diff options
Diffstat (limited to 'extraconfig')
38 files changed, 771 insertions, 523 deletions
diff --git a/extraconfig/all_nodes/mac_hostname.j2.yaml b/extraconfig/all_nodes/mac_hostname.j2.yaml index 75ffc9e6..fcf022ae 100644 --- a/extraconfig/all_nodes/mac_hostname.j2.yaml +++ b/extraconfig/all_nodes/mac_hostname.j2.yaml @@ -1,4 +1,4 @@ -heat_template_version: 2014-10-16 +heat_template_version: ocata description: > Example extra config for cluster config diff --git a/extraconfig/all_nodes/random_string.j2.yaml b/extraconfig/all_nodes/random_string.j2.yaml index 9ce2ca8a..77d4b381 100644 --- a/extraconfig/all_nodes/random_string.j2.yaml +++ b/extraconfig/all_nodes/random_string.j2.yaml @@ -1,4 +1,4 @@ -heat_template_version: 2014-10-16 +heat_template_version: ocata description: > Example extra config for cluster config diff --git a/extraconfig/all_nodes/swap-partition.j2.yaml b/extraconfig/all_nodes/swap-partition.j2.yaml index 36076b0c..b6fef79f 100644 --- a/extraconfig/all_nodes/swap-partition.j2.yaml +++ b/extraconfig/all_nodes/swap-partition.j2.yaml @@ -1,11 +1,7 @@ -heat_template_version: 2014-10-16 +heat_template_version: ocata -description: > - Extra config to add swap space to nodes. +description: Template file to add a swap partition to a node. -# 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: servers: type: json @@ -14,9 +10,7 @@ parameters: description: Swap partition label default: 'swap1' - resources: - SwapConfig: type: OS::Heat::SoftwareConfig properties: @@ -25,8 +19,13 @@ resources: #!/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 + if [ -f "$swap_partition" ]; then + swapon $swap_partition + echo "$swap_partition swap swap defaults 0 0" >> /etc/fstab + else + echo "$swap_partition needs to be a valid path" + echo "Check that $swap_partition_label is a valid partition label" + fi inputs: - name: swap_partition_label description: Swap partition label diff --git a/extraconfig/all_nodes/swap.j2.yaml b/extraconfig/all_nodes/swap.j2.yaml index ce65dacb..044f817c 100644 --- a/extraconfig/all_nodes/swap.j2.yaml +++ b/extraconfig/all_nodes/swap.j2.yaml @@ -1,11 +1,7 @@ -heat_template_version: 2014-10-16 +heat_template_version: ocata -description: > - Extra config to add swap space to nodes. +description: Template file to add a swap file to a node. -# 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: servers: type: json @@ -18,9 +14,7 @@ parameters: description: Full path to location of swap file default: '/swap' - resources: - SwapConfig: type: OS::Heat::SoftwareConfig properties: diff --git a/extraconfig/nova_metadata/krb-service-principals.yaml b/extraconfig/nova_metadata/krb-service-principals.yaml new file mode 100644 index 00000000..c66e6460 --- /dev/null +++ b/extraconfig/nova_metadata/krb-service-principals.yaml @@ -0,0 +1,84 @@ +heat_template_version: ocata +description: 'Generates the relevant service principals for a server' + +parameters: + RoleData: + type: json + description: the list containing the 'role_data' output for the ServiceChain + + # Coming from parameter_defaults + CloudName: + default: overcloud.localdomain + description: The DNS name of this cloud. E.g. ci-overcloud.tripleo.org + type: string + CloudNameInternal: + default: overcloud.internalapi.localdomain + description: > + The DNS name of this cloud's internal API endpoint. E.g. + 'ci-overcloud.internalapi.tripleo.org'. + type: string + CloudNameStorage: + default: overcloud.storage.localdomain + description: > + The DNS name of this cloud's storage endpoint. E.g. + 'ci-overcloud.storage.tripleo.org'. + type: string + CloudNameStorageManagement: + default: overcloud.storagemgmt.localdomain + description: > + The DNS name of this cloud's storage management endpoint. E.g. + 'ci-overcloud.storagemgmt.tripleo.org'. + type: string + CloudNameCtlplane: + default: overcloud.ctlplane.localdomain + description: > + The DNS name of this cloud's storage management endpoint. E.g. + 'ci-overcloud.management.tripleo.org'. + type: string + +resources: + + IncomingMetadataSettings: + type: OS::Heat::Value + properties: + value: + yaql: + # Filter null values and values that contain don't contain + # 'metadata_settings', get the values from that key and get the + # unique ones. + expression: list($.data.where($ != null).where($.containsKey('metadata_settings')).metadata_settings.flatten().distinct()) + data: {get_param: RoleData} + + # Generates entries for nova metadata with the following format: + # 'managed_service_<id>' : <service>/<fqdn> + # Depending on the requested network + IndividualServices: + type: OS::Heat::Value + properties: + value: + yaql: + expression: let(fqdns => $.data.fqdns) -> dict($.data.metadata.where($ != null and $.type = 'vip').select([concat('managed_service_', $.service, $.network), concat($.service, '/', $fqdns.get($.network))])) + data: + metadata: {get_attr: [IncomingMetadataSettings, value]} + fqdns: + external: {get_param: CloudName} + internal_api: {get_param: CloudNameInternal} + storage: {get_param: CloudNameStorage} + storage_mgmt: {get_param: CloudNameStorageManagement} + ctlplane: {get_param: CloudNameCtlplane} + + CompactServices: + type: OS::Heat::Value + properties: + value: + yaql: + expression: dict($.data.where($ != null and $.type = 'node').select([$.service, $.network.replace('_', '')]).groupBy($[0], $[1])) + data: {get_attr: [IncomingMetadataSettings, value]} + +outputs: + metadata: + description: actual metadata