on a Ubuntu 16.04 Virtualbox VM. The installation process worked fine, I could access Horizon and create VMs. The strange thing is that I am unable to use CLI commands, getting this weird error:

ubuntu@ubuntu-VM:~/devstack/devstack/accrc/admin$ pwd

ubuntu@ubuntu-VM:~/devstack/devstack/accrc/admin$ cat admin 
# OpenStack USER ID = 8a8ec35c04b541f8bc9b16bda1692d05
export OS_USERNAME="admin"
# OpenStack project ID = ef38d04b89e046048d5016f0b1fd7f7c
export OS_PROJECT_NAME="admin"
export OS_AUTH_URL=""
export OS_CACERT=""
export NOVA_CERT="/home/ubuntu/devstack/devstack/accrc/cacert.pem"
export OS_AUTH_TYPE=v2password
export OS_PASSWORD="secret"
export OS_USER_DOMAIN_ID=default
export OS_PROJECT_DOMAIN_ID=default

ubuntu@ubuntu-VM:~/devstack/devstack/accrc/admin$ source admin 

ubuntu@ubuntu-VM:~/devstack/devstack/accrc/admin$ nova list
ERROR (AttributeError): 'Namespace' object has no attribute 'os_project_domain_id'

I tried commenting the unset variable statements on adminrc file but nothing changed. Any hint on what is wrong is appreciated.

2016-05-05 12:02:40 -0500 marked best answer heat circular exception - get server ip in user_data

Hi all, I'm trying to test heat templates, but when I reference the IP of a server in the user_data definition of the server itself I see a CircularDependencyException...

Which is the best way to do what I have in mind (basically modify a config file on my server with the actual IP). Here the snippet related with the error:

    type: OS::Nova::Server
      name: serv
      image: image-new
      flavor: m1.small
      key_name: { get_param: KeyName }
          template: |                                                   
            #!/bin/bash -x
        echo hola >> /home/ubuntu/hola.txt
            sudo sed -i s/\<privateIP\>/"%server_ip%"/g /etc/hola/config
            "%server_ip%": {get_attr: [serv, first_address]}         
        - port: { get_resource: server_port }


Is there a way to extract thos info directly from the heat template?

2016-02-25 02:37:26 -0500 received badge  Famous Question (source)
2016-02-23 01:08:03 -0500 received badge  Nice Answer (source)
2016-01-28 02:05:36 -0500 received badge  Notable Question (source)
2016-01-27 08:21:41 -0500 received badge  Popular Question (source)
2016-01-26 08:50:41 -0500 received badge  Notable Question (source)
2016-01-26 08:42:00 -0500 asked a question Multiple neutron plugins

Is it possible to install many plugins on the same neutron installation? Let's suppose I want to use standard OVS approach for a subset of compute nodes, while leveraging networking through Opencontrail on another subnet of compute nodes. Would it be possible?

Thanks for any clue on this topic!

2015-12-09 17:25:08 -0500 received badge  Famous Question (source)
2015-12-09 17:25:08 -0500 received badge  Notable Question (source)
2015-12-09 17:25:08 -0500 received badge  Popular Question (source)
2015-10-22 12:47:07 -0500 marked best answer heat ERROR: You are not authorized to complete this action.

Trying to create a heat-stack I found the following error:

ERROR: You are not authorized to complete this action.

If I perform a simple query I can see I have the heat_stack_owner, heat_stack_user roles...

[root@juno-controller heat-tests(keystone_testadmin)]# keystone user-role-list
|                id                |       name       |             user_id              |            tenant_id             |
| 9fe2ff9ee4384b1894a90878d3e92bab |     _member_     | 230526d5be9c4bb5b72ca8bb5265c458 | 2de156ba27ae404db835d4e955f4f64e |
| f473f1a830c14159833a3ef17aed3cf1 |      admin       | 230526d5be9c4bb5b72ca8bb5265c458 | 2de156ba27ae404db835d4e955f4f64e |
| 84374944abe94ea58f41c59dd3297f58 | heat_stack_owner | 230526d5be9c4bb5b72ca8bb5265c458 | 2de156ba27ae404db835d4e955f4f64e |
| 77ca720b9af24d6bab52b720ac37268a | heat_stack_user  | 230526d5be9c4bb5b72ca8bb5265c458 | 2de156ba27ae404db835d4e955f4f64e |

Am I missing anything? Thanks in advance

2015-10-22 12:47:07 -0500 received badge  Self-Learner (source)
2015-10-16 05:51:11 -0500 received badge  Famous Question (source)
2015-10-14 15:34:32 -0500 received badge  Popular Question (source)
2015-09-29 09:45:06 -0500 marked best answer heat - access created vm - permission denied (publickey)


trying to ssh to a vm created through heat I found this error: Permission denied (publickey)

The strange thing is that if I manually start an identical VM (same image, flavor, security_group, keypair, network) I could perfectly access the VM.

Anyone has encountered this issue before?


ISSUE UPDATE I could not explain myself why VM creation works differently through heat... Is there anything I missed?

ISSUE UPDATE2 I tried the same heat template with a cirros image, and I can access to the VM with the key. The not-working image is an ubuntu12.04 with cloud-init (image downloaded from )

ISSUE UPDATE3 I changed approach, trying to create the keypair at runtime, during stack creation. The key is correctly created and I print the private key as template output. When I use the key to access the VM I get same error as before...

    type: OS::Nova::KeyPair
      name: test-kp
      save_private_key: True

    description: keypair private key
    value: { get_attr: [ test-kp, private_key ] }

ISSUE UPDATE4 Same problem with centos template

It seems like the key is not correctly injected when deploying through Heat...

2015-09-14 01:58:54 -0500 received badge  Notable Question (source)
2015-09-14 00:24:36 -0500 received badge  Popular Question (source)
2015-09-11 09:18:40 -0500 asked a question Ping issue with virtual router

Hi all, I'm trying to use a virtual router inside my Openstack tenant. I created an instance of the router (Vyatta5600) and connected it to two different networks (dc1 - and dc2 -

image description

I then connected two different cirros clients on those networks, setting the corresponding Vyatta interface as their deafult GW. I use default security group configured to allow all inbound/outbound traffic (ICMP, TCP,UDP). From both clients I could ping the default GW.

Pinging from one client to the other one does not work, nor do pinging the vyatta interface on the opposite network.

Is there anything special I should enable at the OS level.



Same thing happens if I change the vyatta instance with a Linux instance acting as a router (as explained in this link)


In this post i found this:

Be aware that the security groups implementation places anti-spoofing rules to prevent a VM sending packets that do not have the source mac or IP address that Neutron assigned to it. Do iptables-save on the compute nodes to see them.

Could it be this thecause of the problem? If yes, is there the possibility to change that behaviour?

2015-08-30 18:25:37 -0500 received badge  Famous Question (source)
2015-08-25 07:00:49 -0500 asked a question RDO - Qrouters lose IP on public network

Hi all, I would ask If this is a correct behaviour for Juno RDO:

for testing purposes I shut down all my infrastructure blades in order to test how my environment react to reboots; all worked as expected, but floating IP functionality do not. It seems qrouter lose their IP on public network whenever the network node reboot. I overcome this problem clearing up the router gateway and setting it again.

Is this the correct behaviour? Is anyone experiencing the same problem?

2015-08-25 06:52:58 -0500 received badge  Notable Question (source)
2015-07-29 12:51:16 -0500 received badge  Famous Question (source)