FastForward is an ultimate DevOps platform.
- The OpenStack bare metal hosts are in MAAS environment(recommend)
- All hosts are two NICs at least(external and internal)
- We assume that you have ceph installed, the cinder bachend default using ceph, the running instace default to using ceph as it's local storage. About ceph please visit: http://docs.ceph.com/docs/master/rbd/rbd-openstack/ or see the (Option)Ceph Guide below.
- For resize instance nova user can be login to each compute node via ssh passwordless(include sudo), and all compute nodes need to restart libvirt-bin to enable live migration
- The FastForward node is the same as ceph-deploy node where can be login to each openstack node passwordless and sudo-passwordless
- The FastForward node default using ~/.ssh/id_rsa ssh private key to logon remote server
- You need to restart the
nova-compute,cinder-volumeandglance-apiservices to finalize the installation if you have selected the ceph as that backend - FastForward support consistency groups for future use but the default LVM and Ceph driver does not support consistency groups yet because the consistency technology is not available at that storage level
Install FastForward on FASTFORWARD-NODE
pip install fastforward
Prepare the OpenStack environment. (NOTE) DO NOT setup eth1 in /etc/network/interfaces
ff --user ubuntu --hosts \
HAPROXY1,\
HAPROXY2,\
CONTROLLER1,\
CONTROLLER2,\
COMPUTE1,\
COMPUTE2,\
COMPUTE3,\
COMPUTE4,\
COMPUTE5,\
COMPUTE6,\
COMPUTE7,\
COMPUTE8,\
COMPUTE9,\
COMPUTE10 \
environment \
prepare-host --public-interface eth1
Deploy to CONTROLLER1
ff --user ubuntu --hosts CONTROLLER1 openstack mysql install
ff --user ubuntu --hosts CONTROLLER1 openstack mysql config --wsrep-cluster-address "gcomm://CONTROLLER1,CONTROLLER2" --wsrep-node-name="galera1" --wsrep-node-address="CONTROLLER1"
Deploy to CONTROLLER2
ff --user ubuntu --hosts CONTROLLER2 openstack mysql install
ff --user ubuntu --hosts CONTROLLER2 openstack mysql config --wsrep-cluster-address "gcomm://CONTROLLER1,CONTROLLER2" --wsrep-node-name="galera2" --wsrep-node-address="CONTROLLER2"
Start the cluster
ff --user ubuntu --hosts CONTROLLER1 openstack mysql manage --wsrep-new-cluster
ff --user ubuntu --hosts CONTROLLER2 openstack mysql manage --start
ff --user ubuntu --hosts CONTROLLER1 openstack mysql manage --change-root-password changeme
Show the cluster status
ff --user ubuntu --hosts CONTROLLER1 openstack mysql manage --show-cluster-status --root-db-pass changeme
Deploy to HAPROXY1
ff --user ubuntu --hosts HAPROXY1 openstack haproxy install
Deploy to HAPROXY2
ff --user ubuntu --hosts HAPROXY2 openstack haproxy install
Generate the HAProxy configuration and upload to target hosts(Do not forget to edit the generated configuration)
ff openstack haproxygen-conf
ff --user ubuntu --hosts HAPROXY1,HAPROXY2 openstack haproxy config --upload-conf haproxy.cfg
Configure Keepalived
ff --user ubuntu --hosts HAPROXY1 openstack haproxy config --configure-keepalived --router_id lb1 --priority 150 --state MASTER --interface eth0 --vip CONTROLLER_VIP
ff --user ubuntu --hosts HAPROXY2 openstack haproxy config --configure-keepalived --router_id lb2 --priority 100 --state SLAVE --interface eth0 --vip CONTROLLER_VIP
Deploy to CONTROLLER1 and CONTROLLER2
ff --user ubuntu --hosts CONTROLLER1,CONTROLLER2 openstack rabbitmq install --erlang-cookie changemechangeme --rabbit-user openstack --rabbit-pass changeme
Create cluster(Ensure CONTROLLER2 can access CONTROLLER1 via hostname)
ff --user ubuntu --hosts CONTROLLER2 openstack rabbitmq join-cluster --name rabbit@CONTROLLER1