entries that will be passed to the server. + value: + map_merge: + - {get_attr: [IndividualServices, value]} + - compact_services: {get_attr: [CompactServices, value]} diff --git a/extraconfig/post_deploy/default.yaml b/extraconfig/post_deploy/default.yaml index ddfe0243..4da54ead 100644 --- a/extraconfig/post_deploy/default.yaml +++ b/extraconfig/post_deploy/default.yaml @@ -1,4 +1,4 @@ -heat_template_version: 2014-10-16 +heat_template_version: ocata description: 'Extra Post Deployment Config' parameters: servers: diff --git a/extraconfig/post_deploy/example.yaml b/extraconfig/post_deploy/example.yaml index f83dff76..8ac7eb73 100644 --- a/extraconfig/post_deploy/example.yaml +++ b/extraconfig/post_deploy/example.yaml @@ -1,4 +1,4 @@ -heat_template_version: 2014-10-16 +heat_template_version: ocata description: > Example extra config for post-deployment diff --git a/extraconfig/post_deploy/example_run_on_update.yaml b/extraconfig/post_deploy/example_run_on_update.yaml index 234488af..738e263b 100644 --- a/extraconfig/post_deploy/example_run_on_update.yaml +++ b/extraconfig/post_deploy/example_run_on_update.yaml @@ -1,4 +1,4 @@ -heat_template_version: 2014-10-16 +heat_template_version: ocata description: > Example extra config for post-deployment, this re-runs every update diff --git a/extraconfig/post_deploy/undercloud_post.sh b/extraconfig/post_deploy/undercloud_post.sh new file mode 100755 index 00000000..8bcae1d3 --- /dev/null +++ b/extraconfig/post_deploy/undercloud_post.sh @@ -0,0 +1,126 @@ +#!/bin/bash +set -eux + +ln -sf /etc/puppet/hiera.yaml /etc/hiera.yaml + + +# WRITE OUT STACKRC +if [ ! -e /root/stackrc ]; then + touch /root/stackrc + chmod 0600 /root/stackrc + +cat >> /root/stackrc <<-EOF_CAT +export OS_PASSWORD=$admin_password +export OS_AUTH_URL=$auth_url +export OS_USERNAME=admin +export OS_TENANT_NAME=admin +export COMPUTE_API_VERSION=1.1 +export NOVA_VERSION=1.1 +export OS_BAREMETAL_API_VERSION=1.15 +export OS_NO_CACHE=True +export OS_CLOUDNAME=undercloud +EOF_CAT + + if [ -n "$ssl_certificate" ]; then +cat >> /root/stackrc <<-EOF_CAT +export PYTHONWARNINGS="ignore:Certificate has no, ignore:A true SSLContext object is not available" +EOF_CAT + fi +fi + +source /root/stackrc + +if [ ! -f /root/.ssh/authorized_keys ]; then + sudo mkdir -p /root/.ssh + sudo chmod 7000 /root/.ssh/ + sudo touch /root/.ssh/authorized_keys + sudo chmod 600 /root/.ssh/authorized_keys +fi + +if [ ! -f /root/.ssh/id_rsa ]; then + ssh-keygen -b 1024 -N '' -f /root/.ssh/id_rsa +fi + +if ! grep "$(cat /root/.ssh/id_rsa.pub)" /root/.ssh/authorized_keys; then + cat /root/.ssh/id_rsa.pub >> /root/.ssh/authorized_keys +fi + +PHYSICAL_NETWORK=ctlplane + +ctlplane_id=$(openstack network list -f csv -c ID -c Name --quote none | tail -n +2 | grep ctlplane | cut -d, -f1) +subnet_ids=$(openstack subnet list -f csv -c ID --quote none | tail -n +2) +subnet_id= + +for subnet_id in $subnet_ids; do + network_id=$(openstack subnet show -f value -c network_id $subnet_id) + if [ "$network_id" = "$ctlplane_id" ]; then + break + fi +done + +net_create=1 +if [ -n "$subnet_id" ]; then + cidr=$(openstack subnet show $subnet_id -f value -c cidr) + if [ "$cidr" = "$undercloud_network_cidr" ]; then + net_create=0 + else + echo "New cidr $undercloud_network_cidr does not equal old cidr $cidr" + echo "Will attempt to delete and recreate subnet $subnet_id" + fi +fi + +if [ "$net_create" -eq "1" ]; then + # Delete the subnet and network to make sure it doesn't already exist + if openstack subnet list | grep start; then + openstack subnet delete $(openstack subnet list | grep start | awk '{print $4}') + fi + if openstack network show ctlplane; then + openstack network delete ctlplane + fi + + + NETWORK_ID=$(openstack network create --provider-network-type=flat --provider-physical-network=ctlplane ctlplane | grep " id " | awk '{print $4}') + + NAMESERVER_ARG="" + if [ -n "${undercloud_nameserver:-}" ]; then + NAMESERVER_ARG="--dns-nameserver $undercloud_nameserver" + fi + + openstack subnet create --network=$NETWORK_ID \ + --gateway=$undercloud_network_gateway \ + --subnet-range=$undercloud_network_cidr \ + --allocation-pool start=$undercloud_dhcp_start,end=$undercloud_dhcp_end \ + --host-route destination=169.254.169.254/32,gateway=$local_ip \ + $NAMESERVER_ARG ctlplane +fi + +# Disable nova quotas +openstack quota set --cores -1 --instances -1 --ram -1 $(openstack project show admin | awk '$2=="id" {print $4}') + +# MISTRAL WORKFLOW CONFIGURATION +if [ "$(hiera mistral_api_enabled)" = "true" ]; then + # load workflows + for workbook in $(openstack workbook list | grep tripleo | cut -f 2 -d ' '); do + openstack workbook delete $workbook + done + for workflow in $(openstack workflow list | grep tripleo | cut -f 2 -d ' '); do + openstack workflow delete $workflow + done + for workbook in $(ls /usr/share/openstack-tripleo-common/workbooks/*); do + openstack workbook create $workbook + done + + # Store the SNMP password in a mistral environment + if ! openstack workflow env show tripleo.undercloud-config &>/dev/null; then + TMP_MISTRAL_ENV=$(mktemp) + echo "{\"name\": \"tripleo.undercloud-config\", \"variables\": {\"undercloud_ceilometer_snmpd_password\": \"$snmp_readonly_user_password\"}}" > $TMP_MISTRAL_ENV + openstack workflow env create $TMP_MISTRAL_ENV + fi + +fi + +# IP forwarding is needed to allow the overcloud nodes access to the outside +# internet in cases where they are on an isolated