1

i am trying to add additional compute node on different virtual machine to the pre-installed openstack. I disabled the firewall services,enable to ping other virtual machine.. but still compute node is not able to register with Rabbitmq service running on controller node.. Here it is my nova.conf file... 

[DEFAULT]
dhcpbridge_flagfile=/etc/nova/nova.conf
dhcpbridge=/usr/bin/nova-dhcpbridge
state_path=/var/lib/nova
lock_path=/var/lock/nova
force_dhcp_release=True
iscsi_helper=tgtadm
libvirt_use_virtio_for_bridges=True
connection_type=libvirt
root_helper=sudo nova-rootwrap /etc/nova/rootwrap.conf
verbose=True
ec2_private_dns_show_ip=True
api_paste_config=/etc/nova/api-paste.ini
volumes_path=/var/lib/nova/volumes
enabled_apis=ec2,osapi_compute,metadata
rpc_backend = rabbit
auth_strategy = keystone
use_neutron = True
firewall_driver = nova.virt.firewall.NoopFirewallDriver
my_ip = #compute node ip
rabbit_host= #controller_node_ip
rabbit_port = 5672
rabbit_userid = stackrabbit
rabbit_password = devstack
rabbit_use_ssl = False
rabbit_virtual_host=/

[keystone_authtoken]
auth_uri = http://controller_node_ip:5000
auth_url = http://controller_node_ip:35357
memcached_servers = controller_node_ip:11211
auth_type = password
project_domain_name = default
user_domain_name = default
project_name = service
username = nova
password = devstack
auth_host = controller_node_ip
auth_port = 35357
auth_protocol = http

[vnc]
enabled = True
vncserver_listen = 0.0.0.0
vncserver_proxyclient_address = $my_ip
novncproxy_base_url = http://controller_node_ip:6080/vnc_auto.html

[glance]
api_servers = http://controller_node_ip:9292

[oslo_concurrency]
lock_path = /var/lib/nova/tmp

Here it is my nova-compute.log:

2016-09-20 19:08:57.701 7201 INFO oslo.messaging._drivers.impl_rabbit [-] Reconnecting to AMQP server on localhost:5672
2016-09-20 19:08:57.701 7201 INFO oslo.messaging._drivers.impl_rabbit [-] Delaying reconnect for 1.0 seconds...
2016-09-20 19:08:58.708 7201 ERROR oslo.messaging._drivers.impl_rabbit [-] AMQP server on localhost:5672 is unreachable: [Errno 111] ECONNREFUSED. Trying again in 30 seconds...

Please suggest me something so that i can resolve this issue... Thank you in advance...

Suneha
  • 121
  • 2
  • 4
  • 10
  • Troubleshoot - 1) Telnet and verify the rabbitmq server running in controller machine is accessible from compute node, 2) Restart rabbitmq server so that compute publisher can register it's connection. – Jayaprakash Sep 23 '16 at 10:08
  • @ Jayprakash thank you for the solution... yeah..i have done this.. and this issue is resolved...but now i m getting a new error: 2016-09-23 16:26:40.305 17907 TRACE nova.openstack.common.threadgroup libvirtError: internal error: client socket is closed – Suneha Sep 23 '16 at 12:08

1 Answers1

0

I encountered this when expanding my nova-compute estate (although I'm not using Devstack).

In my newly created compute server, the following was seen in /var/log/nova/nova-compute.log : -

2017-11-14 11:40:53.287 52408 ERROR oslo.messaging._drivers.impl_rabbit [req-adfd6dc7-fe8c-4de5-8401-58d325c3b4a8 - - - - -] [be6e0302-dfc8-4512-8b48-0d824fc6ea14] AMQP server on 127.0.0.1:5672 is unreachable: [Errno 111] ECONNREFUSED. Trying again in 1 seconds. Client port: None

The solution was quite simple. I checked /var/log/sysinfo (I run ubuntu; /var/log/messages for those on Redhat systems) and could see the following lines:-

Nov 14 12:01:48 compute2 systemd[1]: Started OpenStack Compute.
Nov 14 12:01:49 compute2 nova-compute[3222]: Traceback (most recent call last):
Nov 14 12:01:49 compute2 nova-compute[3222]:   File "/usr/bin/nova-compute", line 10, in <module>
Nov 14 12:01:49 compute2 nova-compute[3222]:     sys.exit(main())
Nov 14 12:01:49 compute2 nova-compute[3222]:   File "/usr/lib/python2.7/dist-packages/nova/cmd/compute.py", line 42, in main
Nov 14 12:01:49 compute2 nova-compute[3222]:     config.parse_args(sys.argv)
Nov 14 12:01:49 compute2 nova-compute[3222]:   File "/usr/lib/python2.7/dist-packages/nova/config.py", line 52, in parse_args
Nov 14 12:01:49 compute2 nova-compute[3222]:     default_config_files=default_config_files)
Nov 14 12:01:49 compute2 nova-compute[3222]:   File "/usr/lib/python2.7/dist-packages/oslo_config/cfg.py", line 2355, in __call__
Nov 14 12:01:49 compute2 nova-compute[3222]:     self._namespace._files_permission_denied)
Nov 14 12:01:49 compute2 nova-compute[3222]: oslo_config.cfg.ConfigFilesPermissionDeniedError: Failed to open some config files: /etc/nova/nova.conf
Nov 14 12:01:49 compute2 systemd[1]: nova-compute.service: Main process exited, code=exited, status=1/FAILURE

Which shows that my /etc/nova/nova.conf file was unreadable. It turns out this was because I used scp to copy the nova.conf from my first compute to my new machine, and the file was read-only to the root user. The solution was to (on my new compute)

cd /etc/nova/
chown nova:nova nova.conf
service nova-compute restart
bthomas
  • 1
  • 1