Create keystone database
ff --user ubuntu --hosts CONTROLLER1 openstack keystone create-keystone-db --root-db-pass changeme --keystone-db-pass changeme
Install keystone on CONTROLLER1 and CONTROLLER2
ff --user ubuntu --hosts CONTROLLER1 openstack keystone install --admin-token changeme --connection mysql+pymysql://keystone:changeme@CONTROLLER_VIP/keystone --memcached-servers CONTROLLER1:11211,CONTROLLER2:11211 --populate
ff --user ubuntu --hosts CONTROLLER2 openstack keystone install --admin-token changeme --connection mysql+pymysql://keystone:changeme@CONTROLLER_VIP/keystone --memcached-servers CONTROLLER1:11211,CONTROLLER2:11211
Create the service entity and API endpoints
ff --user ubuntu --hosts CONTROLLER1 openstack keystone create-entity-and-endpoint --os-token changeme --os-url http://CONTROLLER_VIP:35357/v3 --public-endpoint http://CONTROLLER_VIP:5000/v3 --internal-endpoint http://CONTROLLER_VIP:5000/v3 --admin-endpoint http://CONTROLLER_vip:35357/v3
Create projects, users, and roles
ff --user ubuntu --hosts CONTROLLER1 openstack keystone create-projects-users-roles --os-token changeme --os-url http://CONTROLLER_VIP:35357/v3 --admin-pass changeme --demo-pass changeme
(OPTION) you will need to create OpenStack client environment scripts admin-openrc.sh
export OS_PROJECT_DOMAIN_NAME=default
export OS_USER_DOMAIN_NAME=default
export OS_PROJECT_NAME=admin
export OS_TENANT_NAME=admin
export OS_USERNAME=admin
export OS_PASSWORD=changeme
export OS_AUTH_URL=http://CONTROLLER_VIP:35357/v3
export OS_IDENTITY_API_VERSION=3
export OS_IMAGE_API_VERSION=2
export OS_AUTH_VERSION=3
demo-openrc.sh
export OS_PROJECT_DOMAIN_NAME=default
export OS_USER_DOMAIN_NAME=default
export OS_PROJECT_NAME=demo
export OS_TENANT_NAME=demo
export OS_USERNAME=demo
export OS_PASSWORD=changeme
export OS_AUTH_URL=http://CONTROLLER_VIP:5000/v3
export OS_IDENTITY_API_VERSION=3
export OS_IMAGE_API_VERSION=2
export OS_AUTH_VERSION=3
Create glance database
ff --user ubuntu --hosts CONTROLLER1 openstack glance create-glance-db --root-db-pass changeme --glance-db-pass changeme
Create service credentials
ff --user ubuntu --hosts CONTROLLER1 openstack glance create-service-credentials --os-password changeme --os-auth-url http://CONTROLLER_VIP:35357/v3 --glance-pass changeme --public-endpoint http://CONTROLLER_VIP:9292 --internal-endpoint http://CONTROLLER_VIP:9292 --admin-endpoint http://CONTROLLER_VIP:9292
Install glance on CONTROLLER1 and CONTROLLER2
ff --user ubuntu --hosts CONTROLLER1 openstack glance install --connection mysql+pymysql://glance:GLANCE_PASS@CONTROLLER_VIP/glance --auth-uri http://CONTROLLER_VIP:5000 --auth-url http://CONTROLLER_VIP:35357 --glance-pass changeme --memcached-servers CONTROLLER1:11211,CONTROLLER2:11211 --populate
ff --user ubuntu --hosts CONTROLLER2 openstack glance install --connection mysql+pymysql://glance:GLANCE_PASS@CONTROLLER_VIP/glance --auth-uri http://CONTROLLER_VIP:5000 --auth-url http://CONTROLLER_VIP:35357 --glance-pass changeme --memcached-servers CONTROLLER1:11211,CONTROLLER2:11211
Create nova database
ff --user ubuntu --hosts CONTROLLER1 openstack nova create-nova-db --root-db-pass changeme --nova-db-pass changeme
Create service credentials
ff --user ubuntu --hosts CONTROLLER1 openstack nova create-service-credentials --os-password changeme --os-auth-url http://CONTROLLER_VIP:35357/v3 --nova-pass changeme --public-endpoint 'http://CONTROLLER_VIP:8774/v2.1/%\(tenant_id\)s' --internal-endpoint 'http://CONTROLLER_VIP:8774/v2.1/%\(tenant_id\)s' --admin-endpoint 'http://CONTROLLER_VIP:8774/v2.1/%\(tenant_id\)s'
Install nova on CONTROLLER1
