Ask Your Question
0

Cannot create instances [Devstack + Neutron + Floodlight]

asked 2013-12-13 05:21:49 -0600

muehsi gravatar image

updated 2013-12-16 06:53:23 -0600

Hi folks,

I am using a two node Devstack-setup with Neutron and Floodlight. I tried to use this tutorial: https://wiki.openstack.org/wiki/Neutron/FloodlightPluginSetup The Floodlight-Controller is running on a computer (172.16.150.6). Now I cannot create instances. I am always getting the following error: *Error: Failed to launch instance "vm2": Please try again later [Error: Version 1.9 of Instance is not supported].*

Furthermore, on clicking through the Admin-Panel, I get in System-Info the following error:

Error: 404 Not Found The resource could not be found. There is no data on "Network Agents", not even a empty list.

Compute Services:

nova-conductor  server13    internal    enabled up  0 minutes
nova-cert   server13    internal    enabled up  0 minutes
nova-scheduler  server13    internal    enabled up  0 minutes
nova-consoleauth    server13    internal    enabled up  0 minutes
nova-compute    server01    nova    enabled up  0 minutes

Availability Zones:

Name           Hosts                    Available
internal    server13 (Services Up)        Yes
nova        server01 (Services Up)        Yes

Services:

**Name   Service                       Host             Enabled**
nova     compute                    172.16.34.11        Enabled
neutron  network                    172.16.34.11        Enabled
nova     computev3                  172.16.34.11        Enabled
s3       s3                         172.16.34.11        Enabled
glance   image                      172.16.34.11        Enabled
ec2      ec2                        172.16.34.11        Enabled
keystone identity (native backend)  172.16.34.11        Enabled

Localrc on Controller-Node:

DATABASE_PASSWORD=password
RABBIT_PASSWORD=password
SERVICE_TOKEN=23de8a6122bd3f9afdcf
SERVICE_PASSWORD=74857beec2462815f2a6
ADMIN_PASSWORD=secrete

LOGFILE=/opt/stack/logs/stack.sh.log
VERBOSE=True
LOG_COLOR=True
SCREEN_LOGDIR=/opt/stack/logs


MULTI_HOST=true
HOST_IP=172.16.34.11

# add n-novnc if you want to run nova-compute on the controller node too
ENABLED_SERVICES=g-api,g-reg,key,n-api,n-crt,n-obj,n-sch,n-cauth,horizon,mysql,rabbit,sysstat,n-cond,n-novnc,n-xvnc
disable_service c-api c-sch c-vol cinder
enable_service n-vol

disable_service n-net
enable_service q-svc
enable_service q-dhcp
enable_service q-meta
enable_service neutron

Q_PLUGIN=bigswitch_floodlight
BS_FL_CONTROLLERS_PORT=172.16.150.6:8080
Q_OVS_USE_VETH=true


IMAGE_URLS="https://cloud-images.ubuntu.com/precise/current/precise-server-cloudimg-amd64-disk1.img"

Localrc on compute-Node:

DATABASE_PASSWORD=password
RABBIT_PASSWORD=password
SERVICE_TOKEN=23de8a6122bd3f9afdcf
SERVICE_PASSWORD=74857beec2462815f2a6
ADMIN_PASSWORD=secrete

DATABASE_TYPE=mysql

LOGFILE=/opt/stack/logs/stack.sh.log
VERBOSE=True
LOG_COLOR=True
SCREEN_LOGDIR=/opt/stack/logs

HOST_IP=172.16.34.10
SERVICE_HOST=172.16.34.11
MULTI_HOST=True

ENABLED_SERVICES=n-cpu,n-novnc,neutron

# the following 2 lines are probably not necessary as we configure ovs manually beforehand
Q_PLUGIN=bigswitch_floodlight
Q_HOST=$SERVICE_HOST

enable_service rabbit
enable_service q-l3

MYSQL_HOST=$SERVICE_HOST
RABBIT_HOST=$SERVICE_HOST

VNCSERVER_LISTEN=$HOST_IP
VNCSERVER_PROXYCLIENT_ADDRESS=$HOST_IP

IMAGE_URLS="https://cloud-images.ubuntu.com/precise/current/precise-server-cloudimg-amd64-disk1.img"

