diff options
Diffstat (limited to 'extraconfig')
21 files changed, 692 insertions, 52 deletions
diff --git a/extraconfig/all_nodes/mac_hostname.yaml b/extraconfig/all_nodes/mac_hostname.yaml index 739cbf0a..5883e06a 100644 --- a/extraconfig/all_nodes/mac_hostname.yaml +++ b/extraconfig/all_nodes/mac_hostname.yaml @@ -40,6 +40,7 @@ resources: CollectMacDeploymentsController: type: OS::Heat::SoftwareDeployments properties: + name: CollectMacDeploymentsController servers: {get_param: controller_servers} config: {get_resource: CollectMacConfig} actions: ['CREATE'] # Only do this on CREATE @@ -47,6 +48,7 @@ resources: CollectMacDeploymentsCompute: type: OS::Heat::SoftwareDeployments properties: + name: CollectMacDeploymentsCompute servers: {get_param: compute_servers} config: {get_resource: CollectMacConfig} actions: ['CREATE'] # Only do this on CREATE @@ -54,6 +56,7 @@ resources: CollectMacDeploymentsBlockStorage: type: OS::Heat::SoftwareDeployments properties: + name: CollectMacDeploymentsBlockStorage servers: {get_param: blockstorage_servers} config: {get_resource: CollectMacConfig} actions: ['CREATE'] # Only do this on CREATE @@ -61,6 +64,7 @@ resources: CollectMacDeploymentsObjectStorage: type: OS::Heat::SoftwareDeployments properties: + name: CollectMacDeploymentsObjectStorage servers: {get_param: objectstorage_servers} config: {get_resource: CollectMacConfig} actions: ['CREATE'] # Only do this on CREATE @@ -68,6 +72,7 @@ resources: CollectMacDeploymentsCephStorage: type: OS::Heat::SoftwareDeployments properties: + name: CollectMacDeploymentsCephStorage servers: {get_param: cephstorage_servers} config: {get_resource: CollectMacConfig} actions: ['CREATE'] # Only do this on CREATE @@ -95,6 +100,7 @@ resources: DistributeMacDeploymentsController: type: OS::Heat::SoftwareDeployments properties: + name: DistributeMacDeploymentsController servers: {get_param: controller_servers} config: {get_resource: DistributeMacConfig} input_values: diff --git a/extraconfig/all_nodes/random_string.yaml b/extraconfig/all_nodes/random_string.yaml index b4b30274..49d2d8b6 100644 --- a/extraconfig/all_nodes/random_string.yaml +++ b/extraconfig/all_nodes/random_string.yaml @@ -41,6 +41,7 @@ resources: RandomDeploymentsController: type: OS::Heat::SoftwareDeployments properties: + name: RandomDeploymentsController servers: {get_param: controller_servers} config: {get_resource: RandomConfig} actions: ['CREATE'] # Only do this on CREATE @@ -50,6 +51,7 @@ resources: RandomDeploymentsCompute: type: OS::Heat::SoftwareDeployments properties: + name: RandomDeploymentsCompute servers: {get_param: compute_servers} config: {get_resource: RandomConfig} actions: ['CREATE'] # Only do this on CREATE diff --git a/extraconfig/all_nodes/swap-partition.yaml b/extraconfig/all_nodes/swap-partition.yaml new file mode 100644 index 00000000..89a2adb0 --- /dev/null +++ b/extraconfig/all_nodes/swap-partition.yaml @@ -0,0 +1,90 @@ +heat_template_version: 2014-10-16 + +description: > + Extra config to add swap space to nodes. + +# Parameters passed from the parent template - note if you maintain +# out-of-tree templates they may require additional parameters if the +# in-tree templates add a new role. +parameters: + controller_servers: + type: json + compute_servers: + type: json + blockstorage_servers: + type: json + objectstorage_servers: + type: json + cephstorage_servers: + type: json + swap_partition_label: + type: string + description: Swap partition label + default: 'swap1' + + +resources: + + SwapConfig: + type: OS::Heat::SoftwareConfig + properties: + group: script + config: | + #!