ff --user ubuntu --hosts CONTROLLER1 openstack nova install --connection mysql+pymysql://nova:NOVA_PASS@CONTROLLER_VIP/nova --api-connection mysql+pymysql://nova:NOVA_PASS@CONTROLLER_VIP/nova_api --auth-uri http://CONTROLLER_VIP:5000 --auth-url http://CONTROLLER_VIP:35357 --nova-pass changeme --my-ip MANAGEMENT_IP --memcached-servers CONTROLLER1:11211,CONTROLLER2:11211 --rabbit-hosts CONTROLLER1,CONTROLLER2 --rabbit-user openstack --rabbit-pass changeme --glance-api-servers http://CONTROLLER_VIP:9292 --neutron-endpoint http://CONTROLLER_VIP:9696 --neutron-pass changeme --metadata-proxy-shared-secret changeme --populate
Install nova on CONTROLLER2
ff --user ubuntu --hosts CONTROLLER2 openstack nova install --connection mysql+pymysql://nova:NOVA_PASS@CONTROLLER_VIP/nova --api-connection mysql+pymysql://nova:NOVA_PASS@CONTROLLER_VIP/nova_api --auth-uri http://CONTROLLER_VIP:5000 --auth-url http://CONTROLLER_VIP:35357 --nova-pass changeme --my-ip MANAGEMENT_IP --memcached-servers CONTROLLER1:11211,CONTROLLER2:11211 --rabbit-hosts CONTROLLER1,CONTROLLER2 --rabbit-user openstack --rabbit-pass changeme --glance-api-servers http://CONTROLLER_VIP:9292 --neutron-endpoint http://CONTROLLER_VIP:9696 --neutron-pass changeme --metadata-proxy-shared-secret changeme
Add nova computes(use uuidgen to generate the ceph uuid)
ff --user ubuntu --hosts COMPUTE1 openstack nova-compute install --my-ip MANAGEMENT_IP --rabbit-hosts CONTROLLER1,CONTROLLER2 --rabbit-user openstack --rabbit-pass changeme --auth-uri http://CONTROLLER_VIP:5000 --auth-url http://CONTROLLER_VIP:35357 --nova-pass changeme --novncproxy-base-url http://CONTROLLER_VIP:6080/vnc_auto.html --glance-api-servers http://CONTROLLER_VIP:9292 --neutron-endpoint http://CONTROLLER_VIP:9696 --neutron-pass changeme --rbd-secret-uuid changeme-changeme-changeme-changeme --memcached-servers CONTROLLER1:11211,CONTROLLER2:11211
ff --user ubuntu --hosts COMPUTE2 openstack nova-compute install --my-ip MANAGEMENT_IP --rabbit-hosts CONTROLLER1,CONTROLLER2 --rabbit-user openstack --rabbit-pass changeme --auth-uri http://CONTROLLER_VIP:5000 --auth-url http://CONTROLLER_VIP:35357 --nova-pass changeme --novncproxy-base-url http://CONTROLLER_VIP:6080/vnc_auto.html --glance-api-servers http://CONTROLLER_VIP:9292 --neutron-endpoint http://CONTROLLER_VIP:9696 --neutron-pass changeme --rbd-secret-uuid changeme-changeme-changeme-changeme --memcached-servers CONTROLLER1:11211,CONTROLLER2:11211
The libvirt defaults to using ceph as shared storage, the ceph pool for running instance is vms. if you do not using ceph as it's bachend, you must remove the following param:
images_type = rbd
images_rbd_pool = vms
images_rbd_ceph_conf = /etc/ceph/ceph.conf
rbd_user = cinder
rbd_secret_uuid = changeme-changeme-changeme-changeme
disk_cachemodes="network=writeback"
live_migration_flag="VIR_MIGRATE_UNDEFINE_SOURCE,VIR_MIGRATE_PEER2PEER,VIR_MIGRATE_LIVE,VIR_MIGRATE_PERSIST_DEST,VIR_MIGRATE_TUNNELLED"
Create nova database
ff --user ubuntu --hosts CONTROLLER1 openstack neutron create-neutron-db --root-db-pass changeme --neutron-db-pass changeme
Create service credentials
ff --user ubuntu --hosts CONTROLLER1 openstack neutron create-service-credentials --os-password changeme --os-auth-url http://CONTROLLER_VIP:35357/v3 --neutron-pass changeme --public-endpoint http://CONTROLLER_VIP:9696 --internal-endpoint http://CONTROLLER_VIP:9696 --admin-endpoint http://CONTROLLER_VIP:9696
Install Neutron for self-service
