h1. Openstack Installation TTNN

h2. Pending upstream fixes:

* build-debian-openstack-images, changes should be in next package version:

** Fix extlinux installation on jessie: http://anonscm.debian.org/cgit/openstack/openstack-debian-images.git/commit/?id=1a4ddceb99fb66f42f79e87a183d8e96952e5ac3

** Fix cloud-init setup with configdrive: http://anonscm.debian.org/cgit/openstack/openstack-debian-images.git/commit/?id=15cce87e2fa15b8952d69fb58711bfa88b21ef87

** Use Configdrive first for cloud-init: http://anonscm.debian.org/cgit/openstack/openstack-debian-images.git/commit/?id=1b1e1093d5ee525426382ad29b8219ab58f6656d

* nova: 

** Fix live-migration avec ceph and configdrive:: https://review.openstack.org/#/c/112014/, si la solution est mergé upstream, j'appliquerai le patch au package debian

h2. Installation

Le setup installé est composé de :

* 3 hyperviseurs (compute node) qui contiendront 1 ceph-mon, N ceph-osd, nova-compute (qui lance les VM), neutron-openvswitch-agent (qui configure le réseau des vms)

* 1 VM (controler node), qui contient tous les services de management d'openstack (nova, cinder, glance, neutron)

Les étapes d'installation sont les suivantes:

* Préparation manuelle de la VM de management sur un des hyperviseurs

* Installation de ceph et openstack sur toutes les machines avec puppet

* Déplacement de la VM de management d'openstack dans openstack.

h3. Information sur le puppet master:

La conf et modules puppet utilisé par tetaneutral.net ce trouve la:

> http://chiliproject.tetaneutral.net/projects/git-tetaneutral-net/repository/puppetmaster

> http://chiliproject.tetaneutral.net/projects/git-tetaneutral-net/repository/puppetmaster/revisions/master/entry/manifests/nodes/openstack.pp

Seulement c'est 3 classes sont spécifique à tetaneutral, le reste est réutilisable:

  class { 'ttnn::server': }

  class { 'backup::client': }

  class { 'sexymotd': }

Les modules qui sont nécessaire pour openstack pour le reste du fichier sont:

> https://github.com/puppetlabs/puppetlabs-apache.git 

> https://github.com/puppetlabs/puppetlabs-apt 

> https://github.com/stackforge/puppet-ceph.git 

> https://github.com/stackforge/puppet-cinder.git 

> https://github.com/puppetlabs/puppetlabs-concat.git 

> https://github.com/stackforge/puppet-glance.git 

> https://github.com/stackforge/puppet-horizon.git 

> https://github.com/puppetlabs/puppetlabs-inifile.git 

> https://github.com/stackforge/puppet-keystone.git 

> https://github.com/camptocamp/puppet-kmod.git 

> https://github.com/saz/puppet-memcached.git 

> https://github.com/puppetlabs/puppetlabs-mysql.git 

> https://github.com/stackforge/puppet-neutron.git 

> https://github.com/stackforge/puppet-nova.git 

> https://github.com/puppetlabs/puppetlabs-rabbitmq.git 

> https://github.com/nanliu/puppet-staging.git 

> https://github.com/puppetlabs/puppetlabs-stdlib.git 

> https://github.com/duritong/puppet-sysctl.git 

> https://github.com/puppetlabs/puppetlabs-vcsrepo.git 

> https://github.com/stackforge/puppet-vswitch.git 

> https://github.com/puppetlabs/puppetlabs-xinetd.git 

h3. Préparation réseaux des hosts:

Voici le fichier rc.local d'un hyperviseur (g1)

<pre>

for i in /proc/sys/net/ipv6/conf/*; do for j in autoconf accept_ra; do echo 0 > $i/$j; done;done

# https://www.kernel.org/doc/Documentation/networking/ip-sysctl.txt

for i in /proc/sys/net/ipv4/conf/*/arp_announce; do echo 2 > $i;done           # NOTE(sileht) Est ce que c'est vraiment utilise quand on aura le switch 10G entre les machines ?