/bin/bash + set -eux + swap_partition=$(realpath /dev/disk/by-label/$swap_partition_label) + swapon $swap_partition + echo "$swap_partition swap swap defaults 0 0" >> /etc/fstab + inputs: + - name: swap_partition_label + description: Swap partition label + default: 'swap1' + + ControllerSwapDeployment: + type: OS::Heat::SoftwareDeploymentGroup + properties: + config: {get_resource: SwapConfig} + servers: {get_param: controller_servers} + input_values: + swap_partition_label: {get_param: swap_partition_label} + actions: ["CREATE"] + + ComputeSwapDeployment: + type: OS::Heat::SoftwareDeploymentGroup + properties: + config: {get_resource: SwapConfig} + servers: {get_param: compute_servers} + input_values: + swap_partition_label: {get_param: swap_partition_label} + actions: ["CREATE"] + + BlockStorageSwapDeployment: + type: OS::Heat::SoftwareDeploymentGroup + properties: + config: {get_resource: SwapConfig} + servers: {get_param: blockstorage_servers} + input_values: + swap_partition_label: {get_param: swap_partition_label} + actions: ["CREATE"] + + ObjectStorageSwapDeployment: + type: OS::Heat::SoftwareDeploymentGroup + properties: + config: {get_resource: SwapConfig} + servers: {get_param: objectstorage_servers} + input_values: + swap_partition_label: {get_param: swap_partition_label} + actions: ["CREATE"] + + CephStorageSwapDeployment: + type: OS::Heat::SoftwareDeploymentGroup + properties: + config: {get_resource: SwapConfig} + servers: {get_param: cephstorage_servers} + input_values: + swap_partition_label: {get_param: swap_partition_label} + actions: ["CREATE"] + +outputs: + config_identifier: + value: none diff --git a/extraconfig/all_nodes/swap.yaml b/extraconfig/all_nodes/swap.yaml new file mode 100644 index 00000000..374b1e5d --- /dev/null +++ b/extraconfig/all_nodes/swap.yaml @@ -0,0 +1,108 @@ +heat_template_version: 2014-10-16 + +description: > + Extra config to add swap space to nodes. + +# Parameters passed from the parent template - note if you maintain +# out-of-tree templates they may require additional parameters if the +# in-tree templates add a new role. +parameters: + controller_servers: + type: json + compute_servers: + type: json + blockstorage_servers: + type: json + objectstorage_servers: + type: json + cephstorage_servers: + type: json + swap_size_megabytes: + type: string + description: Amount of swap space to allocate in megabytes + default: '4096' + swap_path: + type: string + description: Full path to location of swap file + default: '/swap' + + +resources: + + SwapConfig: + type: OS::Heat::SoftwareConfig + properties: + group: script + config: | + #!/bin/bash + set -eux + if [ ! -f $swap_path ]; then + dd if=/dev/zero of=$swap_path count=$swap_size_megabytes bs=1M + chmod 0600 $swap_path + mkswap $swap_path + swapon $swap_path + else + echo "$swap_path already exists" + fi + echo "$swap_path swap swap defaults 0 0" >> /etc/fstab + inputs: + - name: swap_size_megabytes + description: Amount of swap space to allocate in megabytes + default: '4096' + - name: swap_path + description: Full path to location of swap file + default: '/swap' + + ControllerSwapDeployment: + type: OS::Heat::SoftwareDeploymentGroup + properties: + config: {get_resource: SwapConfig} + servers: {get_param: controller_servers} + input_values: + swap_size_megabytes: {get_param: swap_size_megabytes} + swap_path: {get_param: swap_path} + actions: ["CREATE"] + + ComputeSwapDeployment: + type: OS::Heat::SoftwareDeploymentGroup + properties: + config: {get_resource: SwapConfig} + servers: {get_param: compute_servers} + input_values: + swap_size_megabytes: {get_param: swap_size_megabytes} + swap_path: {get_param: swap_path} + actions: ["CREATE"] + + BlockStorageSwapDeployment: + type: OS::Heat::SoftwareDeploymentGroup + properties: + config: {get_resource: SwapConfig} + servers: {get_param: blockstorage_servers} + input_values: + swap_size_megabytes: {get_param: swap_size_megabytes} + swap_path: {get_param: swap_path} + actions: ["CREATE"] + + ObjectStorageSwapDeployment: + type: OS::Heat::SoftwareDeploymentGroup + properties: + config: {get_resource: SwapConfig} + servers: {get_param: objectstorage_servers} + input_values: + swap_size_megabytes: {get_param: swap_size_megabytes} + swap_path: {get_param: swap_path} + actions: ["CREATE"] + + CephStorageSwapDeployment: + type: OS::Heat::SoftwareDeploymentGroup + properties: + config: {get_resource: SwapConfig} + servers: {get_param: cephstorage_servers} + input_values: + swap_size_megabytes: {get_param: swap_size_megabytes} + swap_path: {get_param: swap_path} + actions: ["CREATE"] + +outputs: + config_identifier: + value: none diff --git a/extraconfig/post_deploy/example.yaml b/extraconfig/post_deploy/example.yaml index 1d3dca25..f83dff76 100644 --- a/extraconfig/post_deploy/example.yaml +++ b/extraconfig/post_deploy/example.yaml @@ -22,6 +22,7 @@ resources: ExtraDeployments: type: OS::Heat::SoftwareDeployments properties: + name: ExtraDeployments servers: {get_param: servers} config: {get_resource: ExtraConfig} actions: ['CREATE'] # Only do this on CREATE diff --git a/extraconfig/pre_deploy/rhel-registration/rhel-registration.yaml b/extraconfig/pre_deploy/rhel-registration/rhel-registration.yaml index d5160915..a884bdae 100644 --- a/extraconfig/pre_deploy/rhel-registration/rhel-registration.yaml +++ b/extraconfig/pre_deploy/rhel-registration/rhel-registration.yaml @@ -73,6 +73,7 @@ resources: RHELRegistrationDeployment: type: OS::Heat::SoftwareDeployment properties: + name: RHELRegistrationDeployment server: {get_param: server} config: {get_resource: RHELRegistration} actions: ['CREATE'] # Only do this on CREATE @@ -106,6 +107,7 @@ resources: RHELUnregistrationDeployment: type: OS::Heat::SoftwareDeployment properties: + name: RHELUnregistrationDeployment server: {get_param: server} config: {get_resource: RHELUnregistration} actions: ['DELETE'] # Only do this on DELETE diff --git a/extraconfig/pre_deploy/rhel-registration/scripts/rhel-registration b/extraconfig/pre_deploy/rhel-registration/scripts/rhel-registration index cbbd6a1d..76fa63b4 100644 --- a/extraconfig/pre_deploy/rhel-registration/scripts/rhel-registration +++ b/extraconfig/pre_deploy/rhel-registration/scripts/rhel-registration @@ -13,6 +13,7 @@ fi opts= attach_opts= +sat5_opts= repos="repos --enable rhel-7-server-rpms" satellite_repo="rhel-7-server-rh-common-rpms" if [ -n "${REG_AUTO_ATTACH:-}" ]; then @@ -49,6 +50,7 @@ fi if [ -n "${REG_FORCE:-}" ]; then opts="$opts --force" + sat5_opts="$sat5_opts --force" fi if [ -n "${REG_SERVER_URL:-}" ]; then @@ -57,6 +59,7 @@ fi if [ -n "${REG_ACTIVATION_KEY:-}" ]; then opts="$opts --activationkey=$REG_ACTIVATION_KEY" + sat5_opts="$sat5_opts --activationkey=$REG_ACTIVATION_KEY" if [ -z "${REG_ORG:-}" ]; then echo "WARNING: REG_ACTIVATION_KEY set without REG_ORG." @@ -75,10 +78,12 @@ fi if [ -n "${REG_MACHINE_NAME:-}" ]; then opts="$opts --name $REG_MACHINE_NAME" + sat5_opts="$sat5_opts --profilename=$REG_MACHINE_NAME" fi if [ -n "${REG_ORG:-}" ]; then opts="$opts --org=$REG_ORG" + sat5_opts="$sat5_opts --systemorgid=$REG_ORG" fi if [ -n "${REG_REPOS:-}" ]; then @@ -91,6 +96,20 @@ if [ -n "${REG_TYPE:-}" ]; then opts="$opts --type=$REG_TYPE" fi +function detect_satellite_version { + ping_api=$REG_SAT_URL/katello/api/ping + if curl -k -s -D - -o /dev/null $ping_api | grep "200 OK"; then + echo Satellite 6 detected at $REG_SAT_URL + satellite_version=6 + elif curl -k -s -D - -o /dev/null $REG_SAT_URL/rhn/Login.do | grep "200 OK"; then + echo Satellite 5 detected at $REG_SAT_URL + satellite_version=5 + else + echo No Satellite detected at $REG_SAT_URL + exit 1 + fi +} + case "${REG_METHOD:-}" in portal) subscription-manager register $opts @@ -100,13 +119,21 @@ case "${REG_METHOD:-}" in subscription-manager $repos ;; satellite) - repos="$repos --enable ${satellite_repo}" - rpm -Uvh "$REG_SAT_URL/pub/katello-ca-consumer-latest.noarch.rpm" || true - subscription-manager register $opts - subscription-manager $repos - yum install -y katello-agent || true # needed for errata reporting to satellite6 - katello-package-upload - subscription-manager repos --disable ${satellite_repo} + detect_satellite_version + if [ "satellite_version" = "6" ]; then + repos="$repos --enable ${satellite_repo}" + rpm -Uvh "$REG_SAT_URL/pub/katello-ca-consumer-latest.noarch.rpm" || true + subscription-manager register $opts + subscription-manager $repos + yum install -y katello-agent || true # needed for errata reporting