ff --user ubuntu --hosts CONTROLLER1 openstack neutron install --connection mysql+pymysql://neutron:NEUTRON_PASS@CONTROLLER_VIP/neutron --rabbit-hosts CONTROLLER1,CONTROLLER2 --rabbit-user openstack --rabbit-pass changeme --auth-uri http://CONTROLLER_VIP:5000 --auth-url http://CONTROLLER_VIP:35357 --neutron-pass changeme --nova-url http://CONTROLLER_VIP:8774/v2.1 --nova-pass changeme --public-interface eth1 --local-ip MANAGEMENT_INTERFACE_IP --nova-metadata-ip CONTROLLER_VIP --metadata-proxy-shared-secret changeme-changeme-changeme-changeme --memcached-servers CONTROLLER1:11211,CONTROLLER2:11211 --populate
ff --user ubuntu --hosts CONTROLLER2 openstack neutron install --connection mysql+pymysql://neutron:NEUTRON_PASS@CONTROLLER_VIP/neutron --rabbit-hosts CONTROLLER1,CONTROLLER2 --rabbit-user openstack --rabbit-pass changeme --auth-uri http://CONTROLLER_VIP:5000 --auth-url http://CONTROLLER_VIP:35357 --neutron-pass changeme --nova-url http://CONTROLLER_VIP:8774/v2.1 --nova-pass changeme --public-interface eth1 --local-ip MANAGEMENT_INTERFACE_IP --nova-metadata-ip CONTROLLER_VIP --metadata-proxy-shared-secret changeme-changeme-changeme-changeme --memcached-servers CONTROLLER1:11211,CONTROLLER2:11211
Install neutron agent on compute nodes
ff --user ubuntu --hosts COMPUTE1 openstack neutron-agent install --rabbit-hosts CONTROLLER1,CONTROLLER2 --rabbit-user openstack --rabbit-pass changeme --auth-uri http://CONTROLLER_VIP:5000 --auth-url http://CONTROLLER_VIP:35357 --neutron-pass changeme --public-interface eth1 --local-ip MANAGEMENT_INTERFACE_IP --memcached-servers CONTROLLER1:11211,CONTROLLER2:11211
ff --user ubuntu --hosts COMPUTE2 openstack neutron-agent install --rabbit-hosts CONTROLLER1,CONTROLLER2 --rabbit-user openstack --rabbit-pass changeme --auth-uri http://CONTROLLER_VIP:5000 --auth-url http://CONTROLLER_VIP:35357 --neutron-pass changeme --public-interface eth1 --local-ip MANAGEMENT_INTERFACE_IP --memcached-servers CONTROLLER1:11211,CONTROLLER2:11211
Install horizon on controller nodes
ff --user ubuntu --hosts CONTROLLER1,CONTROLLER2 openstack horizon install --openstack-host CONTROLLER_VIP --memcached-servers CONTROLLER1:11211 --time-zone Asia/Shanghai
Create cinder database
ff --user ubuntu --hosts CONTROLLER1 openstack cinder create-cinder-db --root-db-pass changeme --cinder-db-pass changeme
Create cinder service creadentials
ff --user ubuntu --hosts CONTROLLER1 openstack cinder create-service-credentials --os-password changeme --os-auth-url http://CONTROLLER_VIP:35357/v3 --cinder-pass changeme --public-endpoint-v1 'http://CONTROLLER_VIP:8776/v1/%\(tenant_id\)s' --internal-endpoint-v1 'http://CONTROLLER_VIP:8776/v1/%\(tenant_id\)s' --admin-endpoint-v1 'http://CONTROLLER_VIP:8776/v1/%\(tenant_id\)s' --public-endpoint-v2 'http://CONTROLLER_VIP:8776/v2/%\(tenant_id\)s' --internal-endpoint-v2 'http://CONTROLLER_VIP:8776/v2/%\(tenant_id\)s' --admin-endpoint-v2 'http://CONTROLLER_VIP:8776/v2/%\(tenant_id\)s'
Install cinder-api and cinder-volume on controller nodes, the volume backend defaults to ceph (you must have ceph installed)
ff --user ubuntu --hosts CONTROLLER1 openstack cinder install --connection mysql+pymysql://cinder:CINDER_PASS@CONTROLLER_VIP/cinder --rabbit-user openstack --rabbit-pass changeme --rabbit-hosts CONTROLLER1,CONTROLLER2 --auth-uri http://CONTROLLER_VIP:5000 --auth-url http://CONTROLLER_VIP:35357 --cinder-pass changeme --my-ip MANAGEMENT_INTERFACE_IP --glance-api-servers http://CONTROLLER_VIP:9292 --rbd-secret-uuid changeme-changeme-changeme-changeme --memcached-servers CONTROLLER1:11211,CONTROLLER2:11211 --populate