echo 1 > /proc/sys/net/ipv4/ip_forward

echo 1 > /proc/sys/net/ipv6/conf/all/forwarding

ip link set eth0 up

ip link add link eth0 name eth0.3131 type vlan id 3131

ip link set eth0.3131 up

ip addr add 89.234.156.251/32 dev eth0.3131

ip route add 91.224.149.0/24 dev eth0.3131

ip route add default via 91.224.149.254

ip addr add 2a01:6600:8083:fb00::1/56 dev eth0.3131

ip route add default via fe80::31 dev eth0.3131

ip link add link eth0 name eth0.3175 type vlan id 3175

ip link set eth0.3175 up

ip link add link eth0 name eth0.3195 type vlan id 3195

ip link set eth0.3195 up

ip addr add 91.224.148.0/32 dev lo

ip addr add fe80::42/64 dev br-net

ip route add 89.234.156.250/32 dev eth0.3131

ip route add 89.234.156.252/32 dev eth0.3131

ip route add 89.234.156.253/32 dev eth0.3131

exit 0

</pre>

Sur h7.tetaneutral.net le routage est le suivant

<pre>

ip route add 89.234.156.251/32 dev eth0.3131

ip route add 89.234.156.252/32 dev eth0.3131

ip route add 89.234.156.253/32 dev eth0.3131

ip route add 2a01:6600:8083:fb00::/56 dev eth0.3131

ip route add 2a01:6600:8083:fc00::/56 dev eth0.3131

ip route add 2a01:6600:8083:fd00::/56 dev eth0.3131

</pre>

Ainsi que ceci le temps de l'installation

<pre>

ip route add 89.234.156.249/32 via 89.234.156.251 # NOTE(sileht): temporairement pour le setup openstack

ip route add 2a01:6600:8083:f900::/56 via 2a01:6600:8083:fb00:: dev eth0.3131 # NOTE(sileht): temporairement pour le setup openstack

</pre>

h3. Preparation de la VM de management, le controlleur

h4. Installation des prérequis

<pre>

$ apt-get install libvirt-bin openstack-debian-images openvswitch-switch openvswitch-datapath-dkms virtinst

#NOTE(sileht): temporairement appliquer les patches de au-dessus à openstack-debian-images 

</pre>

h4. Preparation de l'image de la VM

<pre>

$ vim hook.sh

set -xv

echo "openstack" > $BODI_CHROOT_PATH/etc/hostname

cat > $BODI_CHROOT_PATH/etc/resolv.conf <<EOF

domain tetaneutral.net

search tetaneutral.net

nameserver 8.8.8.8

EOF

cat > $BODI_CHROOT_PATH/etc/network/interfaces <<EOF

auto lo

iface lo inet loopback

auto eth0

iface eth0 inet manual

  pre-up /sbin/sysctl -w net.ipv6.conf.eth0.accept_ra=0

  pre-up /sbin/sysctl -w net.ipv6.conf.eth0.autoconf=0

  up ip link set eth0 up

  up ip addr add 89.234.156.249/32 dev eth0

  up ip route add 91.224.148.0/32 dev eth0

  up ip route add default via 91.224.148.0

iface eth0 inet6 static

    pre-up /sbin/sysctl -w net.ipv6.conf.eth0.accept_ra=0

    pre-up /sbin/sysctl -w net.ipv6.conf.eth0.autoconf=0

    address 2a01:6600:8083:f900::1

    netmask 56

    gateway fe80::42

EOF

mkdir -p $BODI_CHROOT_PATH/root/.ssh

chmod 600 $BODI_CHROOT_PATH/root/.ssh

cat >> $BODI_CHROOT_PATH/root/.ssh/authorized_keys <<EOF

ssh-dss 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 root@h1

EOF

sed -i -e 's/^.*Port.*$/Port 2222/' -e 's/^[# ]*PasswordAuthentication.*$/PasswordAuthentication no/' $BODI_CHROOT_PATH/etc/ssh/sshd_config

chroot $BODI_CHROOT_PATH dpkg-reconfigure openssh-server

chroot $BODI_CHROOT_PATH apt-get purge -y cloud-init*

</pre>

<pre>

$ chmod +x hook.sh

$ build-openstack-debian-image --image-size 20  --release jessie -u http://apt.tetaneutral.net/debian/ -s http://apt.tetaneutral.net/debian/ --hook-script $(pwd)/hook.sh

$ mv debian-jessie-7.0.0-3-amd64.raw /openstack.raw 

$ rm debian-jessie-7.0.0-3-amd64.qcow2

</pre>

_note: la derniere commande lancée par le script doit être 'qemu-img convert -c -f raw .... , sinon l'image de bootera pas_

