rdo-hangout-multinode-packstack-slides

View Github Repository Open presentation in a new window

larsks

See all presentation from larsks

rdo-hangout-multinode-packstack-slides

5 5

rdo-hangout-multinode-packstack-slides

On Github larsks / rdo-hangout-multinode-packstack-slides

RDO Hangout: Multinode OpenStack with Packstack

February 27, 2014Lars Kellogg-Stedman lars@redhat.com

http://goo.gl/Yvmd0P

What are we going to do today?

A multinode OpenStack install using packstack

  • One controller
  • One network host
  • Two compute hosts

Our tools

What is packstack?

A command-line tool for automating the deployment of simple OpenStack clouds.

  • Single host (--allinone) or multinode
  • Proof of Concept ("PoC") deployments

Not really designed for production use:

  • Servers must be pre-installed
  • No high-availability
  • No scalability
  • No shared filesystems

Supported platforms

  • RHEL
  • CentOS
  • Fedora

Architecture Overview

Specifically:

  • Neutron networking
  • OpenVswitch plugin
  • GRE tenant networks

Controller host

Network host

Compute host

Getting started

Set up ssh

Make sure that you can ssh as root to all your hosts from wherever you're running packstack.

Install packstack

Make the RDO repositories available:

# yum install -y http://rdo.fedorapeople.org/rdo-release.rpm

And install packstack:

# yum -y install openstack-packstack

The answers file

You can set all sorts of parameters on the command line...

packstack --allinone --os-quantum-install=y --provision-demo=n \
  --provision-all-in-one-ovs-bridge=n

...but I like to generate an "answers" file and edit it:

# packstack --gen-answer-file packstack-answers.txt

For this hangout, our packstack-answers.txt file differs from the default like this:

CONFIG_CEILOMETER_INSTALL=n
CONFIG_NOVA_COMPUTE_HOSTS=10.15.0.2,10.15.0.8
CONFIG_NEUTRON_SERVER_HOST=10.15.0.7
CONFIG_NEUTRON_L3_HOSTS=10.15.0.7
CONFIG_NEUTRON_DHCP_HOSTS=10.15.0.7
CONFIG_NEUTRON_LBAAS_HOSTS=10.15.0.7
CONFIG_NEUTRON_METADATA_HOSTS=10.15.0.7
CONFIG_NEUTRON_OVS_TENANT_NETWORK_TYPE=gre
CONFIG_NEUTRON_OVS_TUNNEL_RANGES=1000:3000
CONFIG_NEUTRON_OVS_TUNNEL_IF=eth2

CONFIG_NEUTRON_OVS_TUNNEL_IF is used to configure the source ip address for GRE tunnels via the Facter fact ipaddress_<interfacename> (e.g., ipaddress_eth2).

Run packstack

# packstack --answer-file packstack-answers.txt
Welcome to Installer setup utility
Packstack changed given value  to required value /root/.ssh/id_rsa.pub

Installing:
Clean Up...                                            [ DONE ]
Setting up ssh keys...                                 [ DONE ]
Discovering hosts' details...                          [ DONE ]
Adding pre install manifest entries...                 [ DONE ]
Adding MySQL manifest entries...                       [ DONE ]
Adding QPID manifest entries...                        [ DONE ]
Adding Keystone manifest entries...                    [ DONE ]
Adding Glance Keystone manifest entries...             [ DONE ]
Adding Glance manifest entries...                      [ DONE ]
Installing dependencies for Cinder...                  [ DONE ]
Adding Cinder Keystone manifest entries...             [ DONE ]
.
.
.
Finalizing...                                          [ DONE ]

 **** Installation completed successfully ******


Additional information:
 * Time synchronization installation was skipped. Please note that unsynchronized time on server instances might be problem for some OpenStack components.
 * Did not create a cinder volume group, one already existed
 * File /root/keystonerc_admin has been created on OpenStack client host 10.15.0.4. To use the command line tools you need to source the file.
 * To access the OpenStack Dashboard browse to http://10.15.0.4/dashboard.
Please, find your login credentials stored in the keystonerc_admin in your home directory.
 * The installation log file is available at: /var/tmp/packstack/20140226-155910-suyHHe/openstack-setup.log
 * The generated manifests are available at: /var/tmp/packstack/20140226-155910-suyHHe/manifests

What just happened?

