Contrail Platform Developers
Highlighted
Contrail Platform Developers

[Can't Install] 'fab install_contrail' and 'fab setup_all' getting stuck.

[ Edited ]
‎04-09-2014 11:10 AM

I am attempting to install Contrail with two servers (192.168.1.23 and 192.168.1.24), one to act as the single node. I am running into a couple of issues and hoping someone can point me in the right direction. Both servers' OSs are installed from the 'builder-386-x86_64-DVD.iso' image.  In both of th following cases, I gave it a good 30 min before giving up.


'fab install_contrail' gets stuck here. The node itself does successfully reboot but the node which ran the install never seems to notice.

Node (192.168.1.24) is down... Waiting for node to come back

 

After about 5 min I get this, and then nothing:

 

No handlers could be found for logger "paramiko.transport"




After breaking out of that, I try to run 'fab setup_all' and it gets stuck here:

[root@192.168.1.24] out: [root@192.168.1.24] run: source /opt/contrail/api-venv/bin/activate && python /opt/contrail/utils/provision_vrouter.py --host_name host1.aswitchtoofar.com --host_ip 192.168.1.24 --api_server_ip 192.168.1.24 --oper add --admin_user admin --admin_password secret123 --admin_tenant_name admin



Here is my 'testbed.py' file:

cat /opt/contrail/utils/fabfile/testbeds/testbed.py
from fabric.api import env

#Management ip addresses of hosts in the cluster
host1 = 'root@192.168.1.24'

#External routers if any
#for eg.
#ext_routers = [('mx1', '10.204.216.253')]
ext_routers = []

#Autonomous system number
router_asn = 64512

#Host from which the fab commands are triggered to install and provision
host_build = 'root@192.168.1.23'

#Role definition of the hosts.
env.roledefs = {
    'all': [host1],
    'cfgm': [host1],
    'openstack': [host1],
    'control': [host1],
    'compute': [host1],
    'collector': [host1],
    'webui': [host1],
    'database': [host1],
    'build': [host_build],
}

#Openstack admin password
env.openstack_admin_password = 'secret123'

#Hostnames
env.hostnames = {
    'all': ['host1']
}

env.password = 'secret'
#Passwords of each host
env.passwords = {
    host1: 'secret',

    host_build: 'secret',
}

#For reimage purpose
env.ostypes = {
    host1:'centos',
}

#OPTIONAL BONDING CONFIGURATION
#==============================
#Inferface Bonding
#bond= {
#    host1 : { 'name': 'bond0', 'member': ['p2p0p0','p2p0p1','p2p0p2','p2p0p3'], 'mode':'balance-xor' },
#}

#OPTIONAL SEPARATION OF MANAGEMENT AND CONTROL + DATA
#====================================================
#Control Interface
#control = {
#    host1 : { 'ip': '192.168.10.1/24', 'gw' : '192.168.10.254', 'device':'eth0' },
#}

#Data Interface
#data = {
#    host1 : { 'ip': '192.161.10.1/24', 'gw' : '192.161.10.254', 'device':'bond0' },
#}

#To disable installing contrail interface rename package
#env.interface_rename = False

#To enable multi-tenancy feature
#multi_tenancy = True

#To enable haproxy feature
#haproxy = True

#To Enable prallel execution of task in multiple nodes
#do_parallel = True

# To configure the encapsulation priority. Default: MPLSoGRE
#env.encap_priority =  "'MPLSoUDP','MPLSoGRE','VXLAN'"




Thank you for any assistance you can offer!

4 REPLIES 4
Contrail Platform Developers

Re: [Can't Install] 'fab install_contrail' and 'fab setup_all' getting stuck.