network. +sysctl -w net.ipv4.ip_forward=1 +# Make it persistent +echo "net.ipv4.ip_forward=1" > /etc/sysctl.d/ip-forward.conf diff --git a/extraconfig/post_deploy/undercloud_post.yaml b/extraconfig/post_deploy/undercloud_post.yaml new file mode 100644 index 00000000..38a9181e --- /dev/null +++ b/extraconfig/post_deploy/undercloud_post.yaml @@ -0,0 +1,93 @@ +heat_template_version: ocata + +description: > + Post-deployment for the TripleO undercloud + +parameters: + servers: + type: json + DeployedServerPortMap: + default: {} + type: json + UndercloudDhcpRangeStart: + type: string + default: '192.168.24.5' + UndercloudDhcpRangeEnd: + type: string + default: '192.168.24.24' + UndercloudNetworkCidr: + type: string + default: '192.168.24.0/24' + UndercloudNetworkGateway: + type: string + default: '192.168.24.1' + UndercloudNameserver: + type: string + default: '' + AdminPassword: #supplied by tripleo-undercloud-passwords.yaml + type: string + description: The password for the keystone admin account, used for monitoring, querying neutron etc. + hidden: True + SSLCertificate: + description: > + The content of the SSL certificate (without Key) in PEM format. + type: string + default: "" + hidden: True + SnmpdReadonlyUserPassword: + description: The user password for SNMPd with readonly rights running on all Overcloud nodes + type: string + hidden: true + +conditions: + + ssl_disabled: {equals : [{get_param: SSLCertificate}, ""]} + +resources: + + UndercloudPostConfig: + type: OS::Heat::SoftwareConfig + properties: + group: script + inputs: + - name: deploy_identifier + - name: local_ip + - name: undercloud_dhcp_start + - name: undercloud_dhcp_end + - name: undercloud_network_cidr + - name: undercloud_network_gateway + - name: undercloud_nameserver + - name: admin_password + - name: auth_url + - name: snmp_readonly_user_password + config: {get_file: ./undercloud_post.sh} + + UndercloudPostDeployment: + type: OS::Heat::SoftwareDeployments + properties: + servers: {get_param: servers} + config: {get_resource: UndercloudPostConfig} + input_values: + local_ip: {get_param: [DeployedServerPortMap, 'control_virtual_ip', fixed_ips, 0, ip_address]} + undercloud_dhcp_start: {get_param: UndercloudDhcpRangeStart} + undercloud_dhcp_end: {get_param: UndercloudDhcpRangeEnd} + undercloud_network_cidr: {get_param: UndercloudNetworkCidr} + undercloud_network_gateway: {get_param: UndercloudNetworkGateway} + undercloud_nameserver: {get_param: UndercloudNameserver} + ssl_certificate: {get_param: SSLCertificate} + admin_password: {get_param: AdminPassword} + snmp_readonly_user_password: {get_param: SnmpdReadonlyUserPassword} + # if SSL is enabled we use the public virtual ip as the stackrc endpoint + auth_url: + if: + - ssl_disabled + - list_join: + - '' + - - 'http://' + - {get_param: [DeployedServerPortMap, 'control_virtual_ip', fixed_ips, 0, ip_address]} + - ':5000/v2.0' + - list_join: + - '' + - - 'https://' + - {get_param: [DeployedServerPortMap, 'public_virtual_ip', fixed_ips, 0, ip_address]} + - ':13000/v2.0' diff --git a/extraconfig/pre_deploy/rhel-registration/rhel-registration.yaml b/extraconfig/pre_deploy/rhel-registration/rhel-registration.yaml index 7c65bd8b..fdf2e957 100644 --- a/extraconfig/pre_deploy/rhel-registration/rhel-registration.yaml +++ b/extraconfig/pre_deploy/rhel-registration/rhel-registration.yaml @@ -1,4 +1,4 @@ -heat_template_version: 2014-10-16 +heat_template_version: ocata description: > RHEL Registration and unregistration software deployments. diff --git a/extraconfig/pre_deploy/rhel-registration/scripts/rhel-registration b/extraconfig/pre_deploy/rhel-registration/scripts/rhel-registration index 71ab0767..2650a967 100644 --- a/extraconfig/pre_deploy/rhel-registration/scripts/rhel-registration +++ b/extraconfig/pre_deploy/rhel-registration/scripts/rhel-registration @@ -11,6 +11,7 @@ if [ -e $OK ] ; then exit 0 fi +retryCount=0 opts= attach_opts= sat5_opts= @@ -96,12 +97,28 @@ if [ -n "${REG_TYPE:-}" ]; then opts="$opts --type=$REG_TYPE" fi +function retry() { + if [[ $retryCount < 3 ]]; then + $@ + if ! [[ $? == 0 ]]; then + retryCount=$(echo $retryCount + 1 | bc) + echo "WARN: Failed to connect when running '$@', retrying..." + retry $@ + else + retryCount=0 + fi + else + echo "ERROR: Failed to connect after 3 attempts when running '$@'" + exit 1 + fi +} + function detect_satellite_version { ping_api=$REG_SAT_URL/katello/api/ping - if curl -L -k -s -D - -o /dev/null $ping_api | grep "200 OK"; then + if curl --retry 3 --retry-delay 10 --max-time 30 -L -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 -L -k -s -D - -o /dev/null $REG_SAT_URL/rhn/Login.do | grep "200 OK"; then + elif curl --retry 3 --retry-delay 10 --max-time 30 -L -k -s -D - -o /dev/null $REG_SAT_URL/rhn/Login.do | grep "200 OK"; then echo Satellite 5 detected at $REG_SAT_URL satellite_version=5 else @@ -112,29 +129,29 @@ function detect_satellite_version { case "${REG_METHOD:-}" in portal) - subscription-manager register $opts + retry subscription-manager register $opts if [ -z "${REG_AUTO_ATTACH:-}" -a -z "${REG_ACTIVATION_KEY:-}" ]; then - subscription-manager attach $attach_opts + retry subscription-manager attach $attach_opts fi - subscription-manager repos --disable '*' - subscription-manager $repos + retry subscription-manager repos --disable '*' + retry subscription-manager $repos ;; satellite) detect_satellite_version if [ "$satellite_version" = "6" ]; then repos="$repos --enable ${satellite_repo}" - curl -L -k -O "$REG_SAT_URL/pub/katello-ca-consumer-latest.noarch.rpm" + curl --retry 3 --retry-delay 10 --max-time 30 -L -k -O "$REG_SAT_URL/pub/katello-ca-consumer-latest.noarch.rpm" rpm -Uvh 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 + retry subscription-manager register $opts + retry subscription-manager $repos + retry yum install -y katello-agent || true # needed for errata reporting to satellite6 katello-package-upload - subscription-manager repos --disable ${satellite_repo} + retry subscription-manager repos --disable ${satellite_repo} else pushd /usr/share/rhn/ - curl -k -O $REG_SAT_URL/pub/RHN-ORG-TRUSTED-SSL-CERT + curl --retry 3 --retry-delay 10 --max-time 30 -k -O $REG_SAT_URL/pub/RHN-ORG-TRUSTED-SSL-CERT popd - rhnreg_ks --serverUrl=$REG_SAT_URL/XMLRPC $sat5_opts + retry rhnreg_ks --serverUrl=$REG_SAT_URL/XMLRPC $sat5_opts fi ;; disable) diff --git a/extraconfig/pre_network/ansible_host_config.ansible b/extraconfig/pre_network/ansible_host_config.ansible new file mode 100644 index 00000000..c126c1a1 --- /dev/null +++ b/extraconfig/pre_network/ansible_host_config.ansible @@ -0,0 +1,58 @@ +--- +- name: Configuration to be applied before rebooting the node + connection: local + hosts: localhost + + tasks: + # Kernel Args Configuration + - block: + - name: Ensure the kernel args ( {{ _KERNEL_ARGS_ }} ) is present as TRIPLEO_HEAT_TEMPLATE_KERNEL_ARGS + lineinfile: + dest: /etc/default/grub + regexp: '^TRIPLEO_HEAT_TEMPLATE_KERNEL_ARGS.*' + insertafter: '^GRUB_CMDLINE_LINUX.*' + line: 'TRIPLEO_HEAT_TEMPLATE_KERNEL_ARGS=" {{ _KERNEL_ARGS_ }} "' + - name: Add TRIPLEO_HEAT_TEMPLATE_KERNEL_ARGS to the GRUB_CMDLINE_LINUX parameter + lineinfile: + dest: /etc/default/grub + line: 'GRUB_CMDLINE_LINUX="${GRUB_CMDLINE_LINUX:+$GRUB_CMDLINE_LINUX }${TRIPLEO_HEAT_TEMPLATE_KERNEL_ARGS}"' + insertafter: '^TRIPLEO_HEAT_TEMPLATE_KERNEL_ARGS.*' + - name: Generate grub config file + command: grub2-mkconfig -o /boot/grub2/grub.cfg + become: true + when: _KERNEL_ARGS_|default("") != "" + + # Tune-d Configuration + - block: + - name: Tune-d Configuration + lineinfile: + dest: /etc/tuned/cpu-partitioning-variables.conf + regexp: '^isolated_cores=.*' + line: 'isolated_cores={{ _HOST_CPUS_LIST_ }}' + when: _HOST_CPUS_LIST_|default("") != "" + + - name: Tune-d provile activation + shell: tuned-adm profile {{ _TUNED_PROFILE_NAME_ }} + become: true + when: _TUNED_PROFILE_NAME_|default("") != "" + + # Provisioning Network workaround + # The script will be executed before os-net-config, in which case, only Provisioning network will have IP + # BOOTPROTO of all interface config files (except provisioning), will be set to "none" to avoid reboot failing to acquire IP on other networks + - block: + - find: + paths: /etc/sysconfig/network-scripts/ + patterns: ifcfg-* + register: ifcfg_files + + - replace: + dest: "{{ item.path }}" + regexp: '^BOOTPROTO=.*' + replace: 'BOOTPROTO=none' + when: + - item.path | regex_replace('(^.*ifcfg-)(.*)', '\\2') != "lo" + # This condition will list all the interfaces except the one with valid IP (which is Provisioning network at this stage) + # Simpler Version - hostvars[inventory_hostname]['ansible_' + iface_name ]['ipv4']['address'] is undefined + - hostvars[inventory_hostname]['ansible_' + item.path | regex_replace('(^.*ifcfg-)(.*)', '\\2') ]['ipv4']['address'] is undefined + with_items: + - "{{ ifcfg_files.files }}" diff --git a/extraconfig/pre_network/config_then_reboot.yaml b/extraconfig/pre_network/config_then_reboot.yaml new file mode 100644 index 00000000..bb0b9511 --- /dev/null +++ b/extraconfig/pre_network/config_then_reboot.yaml @@ -0,0 +1,48 @@ +heat_template_version: ocata + +description: > + Do some configuration, then reboot - sometimes needed for early-boot + changes such as modifying kernel configuration + +parameters: + server: + type: string + +resources: + + SomeConfig: + type: OS::Heat::SoftwareConfig + properties: + group: script + config: | + #!/bin/bash + echo "did some config before reboot" > /root/pre-reboot-config + + SomeDeployment: + type: OS::Heat::SoftwareDeployment + properties: + name: SomeDeployment + server: {get_param: server} + config: {get_resource: SomeConfig} + actions: ['CREATE'] # Only do this on CREATE + + RebootConfig: + type: OS::Heat::SoftwareConfig + properties: + group: script + config: | + #!/bin/bash + # Stop os-collect-config to avoid any race collecting another + # deployment before reboot happens + systemctl stop os-collect-config.service + /sbin/reboot + + RebootDeployment: + type: OS::Heat::SoftwareDeployment + depends_on: SomeDeployment + properties: + name: RebootDeployment + server: {get_param: server} + config: {get_resource: RebootConfig} + actions: ['CREATE'] # Only do this on CREATE + signal_transport: NO_SIGNAL diff --git a/extraconfig/pre_network/host_config_and_reboot.role.j2.yaml b/extraconfig/pre_network/host_config_and_reboot.role.j2.yaml new file mode 100644 index 00000000..4ad53cb8 --- /dev/null +++ b/extraconfig/pre_network/host_config_and_reboot.role.j2.yaml @@ -0,0 +1,100 @@ +heat_template_version: ocata + +description: > + Do some configuration, then reboot - sometimes needed for early-boot + changes such as modifying kernel configuration + +parameters: + server: + type: string + {{role}}KernelArgs: + type: string + default: "" + {{role}}TunedProfileName: + type: string + default: "" + {{role}}HostCpusList: + type: string + default: "" + +conditions: + param_exists: + or: + - not: + equals: + - get_param: {{role}}KernelArgs + - "" + - not: + equals: + - get_param: {{role}}TunedProfileName + - "" + +resources: + + HostParametersConfig: + type: OS::Heat::SoftwareConfig + condition: param_exists + properties: + group: ansible + inputs: + - name: _KERNEL_ARGS_ + - name: _TUNED_PROFILE_NAME_ + - name: _HOST_CPUS_LIST_ + outputs: + - name: result + config: + get_file: ansible_host_config.ansible + + HostParametersDeployment: + type: OS::Heat::SoftwareDeployment + condition: param_exists + properties: + name: HostParametersDeployment + server: {get_param: server} + config: {get_resource: HostParametersConfig} + actions: ['CREATE'] # Only do this on CREATE + input_values: + _KERNEL_ARGS_: {get_param: {{role}}KernelArgs} + _TUNED_PROFILE_NAME_: {get_param: {{role}}TunedProfileName} + _HOST_CPUS_LIST_: {get_param: {{role}}HostCpusList} + + RebootConfig: + type: OS::Heat::SoftwareConfig + condition: param_exists + properties: + group: script + config: | + #!/bin/bash + # Stop os-collect-config to avoid any race collecting another + # deployment before reboot happens + systemctl stop os-collect-config.service + /sbin/reboot + + RebootDeployment: + type: OS::Heat::SoftwareDeployment + condition: param_exists + depends_on: HostParametersDeployment + properties: + name: RebootDeployment + server: {get_param: server} + config: {get_resource: RebootConfig} + actions: ['CREATE'] # Only do this on CREATE + signal_transport: NO_SIGNAL + +outputs: + result: + value: + get_attr: [HostParametersDeployment, result] + condition: param_exists + stdout: + value: + get_attr: [HostParametersDeployment, deploy_stdout] + condition: param_exists + stderr: + value: + get_attr: [HostParametersDeployment, deploy_stderr] + condition: param_exists + status_code: + value: + get_attr: [HostParametersDeployment, deploy_status_code] + condition: param_exists diff --git a/extraconfig/tasks/major_upgrade_block_storage.sh b/extraconfig/tasks/major_upgrade_block_storage.sh deleted file mode 100644 index 39861826..00000000 --- a/extraconfig/tasks/major_upgrade_block_storage.sh +++ /dev/null @@ -1,22 +0,0 @@ -#!/bin/bash -# -# This runs an upgrade of Cinder Block Storage nodes. -# -set -eu - -# Special-case OVS for https://bugs.launchpad.net/tripleo/+bug/1635205 -if [[ -n $(rpm -q --scripts openvswitch | awk '/postuninstall/,/*/' | grep "systemctl.*try-restart") ]]; then - echo "Manual upgrade of openvswitch - restart in postun detected" - mkdir OVS_UPGRADE || true - pushd OVS_UPGRADE - echo "Attempting to downloading latest openvswitch with yumdownloader" - yumdownloader --resolve openvswitch - echo "Updating openvswitch with nopostun option" - rpm -U --replacepkgs --nopostun ./*.rpm - popd -else - echo "Skipping manual upgrade of openvswitch - no restart in postun detected" -fi - -yum -y install python-zaqarclient # needed for os-collect-config -yum -y -q update diff --git a/extraconfig/tasks/major_upgrade_ceilometer_wsgi_mitaka_newton.yaml b/extraconfig/tasks/major_upgrade_ceilometer_wsgi_mitaka_newton.yaml index c87e6824..cf5d7a84 100644 --- a/extraconfig/tasks/major_upgrade_ceilometer_wsgi_mitaka_newton.yaml +++ b/extraconfig/tasks/major_upgrade_ceilometer_wsgi_mitaka_newton.yaml @@ -1,4 +1,4 @@ -heat_template_version: 2014-10-16 +heat_template_version: ocata description: > Software-config for ceilometer configuration under httpd during upgrades diff --git a/extraconfig/tasks/major_upgrade_ceph_mon.sh b/extraconfig/tasks/major_upgrade_ceph_mon.sh deleted file mode 100755 index e0d160f1..00000000 --- a/extraconfig/tasks/major_upgrade_ceph_mon.sh +++ /dev/null @@ -1,82 +0,0 @@ -#!/bin/bash -set -eu -set -o pipefail - -echo INFO: starting $(basename "$0") - -# Exit if not running -if ! pidof ceph-mon &> /dev/null; then - echo INFO: ceph-mon is not running, skipping - exit 0 -fi - -# Exit if not Hammer -INSTALLED_VERSION=$(ceph --version | awk '{print $3}') -if ! [[ "$INSTALLED_VERSION" =~ ^0\.94.* ]]; then - echo INFO: version of Ceph installed is not 0.94, skipping - exit 0 -fi - -CEPH_STATUS=$(ceph health | awk '{print $1}') -if [ ${CEPH_STATUS} = HEALTH_ERR ]; then - echo ERROR: Ceph cluster status is HEALTH_ERR, cannot be upgraded - exit 1 -fi - -# Useful when upgrading with OSDs num < replica size -if [[ ${ignore_ceph_upgrade_warnings:-False} != [Tt]rue ]]; then - timeout 300 bash -c "while [ ${CEPH_STATUS} != HEALTH_OK ]; do - echo WARNING: Waiting for Ceph cluster status to go HEALTH_OK; - sleep 30; - CEPH_STATUS=$(ceph health | awk '{print $1}') - done" -fi - -MON_PID=$(pidof ceph-mon) -MON_ID=$(hostname -s) - -# Stop daemon using Hammer sysvinit script -service ceph stop mon.${MON_ID} - -# Ensure it's stopped -timeout 60 bash -c "while kill -0 ${MON_PID} 2> /dev/null; do - sleep 2; -done" - -# Update to Jewel -yum -y -q update ceph-mon ceph - -# Restart/Exit if not on Jewel, only in that case we need the changes -UPDATED_VERSION=$(ceph --version | awk '{print $3}') -if [[ "$UPDATED_VERSION" =~ ^0\.94.* ]]; then - echo WARNING: Ceph was not upgraded, restarting daemons - service ceph start mon.${MON_ID} -elif [[ "$UPDATED_VERSION" =~ ^10\.2.