ff --user ubuntu --hosts CONTROLLER2 openstack cinder install --connection mysql+pymysql://cinder:CINDER_PASS@CONTROLLER_VIP/cinder --rabbit-user openstack --rabbit-pass changeme --rabbit-hosts CONTROLLER1,CONTROLLER2 --auth-uri http://CONTROLLER_VIP:5000 --auth-url http://CONTROLLER_VIP:35357 --cinder-pass changeme --my-ip MANAGEMENT_INTERFACE_IP --glance-api-servers http://CONTROLLER_VIP:9292 --rbd-secret-uuid changeme-changeme-changeme-changeme --memcached-servers CONTROLLER1:11211,CONTROLLER2:11211
Create the Identity service credentials
ff --user ubuntu --hosts CONTROLLER1 openstack swift create-service-credentials --os-password changeme --os-auth-url http://CONTROLLER_VIP:35357/v3 --swift-pass changeme --public-endpoint 'http://CONTROLLER_VIP:8080/v1/AUTH_%\(tenant_id\)s' --internal-endpoint 'http://CONTROLLER_VIP:8080/v1/AUTH_%\(tenant_id\)s' --admin-endpoint http://CONTROLLER_VIP:8080/v1
Install swift proxy
ff --user ubuntu --hosts CONTROLLER1,CONTROLLER2 openstack swift install --auth-uri http://CONTROLLER_VIP:5000 --auth-url http://CONTROLLER_VIP:35357 --swift-pass changeme --memcached-servers CONTROLLER1:11211,CONTROLLER2:11211
Prepare disks on storage node
ff --user ubuntu --hosts OBJECT1,OBJECT2 openstack swift-storage prepare-disks --name sdb,sdc,sdd,sde
Install swift storage on storage node
ff --user ubuntu --hosts OBJECT1 openstack swift-storage install --address MANAGEMENT_INTERFACE_IP --bind-ip MANAGEMENT_INTERFACE_IP
ff --user ubuntu --hosts OBJECT2 openstack swift-storage install --address MANAGEMENT_INTERFACE_IP --bind-ip MANAGEMENT_INTERFACE_IP
Create account ring on controller node
ff --user ubuntu --hosts CONTROLLER1 openstack swift-storage create-account-builder-file --partitions 10 --replicas 3 --moving 1
ff --user ubuntu --hosts CONTROLLER1 openstack swift-storage account-builder-add --region 1 --zone 1 --ip OBJECT1_MANAGEMENT_IP --device sdb --weight 100
ff --user ubuntu --hosts CONTROLLER1 openstack swift-storage account-builder-add --region 1 --zone 1 --ip OBJECT1_MANAGEMENT_IP --device sdc --weight 100
ff --user ubuntu --hosts CONTROLLER1 openstack swift-storage account-builder-add --region 1 --zone 1 --ip OBJECT1_MANAGEMENT_IP --device sdd --weight 100
ff --user ubuntu --hosts CONTROLLER1 openstack swift-storage account-builder-add --region 1 --zone 1 --ip OBJECT1_MANAGEMENT_IP --device sde --weight 100
ff --user ubuntu --hosts CONTROLLER1 openstack swift-storage account-builder-add --region 1 --zone 1 --ip OBJECT2_MANAGEMENT_IP --device sdb --weight 100
ff --user ubuntu --hosts CONTROLLER1 openstack swift-storage account-builder-add --region 1 --zone 1 --ip OBJECT2_MANAGEMENT_IP --device sdc --weight 100
ff --user ubuntu --hosts CONTROLLER1 openstack swift-storage account-builder-add --region 1 --zone 1 --ip OBJECT2_MANAGEMENT_IP --device sdd --weight 100
ff --user ubuntu --hosts CONTROLLER1 openstack swift-storage account-builder-add --region 1 --zone 1 --ip OBJECT2_MANAGEMENT_IP --device sde --weight 100
ff --user ubuntu --hosts CONTROLLER1 openstack swift-storage account-builder-rebalance
Create container ring on controller node
ff --user ubuntu --hosts CONTROLLER1 openstack swift-storage create-container-builder-file --partitions 10 --replicas 3 --moving 1
ff --user ubuntu --hosts CONTROLLER1 openstack swift-storage container-builder-add --region 1 --zone 1 --ip OBJECT1_MANAGEMENT_IP --device sdb --weight 100
ff --user ubuntu --hosts CONTROLLER1 openstack swift-storage container-builder-add --region 1 --zone 1 --ip OBJECT1_MANAGEMENT_IP --device sdc --weight 100
ff --user ubuntu --hosts CONTROLLER1 openstack swift-storage container-builder-add --region 1 --zone 1 --ip OBJECT1_MANAGEMENT_IP --device sdd --weight 100