to satellite6 + katello-package-upload + subscription-manager repos --disable ${satellite_repo} + else + pushd /usr/share/rhn/ + curl -k -O $REG_SAT_URL/pub/RHN-ORG-TRUSTED-SSL-CERT + popd + rhnreg_ks --serverUrl=$REG_SAT_URL/XMLRPC $sat5_opts + fi ;; disable) echo "Disabling RHEL registration" diff --git a/extraconfig/pre_deploy/rhel-registration/scripts/rhel-unregistration b/extraconfig/pre_deploy/rhel-registration/scripts/rhel-unregistration index 1e72e0a6..916f97e3 100644 --- a/extraconfig/pre_deploy/rhel-registration/scripts/rhel-unregistration +++ b/extraconfig/pre_deploy/rhel-registration/scripts/rhel-unregistration @@ -7,6 +7,8 @@ case "${REG_METHOD:-}" in portal|satellite) # Allow unregistration to fail. # We don't want to fail stack deletes if unregistration fails. + # Note that this will be a no-op on satellite 5, which doesn't support + # unregistering from the cli. subscription-manager unregister || true subscription-manager clean || true ;; diff --git a/extraconfig/tasks/major_upgrade_compute.sh b/extraconfig/tasks/major_upgrade_compute.sh new file mode 100644 index 00000000..78628c8c --- /dev/null +++ b/extraconfig/tasks/major_upgrade_compute.sh @@ -0,0 +1,26 @@ +#!/bin/bash +# +# This delivers the compute upgrade script to be invoked as part of the tripleo +# major upgrade workflow. +# +set -eu + +UPGRADE_SCRIPT=/root/tripleo_upgrade_node.sh + +cat > $UPGRADE_SCRIPT << ENDOFCAT +### DO NOT MODIFY THIS FILE +### This file is automatically delivered to the compute nodes as part of the +### tripleo upgrades workflow + +# pin nova to kilo (messaging +-1) for the nova-compute service + +crudini --set /etc/nova/nova.conf upgrade_levels compute $upgrade_level_nova_compute + +yum -y install python-zaqarclient # needed for os-collect-config +yum -y update + +ENDOFCAT + +# ensure the permissions are OK +chmod 0755 $UPGRADE_SCRIPT + diff --git a/extraconfig/tasks/major_upgrade_controller_pacemaker_1.sh b/extraconfig/tasks/major_upgrade_controller_pacemaker_1.sh new file mode 100755 index 00000000..bf2ee330 --- /dev/null +++ b/extraconfig/tasks/major_upgrade_controller_pacemaker_1.sh @@ -0,0 +1,55 @@ +#!/bin/bash + +set -eu + +cluster_sync_timeout=600 + +if pcs status 2>&1 | grep -E '(cluster is not currently running)|(OFFLINE:)'; then + echo_error "ERROR: upgrade cannot start with some cluster nodes being offline" + exit 1 +fi + +if [ "$(hiera -c /etc/puppet/hiera.yaml bootstrap_nodeid)" = "$(facter hostname)" ]; then + pcs resource disable httpd + check_resource httpd stopped 1800 + if pcs status | grep openstack-keystone; then + pcs resource disable openstack-keystone + check_resource openstack-keystone stopped 1800 + fi + pcs resource disable redis + check_resource redis stopped 600 + pcs resource disable mongod + check_resource mongod stopped 600 + pcs resource disable rabbitmq + check_resource rabbitmq stopped 600 + pcs resource disable memcached + check_resource memcached stopped 600 + pcs resource disable galera + check_resource galera stopped 600 + pcs cluster stop --all +fi + +# Swift isn't controled by pacemaker +systemctl_swift stop + +tstart=$(date +%s) +while systemctl is-active pacemaker; do + sleep 5 + tnow=$(date +%s) + if (( tnow-tstart > cluster_sync_timeout )) ; then + echo_error "ERROR: cluster shutdown timed out" + exit 1 + fi +done + +yum -y install python-zaqarclient # needed for os-collect-config +yum -y -q update + +# Pin messages sent to compute nodes to kilo, these will be upgraded later +crudini --set /etc/nova/nova.conf upgrade_levels compute "$upgrade_level_nova_compute" +# https://bugzilla.redhat.com/show_bug.cgi?id=1284047 +# Change-Id: Ib3f6c12ff5471e1f017f28b16b1e6496a4a4b435 +crudini --set /etc/ceilometer/ceilometer.conf DEFAULT rpc_backend rabbit +# https://bugzilla.redhat.com/show_bug.cgi?id=1284058 +# Ifd1861e3df46fad0e44ff9b5cbd58711bbc87c97 Swift Ceilometer middleware no longer exists +crudini --set /etc/swift/proxy-server.conf pipeline:main pipeline "catch_errors healthcheck cache ratelimit tempurl formpost authtoken keystone staticweb proxy-logging proxy-server" diff --git a/extraconfig/tasks/major_upgrade_controller_pacemaker_2.sh b/extraconfig/tasks/major_upgrade_controller_pacemaker_2.sh new file mode 100755 index 00000000..10bea573 --- /dev/null +++ b/extraconfig/tasks/major_upgrade_controller_pacemaker_2.sh @@ -0,0 +1,67 @@ +#!