Errors on Compute-Node:

stack@server01:~/devstack$ cat /opt/stack/logs/screen-n-cpu.log  | grep 'warning\|error'
2013-12-13 12:11:09.872 DEBUG nova.openstack.common.service [-] fatal_exception_format_errors  = False from (pid=28987) log_opt_values /opt/stack/oslo.config/oslo/config/cfg.py:1979
2013-12-13 12:11:09.899 DEBUG nova.openstack.common.service [-] publish_errors                 = False from (pid=28987) log_opt_values /opt/stack/oslo.config/oslo/config/cfg.py:1979
2013-12-13 12:25:04.977 DEBUG nova.virt.disk.mount.api [req-d0df54a2-24f9-4d7d-977b-2173e723a6bc admin demo] Failed to mount filesystem: Unexpected error while running command.
2013-12-13 12:25:05.050 DEBUG nova.virt.disk.vfs.localfs [req-d0df54a2-24f9-4d7d-977b-2173e723a6bc admin demo] Failed to ...
(more)
edit retag flag offensive close merge delete

Comments

What image are you trying to boot? Did you try the cirros image? Are your nodes VMs?

darragh-oreilly gravatar imagedarragh-oreilly ( 2013-12-13 13:24:02 -0600 )edit

I am using ubuntu image (12.04, server, precise) which was working on one-node setup. I got more problems with the Cirros-Image. My nodes are physical servers.

muehsi gravatar imagemuehsi ( 2013-12-16 06:38:36 -0600 )edit

2 answers

Sort by » oldest newest most voted
1

answered 2013-12-13 08:03:14 -0600

dheeru gravatar image

Assuming that this is devstack. It is failing to mount /opt/stack/data/nova/instances/857f1b0e-11ac-4735-b2ea-c28565855e7d/disk. First check this director exist in system. Due to this other errors coming. Just before the 2013-12-13 12:11:09.872 time there must be 'mount' debug statement. Can you look at the statement ? We can try executing this directly and see what happens.

edit flag offensive delete link more

Comments

Yes, I am using Devstack, sorry. The problem is: Meanwhile I did a new unstack/stack. n-CPU on Compute-Node is now in the article. The folder /opt/stack/data/nova/instances/0b703d6b-436c-4d8d-b5ed-c219d28b0a65/disk exists. Thanks for your help in advance

muehsi gravatar imagemuehsi ( 2013-12-16 06:55:00 -0600 )edit

what it the mount statment you see just in the debug. Can you find out from logs ? OpenStack tries something with this command. I would like to run this command manually and see what happens. You can also ping me some time to troubleshoot this issue.

dheeru gravatar imagedheeru ( 2013-12-16 22:56:17 -0600 )edit

What do you mean by "ping me"?^^ Never done this before.. :D I found this: Command: sudo nova-rootwrap /etc/nova/rootwrap.conf mount /dev/nbd0 /tmp/openstack-vfs-localfsIjIT_Q

muehsi gravatar imagemuehsi ( 2013-12-17 06:10:01 -0600 )edit

When I try to use it, it says: *@server13:~/devstack$ sudo nova-rootwrap /etc/nova/rootwrap.conf mount /dev/nbd0 /tmp/openstack-vfs-localfsIjIT_Q mount: Einhängepunkt /tmp/openstack-vfs-localfsIjIT_Q existiert nicht Means: the mount point /dev/nbd0 /tmp/openstack-vgs-localfslijIT_Q does not exist

muehsi gravatar imagemuehsi ( 2013-12-17 07:35:56 -0600 )edit
0

answered 2013-12-17 10:42:33 -0600

muehsi gravatar image

I manager to get instances booted, I don't exactly know how. At least I added the br-int and Controller-IP (I don't know why, usually I did this before but this was deleted). Furthermore I did a git fetch. Now my nova-Versions are different but I don`t know how up update them. I also added GLANCE_HOSTPORT=$SERVICE_HOST:9292 to my localrc on the compute-Node and moved SERVICE_HOST=... to the top of the localrc.

Nevertheless, my instances still do not get IPs and I still have this error: https://ask.openstack.org/en/question/7073/devstack-havana-with-restproxy-plugin-bigswitch_floodlight-not-able-to-ping/ Most probably it is wrong configured, but I don`t know why...

edit flag offensive delete link more

Your Answer

Please start posting anonymously - your entry will be published after you log in or create a new account.

Add Answer

Get to know Ask OpenStack

Resources for moderators

Question Tools

1 follower

Stats

Asked: 2013-12-13 05:21:49 -0600

Seen: 1,683 times

Last updated: Dec 17 '13