* ]]; then - # RPM could own some of these but we can't take risks on the pre-existing files - for d in /var/lib/ceph/mon /var/log/ceph /var/run/ceph /etc/ceph; do - chown -L -R ceph:ceph $d || echo WARNING: chown of $d failed - done - - # Replay udev events with newer rules - udevadm trigger - - # Enable systemd unit - systemctl enable ceph-mon.target - systemctl enable ceph-mon@${MON_ID} - systemctl start ceph-mon@${MON_ID} - - # Wait for daemon to be back in the quorum - timeout 300 bash -c "until (ceph quorum_status | jq .quorum_names | grep -sq ${MON_ID}); do - echo WARNING: Waiting for mon.${MON_ID} to re-join quorum; - sleep 10; - done" - - # if tunables become legacy, cluster status will be HEALTH_WARN causing - # upgrade to fail on following node - ceph osd crush tunables default - - echo INFO: Ceph was upgraded to Jewel -else - echo ERROR: Ceph was upgraded to an unknown release, daemon is stopped, need manual intervention - exit 1 -fi diff --git a/extraconfig/tasks/major_upgrade_ceph_storage.sh b/extraconfig/tasks/major_upgrade_ceph_storage.sh deleted file mode 100644 index d84cad45..00000000 --- a/extraconfig/tasks/major_upgrade_ceph_storage.sh +++ /dev/null @@ -1,116 +0,0 @@ -#!/bin/bash -# -# This delivers the ceph-storage upgrade script to be invoked as part of the tripleo -# major upgrade workflow. -# -set -eu -set -o pipefail - -UPGRADE_SCRIPT=/root/tripleo_upgrade_node.sh - -cat > $UPGRADE_SCRIPT << 'ENDOFCAT' -#!/bin/bash -### DO NOT MODIFY THIS FILE -### This file is automatically delivered to the ceph-storage nodes as part of the -### tripleo upgrades workflow -set -eu - -echo INFO: starting $(basename "$0") - -# Exit if not running -if ! pidof ceph-osd &> /dev/null; then - echo INFO: ceph-osd is not running, skipping - exit 0 -fi - -# Exit if not Hammer -INSTALLED_VERSION=$(ceph --version | awk '{print $3}') -if ! [[ "$INSTALLED_VERSION" =~ ^0\.94.* ]]; then - echo INFO: version of Ceph installed is not 0.94, skipping - exit 0 -fi - -OSD_PIDS=$(pidof ceph-osd) -OSD_IDS=$(ls /var/lib/ceph/osd | awk 'BEGIN { FS = "-" } ; { print $2 }') - -# "so that mirrors aren't rebalanced as if the OSD died" - gfidente / leseb -ceph osd set noout -ceph osd set norebalance -ceph osd set nodeep-scrub -ceph osd set noscrub - -# Stop daemon using Hammer sysvinit script -for OSD_ID in $OSD_IDS; do - service ceph stop osd.${OSD_ID} -done - -# Nice guy will return non-0 only when all failed -timeout 60 bash -c "while kill -0 ${OSD_PIDS} 2> /dev/null; do - sleep 2; -done" - -# Special-case OVS for https://bugs.launchpad.net/tripleo/+bug/1635205 -if [[ -n $(rpm -q --scripts openvswitch | awk '/postuninstall/,/*/' | grep "systemctl.*try-restart") ]]; then - echo "Manual upgrade of openvswitch - restart in postun detected" - mkdir OVS_UPGRADE || true - pushd OVS_UPGRADE - echo "Attempting to downloading latest openvswitch with yumdownloader" - yumdownloader --resolve openvswitch - echo "Updating openvswitch with nopostun option" - rpm -U --replacepkgs --nopostun ./*.rpm - popd -else - echo "Skipping manual upgrade of openvswitch - no restart in postun detected" -fi - -# Update (Ceph to Jewel) -yum -y install python-zaqarclient # needed for os-collect-config -yum -y update - -# Restart/Exit if not on Jewel, only in that case we need the changes -UPDATED_VERSION=$(ceph --version | awk '{print $3}') -if [[ "$UPDATED_VERSION" =~ ^0\.94.* ]]; then - echo WARNING: Ceph was not upgraded, restarting daemon - for OSD_ID in $OSD_IDS; do - service ceph start osd.${OSD_ID} - done -elif [[ "$UPDATED_VERSION" =~ ^10\.2.* ]]; then - # RPM could own some of these but we can't take risks on the pre-existing files - for d in /var/lib/ceph/osd /var/log/ceph /var/run/ceph /etc/ceph; do - chown -L -R ceph:ceph $d || echo WARNING: chown of $d failed - done - - # Replay udev events with newer rules - udevadm trigger && udevadm settle - - # If on ext4, we need to enforce lower values for name and namespace len - # or ceph-osd will refuse to start, see: http://tracker.ceph.com/issues/16187 - for OSD_ID in $OSD_IDS; do - OSD_FS=$(findmnt -n -o FSTYPE -T /var/lib/ceph/osd/ceph-${OSD_ID}) - if [ ${OSD_FS} = ext4 ]; then - crudini --set /etc/ceph/ceph.conf global osd_max_object_name_len 256 - crudini --set /etc/ceph/ceph.conf global osd_max_object_namespace_len 64 - fi - done - - # Enable systemd unit - systemctl enable ceph-osd.target - for OSD_ID in $OSD_IDS; do - systemctl enable ceph-osd@${OSD_ID} - systemctl start ceph-osd@${OSD_ID} - done - - echo INFO: Ceph was upgraded to Jewel -else - echo ERROR: Ceph was upgraded to an unknown release, daemon is stopped, need manual intervention - exit 1 -fi - -ceph osd unset noout -ceph osd unset norebalance -ceph osd unset nodeep-scrub -ceph osd unset noscrub -ENDOFCAT - -# ensure the permissions are OK -chmod 0755 $UPGRADE_SCRIPT diff --git a/extraconfig/tasks/major_upgrade_compute.sh b/extraconfig/tasks/major_upgrade_compute.sh deleted file mode 100644 index b0d42806..00000000 --- a/extraconfig/tasks/major_upgrade_compute.sh +++ /dev/null @@ -1,46 +0,0 @@ -#!/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 - -set -eu - -# pin nova to kilo (messaging +-1) for the nova-compute service - -crudini --set /etc/nova/nova.conf upgrade_levels compute $upgrade_level_nova_compute - -# Special-case OVS for https://bugs.launchpad.net/tripleo/+bug/1635205 -if [[ -n \$(rpm -q --scripts openvswitch | awk '/postuninstall/,/*/' | grep "systemctl.