‎04-09-2014 11:53 AM

In case it helps:

 

[root@host1 ~]# openstack-status
== Nova services ==
openstack-nova-api:           active
openstack-nova-cert:          active
openstack-nova-compute:       inactive (disabled on boot)
openstack-nova-network:       dead (disabled on boot)
openstack-nova-scheduler:     active
openstack-nova-volume:        dead (disabled on boot)
openstack-nova-conductor:     active
== Glance services ==
openstack-glance-api:         active
openstack-glance-registry:    active
== Keystone service ==
openstack-keystone:           active
== Horizon service ==
openstack-dashboard:          404
== Quantum services ==
quantum-server:               active
== Cinder services ==
openstack-cinder-api:         active
openstack-cinder-scheduler:   active
openstack-cinder-volume:      inactive (disabled on boot)
== Support services ==
mysqld:                       active
httpd:                        active
libvirtd:                     active
qpidd:                        active
memcached:                    active
== Keystone users ==
+----------------------------------+---------+---------+---------------------+
|                id                |   name  | enabled |        email        |
+----------------------------------+---------+---------+---------------------+
| 2bfbdb648b5c4c0ab971dec17ef4d7d0 |  admin  |   True  |  admin@example.com  |
| fbe938a23f3841e1a2837d9d58136157 |  cinder |   True  |  cinder@example.com |
| 261baf081bab498db2458fa2c11e7320 |   demo  |   True  |   demo@example.com  |
| 2747fdb3dccc406f8be69ed60b2761f6 |  glance |   True  |  glance@example.com |
| b83a8db42ab94960819f3993602d3fbd |   nova  |   True  |   nova@example.com  |
| b8a07e5efc5a4d15bde35b66902336e2 | quantum |   True  | quantum@example.com |
+----------------------------------+---------+---------+---------------------+
== Glance images ==
ID                                   Name                           Disk Format          Container Format     Size          
------------------------------------ ------------------------------ -------------------- -------------------- --------------
== Nova instance flavors ==
m1.medium: Memory: 4096MB, VCPUS: 2, Root: 40GB, Ephemeral: 0Gb, FlavorID: 3, Swap: 0MB, RXTX Factor: 1.0, public, ExtraSpecs {}
m1.large: Memory: 8192MB, VCPUS: 4, Root: 80GB, Ephemeral: 0Gb, FlavorID: 4, Swap: 0MB, RXTX Factor: 1.0, public, ExtraSpecs {}
m1.tiny: Memory: 512MB, VCPUS: 1, Root: 0GB, Ephemeral: 0Gb, FlavorID: 1, Swap: 0MB, RXTX Factor: 1.0, public, ExtraSpecs {}
m1.xlarge: Memory: 16384MB, VCPUS: 8, Root: 160GB, Ephemeral: 0Gb, FlavorID: 5, Swap: 0MB, RXTX Factor: 1.0, public, ExtraSpecs {}
m1.small: Memory: 2048MB, VCPUS: 1, Root: 20GB, Ephemeral: 0Gb, FlavorID: 2, Swap: 0MB, RXTX Factor: 1.0, public, ExtraSpecs {}
== Nova instances ==

 

 

 

I suspected that the vRouter part died because compute didnt start up.  Despite horizon showing as '404' I can reach the web dashboard.  'openstack-nova-scheduler' hadnot been enabled in chkconfig.  I enabled and started it and ran 'fab setup_all' again but it made no difference.

Contrail Platform Developers

Re: [Can't Install] 'fab install_contrail' and 'fab setup_all' getting stuck.

‎04-23-2014 11:22 AM

Hi Tom,

  I am assuming you are running fab from /opt/contrail/utils on host1. 

If so, In testbed.py, can you update host_build to be 192.168.1.24 instead of 192.168.1.23  and rerun setup_all ?

 

Also, in the env.hostnames, please specify the output of 'hostname' command . Is it just 'host1' ? 

 

setup_all got stuck mostly because Contrail API Server did not start fully. It requires its dependent processes like redis, database to be up. 

Can you check if redis-config process is up and cassandra is up on port 9160 ? 

Can you attach the output of setup_all also

 

Vedu

Contrail Platform Developers

Re: [Can't Install] 'fab install_contrail' and 'fab setup_all' getting stuck.

‎06-02-2014 08:53 AM

Also, in the env.hostnames, please specify the output of 'hostname' command . Is it just 'host1' ?

 

Vedu


 

This was the answer.  After adding the FQDN to my local hosts file rather than only "host1" the fab install completed.

Contrail Platform Developers

Re: [Can't Install] 'fab install_contrail' and 'fab setup_all' getting stuck.

‎02-02-2015 08:14 AM

Hi there,

 

Same situation for me - I am stuck at the last process where vrouter.py provisions happens - I am trying to run that in ubuntu server 12.04. Any help would be appriciated!!

 

Thanks,

Furqan