[ English | English (United Kingdom) | Indonesia | 한국어 (대한민국) | Deutsch ]

Exercise the Cloud

Once OpenStack-Helm has been deployed, the cloud can be exercised either with the OpenStack client, or the same heat templates that are used in the validation gates.

#!/bin/bash

export OS_CLOUD=openstack_helm

: ${OSH_EXT_NET_NAME:="public"}
: ${OSH_EXT_SUBNET_NAME:="public-subnet"}
: ${OSH_EXT_SUBNET:="172.24.4.0/24"}
: ${OSH_BR_EX_ADDR:="172.24.4.1/24"}
openstack stack create --wait \
  --parameter network_name=${OSH_EXT_NET_NAME} \
  --parameter physical_network_name=public \
  --parameter subnet_name=${OSH_EXT_SUBNET_NAME} \
  --parameter subnet_cidr=${OSH_EXT_SUBNET} \
  --parameter subnet_gateway=${OSH_BR_EX_ADDR%/*} \
  -t ./tools/gate/files/heat-public-net-deployment.yaml \
  heat-public-net-deployment

: ${OSH_PRIVATE_SUBNET_POOL:="10.0.0.0/8"}
: ${OSH_PRIVATE_SUBNET_POOL_NAME:="shared-default-subnetpool"}
: ${OSH_PRIVATE_SUBNET_POOL_DEF_PREFIX:="24"}
openstack stack create --wait \
  --parameter subnet_pool_name=${OSH_PRIVATE_SUBNET_POOL_NAME} \
  --parameter subnet_pool_prefixes=${OSH_PRIVATE_SUBNET_POOL} \
  --parameter subnet_pool_default_prefix_length=${OSH_PRIVATE_SUBNET_POOL_DEF_PREFIX} \
  -t ./tools/gate/files/heat-subnet-pool-deployment.yaml \
  heat-subnet-pool-deployment

: ${OSH_EXT_NET_NAME:="public"}
: ${OSH_VM_KEY_STACK:="heat-vm-key"}
: ${OSH_PRIVATE_SUBNET:="10.0.0.0/24"}
# NOTE(portdirect): We do this fancy, and seemingly pointless, footwork to get
# the full image name for the cirros Image without having to be explicit.
IMAGE_NAME=$(openstack image show -f value -c name \
  $(openstack image list -f csv | awk -F ',' '{ print $2 "," $1 }' | \
    grep "^\"Cirros" | head -1 | awk -F ',' '{ print $2 }' | tr -d '"'))

# Setup SSH Keypair in Nova
mkdir -p ${HOME}/.ssh
openstack keypair create --private-key ${HOME}/.ssh/osh_key ${OSH_VM_KEY_STACK}
chmod 600 ${HOME}/.ssh/osh_key

openstack stack create --wait \
    --parameter public_net=${OSH_EXT_NET_NAME} \
    --parameter image="${IMAGE_NAME}" \
    --parameter ssh_key=${OSH_VM_KEY_STACK} \
    --parameter cidr=${OSH_PRIVATE_SUBNET} \
    --parameter dns_nameserver=${OSH_BR_EX_ADDR%/*} \
    -t ./tools/gate/files/heat-basic-vm-deployment.yaml \
    heat-basic-vm-deployment

FLOATING_IP=$(openstack stack output show \
    heat-basic-vm-deployment \
    floating_ip \
    -f value -c output_value)

function wait_for_ssh_port {
  # Default wait timeout is 300 seconds
  set +x
  end=$(date +%s)
  if ! [ -z $2 ]; then
   end=$((end + $2))
  else
   end=$((end + 300))
  fi
  while true; do
      # Use Nmap as its the same on Ubuntu and RHEL family distros
      nmap -Pn -p22 $1 | awk '$1 ~ /22/ {print $2}' | grep -q 'open' && \
          break || true
      sleep 1
      now=$(date +%s)
      [ $now -gt $end ] && echo "Could not connect to $1 port 22 in time" && exit -1
  done
  set -x
}
wait_for_ssh_port $FLOATING_IP

# SSH into the VM and check it can reach the outside world
ssh-keyscan "$FLOATING_IP" >> ~/.ssh/known_hosts
ssh -i ${HOME}/.ssh/osh_key cirros@${FLOATING_IP} ping -q -c 1 -W 2 ${OSH_BR_EX_ADDR%/*}

# Check the VM can reach the metadata server
ssh -i ${HOME}/.ssh/osh_key cirros@${FLOATING_IP} curl --verbose --connect-timeout 5 169.254.169.254

# Check the VM can reach the keystone server
ssh -i ${HOME}/.ssh/osh_key cirros@${FLOATING_IP} curl --verbose --connect-timeout 5 keystone.openstack.svc.cluster.local

# Check to see if cinder has been deployed, if it has then perform a volume attach.
if openstack service list -f value -c Type | grep -q "^volume"; then
  INSTANCE_ID=$(openstack stack output show \
      heat-basic-vm-deployment \
      instance_uuid \
      -f value -c output_value)

  # Get the devices that are present on the instance
  DEVS_PRE_ATTACH=$(mktemp)
  ssh -i ${HOME}/.ssh/osh_key cirros@${FLOATING_IP} lsblk > ${DEVS_PRE_ATTACH}

  # Create and attach a block device to the instance
  openstack stack create --wait \
    --parameter instance_uuid=${INSTANCE_ID} \
    -t ./tools/gate/files/heat-vm-volume-attach.yaml \
    heat-vm-volume-attach

  # Get the devices that are present on the instance
  DEVS_POST_ATTACH=$(mktemp)
  ssh -i ${HOME}/.ssh/osh_key cirros@${FLOATING_IP} lsblk > ${DEVS_POST_ATTACH}

  # Check that we have the expected number of extra devices on the instance post attach
  if ! [ "$(comm -13 ${DEVS_PRE_ATTACH} ${DEVS_POST_ATTACH} | wc -l)" -eq "1" ]; then
    echo "Volume not successfully attached"
    exit 1
  fi
fi

Alternatively, this step can be performed by running the script directly:

./tools/deployment/developer/common/900-use-it.sh

To run further commands from the CLI manually, execute the following to set up authentication credentials:

export OS_CLOUD=openstack_helm

Note that this command will only enable you to auth successfully using the python-openstackclient CLI. To use legacy clients like the python-novaclient from the CLI, reference the auth values in /etc/openstack/clouds.yaml and run:

export OS_USERNAME='admin'
export OS_PASSWORD='password'
export OS_PROJECT_NAME='admin'
export OS_PROJECT_DOMAIN_NAME='default'
export OS_USER_DOMAIN_NAME='default'
export OS_AUTH_URL='http://keystone.openstack.svc.cluster.local/v3'

The example above uses the default values used by openstack-helm-infra.