/bin/bash + +set -eu + +cluster_form_timeout=600 +cluster_settle_timeout=600 +galera_sync_timeout=600 + +if [ "$(hiera -c /etc/puppet/hiera.yaml bootstrap_nodeid)" = "$(facter hostname)" ]; then + pcs cluster start --all + + tstart=$(date +%s) + while pcs status 2>&1 | grep -E '(cluster is not currently running)|(OFFLINE:)'; do + sleep 5 + tnow=$(date +%s) + if (( tnow-tstart > cluster_form_timeout )) ; then + echo_error "ERROR: timed out forming the cluster" + exit 1 + fi + done + + if ! timeout -k 10 $cluster_settle_timeout crm_resource --wait; then + echo_error "ERROR: timed out waiting for cluster to finish transition" + exit 1 + fi + + pcs resource enable galera + check_resource galera started 600 + pcs resource enable mongod + check_resource mongod started 600 + + tstart=$(date +%s) + while ! clustercheck; do + sleep 5 + tnow=$(date +%s) + if (( tnow-tstart > galera_sync_timeout )) ; then + echo_error "ERROR galera sync timed out" + exit 1 + fi + done + + # Run all the db syncs + # TODO: check if this can be triggered in puppet and removed from here + ceilometer-dbsync --config-file=/etc/ceilometer/ceilometer.conf + cinder-manage db sync + glance-manage --config-file=/etc/glance/glance-registry.conf db_sync + heat-manage --config-file /etc/heat/heat.conf db_sync + keystone-manage db_sync + neutron-db-manage --config-file /etc/neutron/neutron.conf --config-file /etc/neutron/plugin.ini upgrade head + nova-manage db sync + + pcs resource enable memcached + check_resource memcached started 600 + pcs resource enable rabbitmq + check_resource rabbitmq started 600 + pcs resource enable redis + check_resource redis started 600 + if pcs status | grep openstack-keystone; then + pcs resource enable openstack-keystone + check_resource openstack-keystone started 1800 + fi + pcs resource enable httpd + check_resource httpd started 1800 +fi + +# Swift isn't controled by heat +systemctl_swift start diff --git a/extraconfig/tasks/major_upgrade_object_storage.sh b/extraconfig/tasks/major_upgrade_object_storage.sh new file mode 100644 index 00000000..0f6d091e --- /dev/null +++ b/extraconfig/tasks/major_upgrade_object_storage.sh @@ -0,0 +1,38 @@ +#!/bin/bash +# +# This delivers the swift-storage upgrade script to be invoked as part of the tripleo +# major upgrade workflow. +# +set -eu + +UPGRADE_SCRIPT=/root/tripleo_upgrade_node.sh + +cat > $UPGRADE_SCRIPT << ENDOFCAT +### DO NOT MODIFY THIS FILE +### This file is automatically delivered to the swift-storage nodes as part of the +### tripleo upgrades workflow + + +function systemctl_swift { + action=$1 + for S in openstack-swift-account-auditor openstack-swift-account-reaper openstack-swift-account-replicator openstack-swift-account \ + openstack-swift-container-auditor openstack-swift-container-replicator openstack-swift-container-updater openstack-swift-container \ + openstack-swift-object-auditor openstack-swift-object-replicator openstack-swift-object-updater openstack-swift-object openstack-swift-proxy; do + systemctl $action $S + done +} + + +systemctl_swift stop + +yum -y update + +systemctl_swift start + + + +ENDOFCAT + +# ensure the permissions are OK +chmod 0755 $UPGRADE_SCRIPT + diff --git a/extraconfig/tasks/major_upgrade_pacemaker.yaml b/extraconfig/tasks/major_upgrade_pacemaker.yaml new file mode 100644 index 00000000..b867d107 --- /dev/null +++ b/extraconfig/tasks/major_upgrade_pacemaker.yaml @@ -0,0 +1,71 @@ +heat_template_version: 2014-10-16 +description: 'Upgrade for Pacemaker deployments' + +parameters: + controller_servers: + type: json + compute_servers: + type: json + blockstorage_servers: + type: json + objectstorage_servers: + type: json + cephstorage_servers: + type: json + input_values: + type: json + description: input values for the software deployments + + UpgradeLevelNovaCompute: + type: string + description: Nova Compute upgrade level + default: '' + +resources: + # TODO(jistr): for Mitaka->Newton upgrades and further we can use + # map_merge with input_values instead of feeding params into scripts + # via str_replace on bash snippets + + ControllerPacemakerUpgradeConfig_Step1: + type: OS::Heat::SoftwareConfig + properties: + group: script + config: + list_join: + - '' + - - str_replace: + template: | + #!