*try-restart") ]]; then - echo "Manual upgrade of openvswitch - restart in postun detected" - mkdir OVS_UPGRADE || true - pushd OVS_UPGRADE - echo "Attempting to downloading latest openvswitch with yumdownloader" - yumdownloader --resolve openvswitch - echo "Updating openvswitch with nopostun option" - rpm -U --replacepkgs --nopostun ./*.rpm - popd -else - echo "Skipping manual upgrade of openvswitch - no restart in postun detected" -fi - -yum -y install python-zaqarclient # needed for os-collect-config -yum -y update - -# Due to bug#1640177 we need to restart compute agent -echo "Restarting openstack ceilometer agent compute" -systemctl restart openstack-ceilometer-compute - -ENDOFCAT - -# ensure the permissions are OK -chmod 0755 $UPGRADE_SCRIPT - diff --git a/extraconfig/tasks/major_upgrade_controller_pacemaker_2.sh b/extraconfig/tasks/major_upgrade_controller_pacemaker_2.sh index 7cc6735f..6bfe1239 100755 --- a/extraconfig/tasks/major_upgrade_controller_pacemaker_2.sh +++ b/extraconfig/tasks/major_upgrade_controller_pacemaker_2.sh @@ -100,18 +100,7 @@ if [ $DO_MYSQL_UPGRADE -eq 1 ]; then fi # Special-case OVS for https://bugs.launchpad.net/tripleo/+bug/1635205 -if [[ -n $(rpm -q --scripts openvswitch | awk '/postuninstall/,/*/' | grep "systemctl.*try-restart") ]]; then - echo "Manual upgrade of openvswitch - restart in postun detected" - mkdir OVS_UPGRADE || true - pushd OVS_UPGRADE - echo "Attempting to downloading latest openvswitch with yumdownloader" - yumdownloader --resolve openvswitch - echo "Updating openvswitch with nopostun option" - rpm -U --replacepkgs --nopostun ./*.rpm - popd -else - echo "Skipping manual upgrade of openvswitch - no restart in postun detected" -fi +special_case_ovs_upgrade_if_needed yum -y install python-zaqarclient # needed for os-collect-config yum -y -q update diff --git a/extraconfig/tasks/major_upgrade_controller_pacemaker_3.sh b/extraconfig/tasks/major_upgrade_controller_pacemaker_3.sh index 6748f891..a3cbd945 100755 --- a/extraconfig/tasks/major_upgrade_controller_pacemaker_3.sh +++ b/extraconfig/tasks/major_upgrade_controller_pacemaker_3.sh @@ -57,10 +57,10 @@ if [[ -n $(is_bootstrap_node) ]]; then # TODO: check if this can be triggered in puppet and removed from here ceilometer-upgrade --config-file=/etc/ceilometer/ceilometer.conf --skip-gnocchi-resource-types cinder-manage db sync - glance-manage --config-file=/etc/glance/glance-registry.conf db_sync + glance-manage 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 + neutron-db-manage upgrade heads nova-manage db sync nova-manage api_db sync nova-manage db online_data_migrations diff --git a/extraconfig/tasks/major_upgrade_object_storage.sh b/extraconfig/tasks/major_upgrade_object_storage.sh deleted file mode 100644 index 2667bb16..00000000 --- a/extraconfig/tasks/major_upgrade_object_storage.sh +++ /dev/null @@ -1,53 +0,0 @@ -#!/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 - -set -eu - -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; do - systemctl \$action \$S - done -} - -# Special-case OVS for https://bugs.launchpad.net/tripleo/+bug/1635205 -if [[ -n \$(rpm -q --scripts openvswitch | awk '/postuninstall/,/*/' | grep "systemctl.*try-restart") ]]; then - echo "Manual upgrade of openvswitch - restart in postun detected" - mkdir OVS_UPGRADE || true - pushd OVS_UPGRADE - echo "Attempting to downloading latest openvswitch with yumdownloader" - yumdownloader --resolve openvswitch - echo "Updating openvswitch with nopostun option" - rpm -U --replacepkgs --nopostun ./*.rpm - popd -else - echo "Skipping manual upgrade of openvswitch - no restart in postun detected" -fi - -systemctl_swift stop - -yum -y install python-zaqarclient # needed for os-collect-config -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 index b0418a56..8c91027d 100644 --- a/extraconfig/tasks/major_upgrade_pacemaker.yaml +++ b/extraconfig/tasks/major_upgrade_pacemaker.yaml @@ -1,4 +1,4 @@ -heat_template_version: 2016-10-14 +heat_template_version: ocata description: 'Upgrade for Pacemaker deployments' parameters: @@ -33,33 +33,6 @@ resources: # map_merge with input_values instead of feeding params into scripts # via str_replace on bash snippets - CephMonUpgradeConfig: - type: OS::Heat::SoftwareConfig - properties: - group: script - config: - list_join: - - '' - - - str_replace: - template: | - #!/bin/bash - ignore_ceph_upgrade_warnings='IGNORE_CEPH_UPGRADE_WARNINGS' - params: - IGNORE_CEPH_UPGRADE_WARNINGS: {get_param: IgnoreCephUpgradeWarnings} - - get_file: major_upgrade_ceph_mon.sh - - CephMonUpgradeDeployment: - type: OS::Heat::SoftwareDeploymentGroup - properties: - servers: {get_param: [servers, Controller]} - config: {get_resource: CephMonUpgradeConfig} - input_values: {get_param: input_values} - update_policy: - batch_create: - max_batch_size: 1 - rolling_update: - max_batch_size: 1 - ControllerPacemakerUpgradeConfig_Step1: type: OS::Heat::SoftwareConfig properties: @@ -86,26 +59,11 @@ resources: ControllerPacemakerUpgradeDeployment_Step1: type: OS::Heat::SoftwareDeploymentGroup - depends_on: CephMonUpgradeDeployment properties: servers: {get_param: [servers, Controller]} config: {get_resource: ControllerPacemakerUpgradeConfig_Step1} input_values: {get_param: input_values} - BlockStorageUpgradeConfig: - type: OS::Heat::SoftwareConfig - depends_on: ControllerPacemakerUpgradeDeployment_Step1 - properties: - group: script - config: {get_file: major_upgrade_block_storage.sh} - - BlockStorageUpgradeDeployment: - type: OS::Heat::SoftwareDeploymentGroup - properties: - servers: {get_param: [servers, BlockStorage]} - config: {get_resource: BlockStorageUpgradeConfig} - input_values: {get_param: input_values} - ControllerPacemakerUpgradeConfig_Step2: type: OS::Heat::SoftwareConfig