ff --user ubuntu --hosts CONTROLLER1 openstack swift-storage container-builder-add --region 1 --zone 1 --ip OBJECT1_MANAGEMENT_IP --device sde --weight 100
ff --user ubuntu --hosts CONTROLLER1 openstack swift-storage container-builder-add --region 1 --zone 1 --ip OBJECT2_MANAGEMENT_IP --device sdb --weight 100
ff --user ubuntu --hosts CONTROLLER1 openstack swift-storage container-builder-add --region 1 --zone 1 --ip OBJECT2_MANAGEMENT_IP --device sdc --weight 100
ff --user ubuntu --hosts CONTROLLER1 openstack swift-storage container-builder-add --region 1 --zone 1 --ip OBJECT2_MANAGEMENT_IP --device sdd --weight 100
ff --user ubuntu --hosts CONTROLLER1 openstack swift-storage container-builder-add --region 1 --zone 1 --ip OBJECT2_MANAGEMENT_IP --device sde --weight 100
ff --user ubuntu --hosts CONTROLLER1 openstack swift-storage container-builder-rebalance
Create object ring on controller node
ff --user ubuntu --hosts CONTROLLER1 openstack swift-storage create-object-builder-file --partitions 10 --replicas 3 --moving 1
ff --user ubuntu --hosts CONTROLLER1 openstack swift-storage object-builder-add --region 1 --zone 1 --ip OBJECT1_MANAGEMENT_IP --device sdb --weight 100
ff --user ubuntu --hosts CONTROLLER1 openstack swift-storage object-builder-add --region 1 --zone 1 --ip OBJECT1_MANAGEMENT_IP --device sdc --weight 100
ff --user ubuntu --hosts CONTROLLER1 openstack swift-storage object-builder-add --region 1 --zone 1 --ip OBJECT1_MANAGEMENT_IP --device sdd --weight 100
ff --user ubuntu --hosts CONTROLLER1 openstack swift-storage object-builder-add --region 1 --zone 1 --ip OBJECT1_MANAGEMENT_IP --device sde --weight 100
ff --user ubuntu --hosts CONTROLLER1 openstack swift-storage object-builder-add --region 1 --zone 1 --ip OBJECT2_MANAGEMENT_IP --device sdb --weight 100
ff --user ubuntu --hosts CONTROLLER1 openstack swift-storage object-builder-add --region 1 --zone 1 --ip OBJECT2_MANAGEMENT_IP --device sdc --weight 100
ff --user ubuntu --hosts CONTROLLER1 openstack swift-storage object-builder-add --region 1 --zone 1 --ip OBJECT2_MANAGEMENT_IP --device sdd --weight 100
ff --user ubuntu --hosts CONTROLLER1 openstack swift-storage object-builder-add --region 1 --zone 1 --ip OBJECT2_MANAGEMENT_IP --device sde --weight 100
ff --user ubuntu --hosts CONTROLLER1 openstack swift-storage object-builder-rebalance
Sync the builder file from controller node to each storage node and other any proxy node
ff --user ubuntu --host CONTROLLER1 openstack swift-storage sync-builder-file --to CONTROLLER2,OBJECT1,OBJECT2
Finalize installation on all nodes
ff --user ubuntu --hosts CONTROLLER1,CONTROLLER2,OBJECT1,OBJECT2 openstack swift finalize-install --swift-hash-path-suffix changeme --swift-hash-path-prefix changeme
For more information about ceph backend visit:
On Xenial please using ceph-deploy version 1.5.34
Install ceph-deploy(1.5.34)
wget -q -O- 'https://download.ceph.com/keys/release.asc' | sudo apt-key add -
echo deb http://download.ceph.com/debian-jewel/ $(lsb_release -sc) main | sudo tee /etc/apt/sources.list.d/ceph.list
sudo apt-get update && sudo apt-get install ceph-deploy
Create ceph cluster directory
mkdir ceph-cluster
cd ceph-cluster
Create cluster and add initial monitor(s) to the ceph.conf
ceph-deploy new CONTROLLER1 CONTROLLER2 COMPUTE1 COMPUTE2 BLOCK1 BLOCK2
echo "osd pool default size = 2" | tee -a ceph.conf
Install ceph client(Optionaly you can use --release jewel to install jewel version, the ceph-deploy 1.5.34 default release is jewel) and you can use --repo-url http://your-local-repo.example.org/mirror/download.ceph.com/debian-jewel to specify the local repository.