/bin/bash + upgrade_level_nova_compute='UPGRADE_LEVEL_NOVA_COMPUTE' + params: + UPGRADE_LEVEL_NOVA_COMPUTE: {get_param: UpgradeLevelNovaCompute} + - get_file: pacemaker_common_functions.sh + - get_file: major_upgrade_pacemaker_migrations.sh + - get_file: major_upgrade_controller_pacemaker_1.sh + + ControllerPacemakerUpgradeDeployment_Step1: + type: OS::Heat::SoftwareDeploymentGroup + properties: + servers: {get_param: controller_servers} + config: {get_resource: ControllerPacemakerUpgradeConfig_Step1} + input_values: {get_param: input_values} + + ControllerPacemakerUpgradeConfig_Step2: + type: OS::Heat::SoftwareConfig + properties: + group: script + config: + list_join: + - '' + - - get_file: pacemaker_common_functions.sh + - get_file: major_upgrade_pacemaker_migrations.sh + - get_file: major_upgrade_controller_pacemaker_2.sh + + ControllerPacemakerUpgradeDeployment_Step2: + type: OS::Heat::SoftwareDeploymentGroup + depends_on: ControllerPacemakerUpgradeDeployment_Step1 + properties: + servers: {get_param: controller_servers} + config: {get_resource: ControllerPacemakerUpgradeConfig_Step2} + input_values: {get_param: input_values} + diff --git a/extraconfig/tasks/major_upgrade_pacemaker_migrations.sh b/extraconfig/tasks/major_upgrade_pacemaker_migrations.sh new file mode 100644 index 00000000..7fd26945 --- /dev/null +++ b/extraconfig/tasks/major_upgrade_pacemaker_migrations.sh @@ -0,0 +1,14 @@ +#!/bin/bash + +# Special pieces of upgrade migration logic go into this +# file. E.g. Pacemaker cluster transitions for existing deployments, +# matching changes to overcloud_controller_pacemaker.pp (Puppet +# handles deployment, this file handles migrations). +# +# This file shouldn't execute any action on its own, all logic should +# be wrapped into bash functions. Upgrade scripts will source this +# file and call the functions defined in this file where appropriate. +# +# The migration functions should be idempotent. If the migration has +# been already applied, it should be possible to call the function +# again without damaging the deployment or failing the upgrade. diff --git a/extraconfig/tasks/major_upgrade_script_delivery.yaml b/extraconfig/tasks/major_upgrade_script_delivery.yaml new file mode 100644 index 00000000..f7faa7fc --- /dev/null +++ b/extraconfig/tasks/major_upgrade_script_delivery.yaml @@ -0,0 +1,65 @@ +heat_template_version: 2014-10-16 +description: 'Upgrade for Pacemaker deployments' + +parameters: + + controller_servers: + type: json + compute_servers: + type: json + blockstorage_servers: + type: json + objectstorage_servers: + type: json + cephstorage_servers: + type: json + input_values: + type: json + description: input values for the software deployments + + UpgradeLevelNovaCompute: + type: string + description: Nova Compute upgrade level + default: '' + +resources: + # TODO(jistr): for Mitaka->Newton upgrades and further we can use + # map_merge with input_values instead of feeding params into scripts + # via str_replace on bash snippets + + ComputeDeliverUpgradeScriptConfig: + type: OS::Heat::SoftwareConfig + properties: + group: script + config: + list_join: + - '' + - - str_replace: + template: | + #!