_note(jessie): si l'erreur est celle ci-dessous, relancer manuellement 'kpartx -d debian-jessie-7.0.0-3-amd64.raw' jusqu'a ce que l'erreur disparaissent et le .raw est utilisable:_

<pre>

+ 'kpartx -d debian-jessie-7.0.0-3-amd64.raw'

device-mapper: remove ioctl on loop0p1 failed: Device or resource busy

loop deleted : /dev/loop0

</pre>

On garde l'image raw qui sera importable dans ceph plus tard et on la mets à la racine pour qui l'utilisateur libvirt puisse y accéder.

h4. Préparation du réseau

<pre>

$ ovs-vsctl add-br br-net

$ ip link set br-net up

$ ovs-vsctl br-set-external-id br-net bridge-id br-net

$ ip addr add fe80::42/64 dev br-net

$ ip route add 89.234.156.249/32 dev br-net proto 42

$ ip -6 route add 2a01:6600:8083:f900::/56 dev br-net proto 42

</pre>

h4. Installation de la VM

<pre>

$ virt-install --name openstack --ram 2048 --network bridge=br-net --nographics --serial pty --wait -1 --noreboot --autostart --disk /openstack.raw,bus=virtio,cache=none,io=native --import

$ EDITOR="sed -i -e \"s,<source bridge='br-net'/>,<source bridge='br-net'/><virtualport type='openvswitch' />,g\"" virsh edit openstack  # Openvswitch is not yet supported by virt-install

$ virsh  start openstack

$ ssh root@openstack.tetaneutral.net -p 2222

</pre>

Une fois connecté à la VM:

<pre>

$ apt-get install puppet

$ puppet agent --enable

$ puppet agent -vt --server puppet.tetaneutral.net --certname openstack.tetaneutral.net --pluginsync 

Info: Caching certificate for ca

Info: csr_attributes file loading from /etc/puppet/csr_attributes.yaml

Info: Creating a new SSL certificate request for openstack.tetaneutral.net

Info: Certificate Request fingerprint (SHA256): AE:72:47:40:A0:E2:F4:59:BA:39:FA:3D:C2:A7:C9:1B:9F:87:A5:B7:65:3A:F9:D4:DE:AF:E2:A3:02:41:0F:2E

Info: Caching certificate for ca

Exiting; no certificate found and waitforcert is disabled

</pre>

Sur le puppetmaster:

<pre>

$ puppet ca sign openstack.tetaneutral.net 

</pre>

_Note: si un vieux cert existe: puppet node clean openstack.tetaneutral.net _

De nouveau sur la VM:

<pre>

$ puppet agent -vt --server puppet.tetaneutral.net --certname openstack.tetaneutral.net --pluginsync 

$ puppet agent -vt --server puppet.tetaneutral.net --certname openstack.tetaneutral.net --pluginsync 

$ puppet agent -vt --server puppet.tetaneutral.net --certname openstack.tetaneutral.net --pluginsync 

</pre>

Il ne doit plus y avoir d'erreur la 3°/4° fois.

_note: Un paquet est manquant en ce moment (06/08/2014) dans jessie 'nullmailer', pour l'installer manuellement:_

<pre>

wget http://ftp.fr.debian.org/debian/pool/main/n/nullmailer/nullmailer_1.11-2.1_amd64.deb

apt-get purge exim4-base          exim4-config        exim4-daemon-light

dpkg -i nullmailer_1.11-2.1_amd64.deb  # Répondre n'importe quoi, ca sera ecrasé par puppet

apt-get install bsd-mailx

$ puppet agent -vt --server puppet.tetaneutral.net --certname openstack.tetaneutral.net --pluginsync 

</pre>

h3. Configuration des hyperviseurs

h2. Management d'openstack

h3. Build and upload debian image:

<pre>

$ apt-get install openstack-debian-images

#NOTE(sileht): temporairement appliquer les patches de au-dessus

$ build-openstack-debian-image -m -r jessie  -u http://apt.tetaneutral.net/debian/ -s http://apt.tetaneutral.net/debian/

$ glance image-create --file debian-jessie-7.0.0-3-amd64.raw --is-public True --human-readable --progress --disk-format raw --container-format bare --name debian-jessie-amd64

</pre>