ceph-deploy install PLAYBACK-NODE CONTROLLER1 CONTROLLER2 COMPUTE1 COMPUTE2 BLOCK1 BLOCK2
Add the initial monitor(s) and gather the keys
ceph-deploy mon create-initial
If you want to add additional monitors, do that
ceph-deploy mon add {additional-monitor}
Add ceph osd(s)
ceph-deploy osd create --zap-disk BLOCK1:/dev/sdb
ceph-deploy osd create --zap-disk BLOCK1:/dev/sdc
ceph-deploy osd create --zap-disk BLOCK2:/dev/sdb
ceph-deploy osd create --zap-disk BLOCK2:/dev/sdc
Sync admin key
ceph-deploy admin PLAYBACK-NODE CONTROLLER1 CONTROLLER2 COMPUTE1 COMPUTE2 BLOCK1 BLOCK2
sudo chmod +r /etc/ceph/ceph.client.admin.keyring # On all ceph clients node
Create osd pool for cinder and running instance
ceph osd pool create volumes 512
ceph osd pool create vms 512
ceph osd pool create images 512
Setup ceph client authentication
ceph auth get-or-create client.cinder mon 'allow r' osd 'allow class-read object_prefix rbd_children, allow rwx pool=volumes, allow rwx pool=vms, allow rx pool=images'
ceph auth get-or-create client.glance mon 'allow r' osd 'allow class-read object_prefix rbd_children, allow rwx pool=images'
Add the keyrings for client.cinder and client.glance to appropriate nodes and change their ownership
ceph auth get-or-create client.cinder | sudo tee /etc/ceph/ceph.client.cinder.keyring # On all cinder-volume nodes
sudo chown cinder:cinder /etc/ceph/ceph.client.cinder.keyring" # On all cinder-volume nodes
ceph auth get-or-create client.glance | sudo tee /etc/ceph/ceph.client.glance.keyring # On all glance-api nodes
sudo chown glance:glance /etc/ceph/ceph.client.glance.keyring" # On all glance-api nodes
Nodes running nova-compute need the keyring file for the nova-compute process
ceph auth get-or-create client.cinder | sudo tee /etc/ceph/ceph.client.cinder.keyring # On all nova-compute nodes
They also need to store the secret key of the client.cinder user in libvirt. The libvirt process needs it to access the cluster while attaching a block device from Cinder.
Create a temporary copy of the secret key on the nodes running nova-compute
ceph auth get-key client.cinder | tee client.cinder.key # On all nova-compute nodes
Then, on the compute nodes, add the secret key to libvirt and remove the temporary copy of the key(the uuid is the same as your --rbd-secret-uuid option, you have to save the uuid for later)
uuidgen
457eb676-33da-42ec-9a8c-9293d545c337
# The following steps on all nova-compute nodes
cat > secret.xml <<EOF
<secret ephemeral='no' private='no'>
<uuid>457eb676-33da-42ec-9a8c-9293d545c337</uuid>
<usage type='ceph'>
<name>client.cinder secret</name>
</usage>
</secret>
EOF
sudo virsh secret-define --file secret.xml
Secret 457eb676-33da-42ec-9a8c-9293d545c337 created
sudo virsh secret-set-value --secret 457eb676-33da-42ec-9a8c-9293d545c337 --base64 $(cat client.cinder.key) && rm client.cinder.key secret.xml
(optional)Now on every compute nodes edit your Ceph configuration file, add the client section
[client]
rbd cache = true
rbd cache writethrough until flush = true
rbd concurrent management ops = 20
[client.cinder]
keyring = /etc/ceph/ceph.client.cinder.keyring
(optional)On every glance-api nodes edit your Ceph configuration file, add the client section
[client.glance]
keyring= /etc/ceph/ceph.client.glance.keyring
(optional)If you want to remove osd
sudo stop ceph-mon-all && sudo stop ceph-osd-all # On osd node
ceph osd out {OSD-NUM}
ceph osd crush remove osd.{OSD-NUM}
ceph auth del osd.{OSD-NUM}
ceph osd rm {OSD-NUM}
ceph osd crush remove {HOST}
(optional)If you want to remove monitor
ceph mon remove {MON-ID}
Notes: you need to restart the nova-compute, cinder-volume and glance-api services to finalize the installation.