/bin/bash + upgrade_level_nova_compute='UPGRADE_LEVEL_NOVA_COMPUTE' + params: + UPGRADE_LEVEL_NOVA_COMPUTE: {get_param: UpgradeLevelNovaCompute} + - get_file: major_upgrade_compute.sh + + ComputeDeliverUpgradeScriptDeployment: + type: OS::Heat::SoftwareDeploymentGroup + properties: + servers: {get_param: compute_servers} + config: {get_resource: ComputeDeliverUpgradeScriptConfig} + input_values: {get_param: input_values} + + + ObjectStoreDeliverUpgradeScriptConfig: + type: OS::Heat::SoftwareConfig + properties: + group: script + config: {get_file: major_upgrade_object_storage.sh} + + ObjectStoreDeliverUpgradeScriptDeployment: + type: OS::Heat::SoftwareDeploymentGroup + properties: + servers: {get_param: objectstorage_servers} + config: {get_resource: ObjectStoreDeliverUpgradeScriptConfig} + input_values: {get_param: input_values} + diff --git a/extraconfig/tasks/noop.yaml b/extraconfig/tasks/noop.yaml index 0cff7469..dbb863be 100644 --- a/extraconfig/tasks/noop.yaml +++ b/extraconfig/tasks/noop.yaml @@ -4,6 +4,22 @@ description: 'No-op task' parameters: servers: type: json + default: [] + controller_servers: + type: json + default: [] + compute_servers: + type: json + default: [] + blockstorage_servers: + type: json + default: [] + objectstorage_servers: + type: json + default: [] + cephstorage_servers: + type: json + default: [] input_values: type: json default: {} diff --git a/extraconfig/tasks/pacemaker_common_functions.sh b/extraconfig/tasks/pacemaker_common_functions.sh new file mode 100755 index 00000000..ee3216e4 --- /dev/null +++ b/extraconfig/tasks/pacemaker_common_functions.sh @@ -0,0 +1,48 @@ +#!/bin/bash + +set -eu + +function check_resource { + + if [ "$#" -ne 3 ]; then + echo_error "ERROR: check_resource function expects 3 parameters, $# given" + exit 1 + fi + + service=$1 + state=$2 + timeout=$3 + + if [ "$state" = "stopped" ]; then + match_for_incomplete='Started' + else # started + match_for_incomplete='Stopped' + fi + + if timeout -k 10 $timeout crm_resource --wait; then + node_states=$(pcs status --full | grep "$service" | grep -v Clone) + if echo "$node_states" | grep -q "$match_for_incomplete"; then + echo_error "ERROR: cluster finished transition but $service was not in $state state, exiting." + exit 1 + else + echo "$service has $state" + fi + else + echo_error "ERROR: cluster remained unstable for more than $timeout seconds, exiting." + exit 1 + fi + +} + +function echo_error { + echo "$@" | tee /dev/fd2 +} + +function systemctl_swift { + action=$1 + for S in openstack-swift-account-auditor openstack-swift-account-reaper openstack-swift-account-replicator openstack-swift-account \ + openstack-swift-container-auditor openstack-swift-container-replicator openstack-swift-container-updater openstack-swift-container \ + openstack-swift-object-auditor openstack-swift-object-replicator openstack-swift-object-updater openstack-swift-object openstack-swift-proxy; do + systemctl $action $S + done +} diff --git a/extraconfig/tasks/pacemaker_resource_restart.sh b/extraconfig/tasks/pacemaker_resource_restart.sh index 4e8b20fd..b2bdc55a 100755 --- a/extraconfig/tasks/pacemaker_resource_restart.sh +++ b/extraconfig/tasks/pacemaker_resource_restart.sh @@ -3,37 +3,6 @@ set -eux pacemaker_status=$(systemctl is-active pacemaker) -check_interval=3 - -function check_resource { - - service=$1 - state=$2 - timeout=$3 - tstart=$(date +%s) - tend=$(( $tstart + $timeout )) - - if [ "$state" = "stopped" ]; then - match_for_incomplete='Started' - else # started - match_for_incomplete='Stopped' - fi - - while (( $(date +%s) < $tend )); do - node_states=$(pcs status --full | grep "$service" | grep -v Clone) - if echo "$node_states" | grep -q "$match_for_incomplete"; then - echo "$service not yet $state, sleeping $check_interval seconds." - sleep $check_interval - else - echo "$service has $state" - return - fi - done - - echo "$service never $state after $timeout seconds" | tee /dev/fd/2 - exit 1 - -} # Run if pacemaker is running, we're the bootstrap node, # and we're updating the deployment (not creating). @@ -50,7 +19,7 @@ if [ "$pacemaker_status" = "active" -a \ pcs resource disable httpd check_resource httpd stopped 300 pcs resource disable openstack-keystone - check_resource openstack-keystone stopped 1200 + check_resource openstack-keystone stopped 