  • Lots of packages installed
  • Configuration in /etc/{nova,glance,cinder,neutron,...}/
  • Firewall rules for all services
  • MySQL users and databases created
  • Services started

Post-install Configuration

Fix Horizon

You may need to fix ALLOWED_HOSTS:

# sed -i '/^ALLOWED_HOSTS/ s/=.*/= [ "*" ]/' \
  /etc/openstack-dashboard/local_settings 
# service httpd restart

Source your admin credentials:

# . /root/keystonerc_admin

Create a disk image:

glance image-create \
  --copy-from http://download.cirros-cloud.net/0.3.1/cirros-0.3.1-x86_64-disk.img \
  --is-public true \
  --container-format bare \
  --disk-format qcow2 \
  --name cirros

Create external network:

# neutron net-create external --router:external=True
# neutron subnet-create --disable-dhcp external 172.16.13.0/24

Create a flavor for testing:

# nova flavor-create m1.nano auto 128 1 1

This flavor consumes minimal memory and disk so it is better than the default flavors for testing in constrained environments.

Create a non-admin user:

# keystone tenant-create --name demo
# keystone user-create --name demo --tenant demo --pass demo

And store the credentials in /root/keystonerc_demo:

export OS_USERNAME=demo
export OS_TENANT_NAME=demo
export OS_PASSWORD=demo
export OS_AUTH_URL=http://10.15.0.7:35357/v2.0/
export PS1='[\u@\h \W(keystone_demo)]\$ '

Switch credentials

From this point on we're going to be operating as the "demo" user:

# . /root/keystonerc_demo

Create an ssh keypair

Create a keypair:

# ssh-keygen -t rsa -b 2048 -N '' -f id_rsa_demo

Upload the public key to Nova:

# nova keypair-add --pub-key id_rsa_demo.pub demo

We'll use this later on when logging in instances.

Create tenant networks

Create a private network:

# neutron net-create net0
# neutron subnet-create --name net0-subnet0 \
  --dns-nameserver 8.8.8.8 net0 10.0.0.0/24

Create a router and connect it to the private network and the external network:

# neutron router-create extrouter
# neutron router-gateway-set extrouter external
# neutron router-interface-add extrouter net0-subnet0

Now we should have something like:

# neutron net-list
+--------------------------------------+----------+--------------------------------------------------+
| id                                   | name     | subnets                                          |
+--------------------------------------+----------+--------------------------------------------------+
| 77cafb07-a793-41cb-8a96-58d04408e10d | net0     | f0beab82-0673-40eb-8934-68acc6bd635a 10.0.0.0/24 |
| e1de0593-73d4-427d-89f6-9c7b0e7c7ef9 | external | 57c65000-0782-40c3-906e-09d9a4ad5113             |
+--------------------------------------+----------+--------------------------------------------------+

Create security rules

Make sure we allow ICMP and SSH traffic to instances:

# neutron security-group-rule-create --protocol icmp default
# neutron security-group-rule-create --protocol tcp \
  --port-range-min 22 --port-range-max 22 default

Booting an instance

We'll need the UUID for network net0 that we created in the previous step:

# nova boot --poll --flavor m1.nano --image cirros \
  --nic net-id=77cafb07-a793-41cb-8a96-58d04408e10d \
  --key-name demo test0

You can't get there from here...

The instance is up and has an ip on an internal network...

...but there's no way for us to get there from here.

...or can you?

Actually, we can get at the instance from inside the appropriate network namespace on the network host:

# ip netns exec dhcp-77cafb07-a793-41cb-8a96-58d04408e10d \
  ping ...

Create and assign a floating ip

Allocate a floating ip address from the external network:

# nova floating-ip-create external
+-------------+-------------+----------+----------+
| Ip          | Instance Id | Fixed Ip | Pool     |
+-------------+-------------+----------+----------+
| 172.16.13.3 | None        | None     | external |
+-------------+-------------+----------+----------+

Assign it to the new instance:

# nova add-floating-ip test0 172.16.13.3

External network connectivity

In the real world:

In our demo:

# ip addr add 172.16.13.1/24 dev br-ex
# iptables -t nat -I POSTROUTING 1 -s 172.16.13.0/24 -j MASQUERADE

About this hangout