Create manila database and service credentials
ff --user ubuntu --hosts CONTROLLER1 openstack manila create-manila-db --root-db-pass CHANGEME --manila-db-pass CHANGEME
ff --user ubuntu --hosts CONTROLLER1 openstack manila create-service-credentials --os-password CHANGEME --os-auth-url http://CONTROLLER_VIP:35357/v3 --manila-pass CHANGEME --public-endpoint-v1 "http://CONTROLLER_VIP:8786/v1/%\(tenant_id\)s" --internal-endpoint-v1 "http://CONTROLLER_VIP:8786/v1/%\(tenant_id\)s" --admin-endpoint-v1 "http://CONTROLLER_VIP:8786/v1/%\(tenant_id\)s" --public-endpoint-v2 "http://CONTROLLER_VIP:8786/v2/%\(tenant_id\)s" --internal-endpoint-v2 "http://CONTROLLER_VIP:8786/v2/%\(tenant_id\)s" --admin-endpoint-v2 "http://CONTROLLER_VIP:8786/v2/%\(tenant_id\)s"
Install manila on CONTROLLER1 and CONTROLLER2
ff --user ubuntu --hosts CONTROLLER1 openstack manila install --connection mysql+pymysql://manila:CHANGEME@CONTROLLER_VIP/manila --auth-uri http://CONTROLLER_VIP:5000 --auth-url http://CONTROLLER_VIP:35357 --manila-pass CHANGEME --my-ip CONTROLLER1 --memcached-servers CONTROLLER1:11211,CONTROLLER2:11211 --rabbit-hosts CONTROLLER1,CONTROLLER2 --rabbit-user openstack --rabbit-pass CHANGEME --populate
ff --user ubuntu --hosts CONTROLLER2 openstack manila install --connection mysql+pymysql://manila:CHANGEME@CONTROLLER_VIP/manila --auth-uri http://CONTROLLER_VIP:5000 --auth-url http://CONTROLLER_VIP:35357 --manila-pass CHANGEME --my-ip CONTROLLER2 --memcached-servers CONTROLLER1:11211,CONTROLLER2:11211 --rabbit-hosts CONTROLLER1,CONTROLLER2 --rabbit-user openstack --rabbit-pass CHANGEME
Install manila share on CONTROLLER1 and CONTROLLER2
ff --user ubuntu --hosts CONTROLLER1 openstack manila-share install --connection mysql+pymysql://manila:CHANGEME@CONTROLLER_VIP/manila --auth-uri http://CONTROLLER_VIP:5000 --auth-url http://CONTROLLER_VIP:35357 --manila-pass CHANGEME --my-ip CONTROLLER1 --memcached-servers CONTROLLER1:11211,CONTROLLER2:11211 --rabbit-hosts CONTROLLER1,CONTROLLER2 --rabbit-user openstack --rabbit-pass CHANGEME --neutron-endpoint http://CONTROLLER_VIP:9696 --neutron-pass CHANGEME --nova-pass CHANGEME --cinder-pass CHANGEME
ff --user ubuntu --hosts CONTROLLER2 openstack manila-share install --connection mysql+pymysql://manila:CHANGEME@CONTROLLER_VIP/manila --auth-uri http://CONTROLLER_VIP:5000 --auth-url http://CONTROLLER_VIP:35357 --manila-pass CHANGEME --my-ip CONTROLLER2 --memcached-servers CONTROLLER1:11211,CONTROLLER2:11211 --rabbit-hosts CONTROLLER1,CONTROLLER2 --rabbit-user openstack --rabbit-pass CHANGEME --neutron-endpoint http://CONTROLLER_VIP:9696 --neutron-pass CHANGEME --nova-pass CHANGEME --cinder-pass CHANGEME
Create the service image for manila
http://docs.openstack.org/mitaka/install-guide-ubuntu/launch-instance-manila.html
Create shares with share servers management support
http://docs.openstack.org/mitaka/install-guide-ubuntu/launch-instance-manila-dhss-true-option2.html