1800 if pcs status | grep haproxy-clone; then pcs resource restart haproxy-clone @@ -62,7 +31,7 @@ if [ "$pacemaker_status" = "active" -a \ pcs resource restart galera-master pcs resource enable openstack-keystone - check_resource openstack-keystone started 300 + check_resource openstack-keystone started 1800 pcs resource enable httpd check_resource httpd started 800 diff --git a/extraconfig/tasks/post_puppet_pacemaker.yaml b/extraconfig/tasks/post_puppet_pacemaker.yaml index 7de41d94..fbed9ce5 100644 --- a/extraconfig/tasks/post_puppet_pacemaker.yaml +++ b/extraconfig/tasks/post_puppet_pacemaker.yaml @@ -33,7 +33,11 @@ resources: type: OS::Heat::SoftwareConfig properties: group: script - config: {get_file: pacemaker_resource_restart.sh} + config: + list_join: + - '' + - - get_file: pacemaker_common_functions.sh + - get_file: pacemaker_resource_restart.sh ControllerPostPuppetRestartDeployment: type: OS::Heat::SoftwareDeployments diff --git a/extraconfig/tasks/yum_update.sh b/extraconfig/tasks/yum_update.sh index 2d6b8cc2..59e4be45 100755 --- a/extraconfig/tasks/yum_update.sh +++ b/extraconfig/tasks/yum_update.sh @@ -123,10 +123,13 @@ openstack-nova-scheduler" echo "Setting resource start/stop timeouts" for service in $SERVICES; do - pcs -f $pacemaker_dumpfile resource update $service op start timeout=100s op stop timeout=100s + pcs -f $pacemaker_dumpfile resource update $service op start timeout=200s op stop timeout=200s done # mongod start timeout is higher, setting only stop timeout - pcs -f $pacemaker_dumpfile resource update mongod op stop timeout=100s + pcs -f $pacemaker_dumpfile resource update mongod op start timeout=370s op stop timeout=200s + + echo "Making sure rabbitmq has the notify=true meta parameter" + pcs -f $pacemaker_dumpfile resource update rabbitmq meta notify=true echo "Applying new Pacemaker config" if ! pcs cluster cib-push $pacemaker_dumpfile; then @@ -150,15 +153,15 @@ openstack-nova-scheduler" kill $(ps ax | grep -e "keepalived.*\.pid-vrrp" | awk '{print $1}') 2>/dev/null || : kill $(ps ax | grep -e "radvd.*\.pid\.radvd" | awk '{print $1}') 2>/dev/null || : else - echo "Excluding upgrading packages that are handled by config management tooling" - command_arguments="$command_arguments --skip-broken" - for exclude in $(cat /var/lib/tripleo/installed-packages/* | sort -u); do - command_arguments="$command_arguments --exclude $exclude" - done + echo "Upgrading openstack-puppet-modules" + yum -q -y update openstack-puppet-modules + echo "Upgrading other packages is handled by config management tooling" + echo -n "true" > $heat_outputs_path.update_managed_packages + exit 0 fi command=${command:-update} -full_command="yum -y $command $command_arguments" +full_command="yum -q -y $command $command_arguments" echo "Running: $full_command" result=$($full_command) @@ -199,9 +202,6 @@ if [[ "$pacemaker_status" == "active" ]] ; then fi pcs status - -else - echo -n "true" > $heat_outputs_path.update_managed_packages fi echo "Finished yum_update.sh on server $deploy_server_id at `date`" diff --git a/extraconfig/tasks/yum_update_noop.yaml b/extraconfig/tasks/yum_update_noop.yaml new file mode 100644 index 00000000..b759d9c5 --- /dev/null +++ b/extraconfig/tasks/yum_update_noop.yaml @@ -0,0 +1,29 @@ +heat_template_version: 2014-10-16 +description: 'No-op yum update task' + +resources: + + config: + type: OS::Heat::SoftwareConfig + properties: + group: script + config: | + #!/bin/bash + echo -n "false" > $heat_outputs_path.update_managed_packages + inputs: + - name: update_identifier + description: yum will only run for previously unused values of update_identifier + default: '' + - name: command + description: yum sub-command to run, defaults to "update" + default: update + - name: command_arguments + description: yum command arguments, defaults to "" + default: '' + outputs: + - name: update_managed_packages + description: boolean value indicating whether to upgrade managed packages + +outputs: + OS::stack_id: + value: {get_resource: config} |