properties: @@ -132,7 +90,7 @@ resources: ControllerPacemakerUpgradeDeployment_Step2: type: OS::Heat::SoftwareDeploymentGroup - depends_on: BlockStorageUpgradeDeployment + depends_on: ControllerPacemakerUpgradeDeployment_Step1 properties: servers: {get_param: [servers, Controller]} config: {get_resource: ControllerPacemakerUpgradeConfig_Step2} diff --git a/extraconfig/tasks/major_upgrade_pacemaker_init.j2.yaml b/extraconfig/tasks/major_upgrade_pacemaker_init.j2.yaml deleted file mode 100644 index f6aa3066..00000000 --- a/extraconfig/tasks/major_upgrade_pacemaker_init.j2.yaml +++ /dev/null @@ -1,87 +0,0 @@ -heat_template_version: 2014-10-16 -description: 'Upgrade for Pacemaker deployments' - -parameters: - - servers: - type: json - input_values: - type: json - description: input values for the software deployments - - UpgradeInitCommand: - type: string - description: | - Command or script snippet to run on all overcloud nodes to - initialize the upgrade process. E.g. a repository switch. - default: '' - UpgradeLevelNovaCompute: - type: string - description: Nova Compute upgrade level - default: '' - -resources: - - # For the UpgradeInit also rename /etc/resolv.conf.save for +bug/1567004 - - UpgradeInitConfig: - type: OS::Heat::SoftwareConfig - properties: - group: script - config: - list_join: - - '' - - - "#!/bin/bash\n\n" - - "if [[ -f /etc/resolv.conf.save ]] ; then rm /etc/resolv.conf.save; fi\n\n" - - get_param: UpgradeInitCommand - - # 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 - - # FIXME(shardy) we have hard-coded per-role *ScriptConfig's here - # Would be better to have a common config for all roles - 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 - - ObjectStorageDeliverUpgradeScriptConfig: - type: OS::Heat::SoftwareConfig - properties: - group: script - config: {get_file: major_upgrade_object_storage.sh} - - CephStorageDeliverUpgradeScriptConfig: - type: OS::Heat::SoftwareConfig - properties: - group: script - config: {get_file: major_upgrade_ceph_storage.sh} - -{% for role in roles %} - UpgradeInit{{role.name}}Deployment: - type: OS::Heat::SoftwareDeploymentGroup - properties: - servers: {get_param: [servers, {{role.name}}]} - config: {get_resource: UpgradeInitConfig} - input_values: {get_param: input_values} - - {% if not role.name in ['Controller', 'BlockStorage'] %} - {{role.name}}DeliverUpgradeScriptDeployment: - type: OS::Heat::SoftwareDeploymentGroup - properties: - servers: {get_param: [servers, {{role.name}}]} - config: {get_resource: {{role.name}}DeliverUpgradeScriptConfig} - input_values: {get_param: input_values} - {% endif %} -{% endfor %} diff --git a/extraconfig/tasks/major_upgrade_pacemaker_migrations.sh b/extraconfig/tasks/major_upgrade_pacemaker_migrations.sh index 6d02acc8..ae22a1e7 100644 --- a/extraconfig/tasks/major_upgrade_pacemaker_migrations.sh +++ b/extraconfig/tasks/major_upgrade_pacemaker_migrations.sh @@ -83,7 +83,6 @@ function services_to_migrate { openstack-cinder-api-clone openstack-cinder-scheduler-clone openstack-glance-api-clone - openstack-glance-registry-clone openstack-gnocchi-metricd-clone openstack-gnocchi-statsd-clone openstack-heat-api-cfn-clone diff --git a/extraconfig/tasks/mitaka_to_newton_aodh_data_migration.yaml b/extraconfig/tasks/mitaka_to_newton_aodh_data_migration.yaml index b9a87d33..45933fb7 100644 --- a/extraconfig/tasks/mitaka_to_newton_aodh_data_migration.yaml +++ b/extraconfig/tasks/mitaka_to_newton_aodh_data_migration.yaml @@ -1,4 +1,4 @@ -heat_template_version: 2014-10-16 +heat_template_version: ocata description: > Software-config for performing aodh data migration diff --git a/extraconfig/tasks/pacemaker_common_functions.sh b/extraconfig/tasks/pacemaker_common_functions.sh index 2c7dfc35..aae4a2de 100755 --- a/extraconfig/tasks/pacemaker_common_functions.sh +++ b/extraconfig/tasks/pacemaker_common_functions.sh @@ -297,3 +297,27 @@ function systemctl_swift { manage_systemd_service $action $service done } + +# Special-case OVS for https://bugs.launchpad.net/tripleo/+bug/1635205 +function special_case_ovs_upgrade_if_needed { + if [[ -n $(rpm -q --scripts openvswitch | awk '/postuninstall/,/*/' | grep "systemctl.*try-restart") ]]; then + echo "Manual upgrade of openvswitch - restart in postun detected" + rm -rf OVS_UPGRADE + mkdir OVS_UPGRADE && pushd OVS_UPGRADE + echo "Attempting to downloading latest openvswitch with yumdownloader" + yumdownloader --resolve openvswitch + for pkg in $(ls -1 *.rpm); do + if rpm -U --test $pkg 2>&1 | grep "already installed" ; then + echo "Looks like newer version of $pkg is already installed, skipping" + else + echo "Updating $pkg with nopostun option" + rpm -U --replacepkgs --nopostun $pkg + fi + done + popd + else + echo "Skipping manual upgrade of openvswitch - no restart in postun detected" + fi + +} + diff --git a/extraconfig/tasks/post_puppet_pacemaker.yaml b/extraconfig/tasks/post_puppet_pacemaker.yaml index b62502f8..a63868c9 100644 --- a/extraconfig/tasks/post_puppet_pacemaker.yaml +++ b/extraconfig/tasks/post_puppet_pacemaker.yaml @@ -1,4 +1,4 @@ -heat_template_version: 2014-10-16 +heat_template_version: ocata description: 'Post-Puppet Config for Pacemaker deployments' parameters: diff --git a/extraconfig/tasks/post_puppet_pacemaker_restart.yaml b/extraconfig/tasks/post_puppet_pacemaker_restart.yaml index 52760c87..475a6688 100644 --- a/extraconfig/tasks/post_puppet_pacemaker_restart.yaml +++ b/extraconfig/tasks/post_puppet_pacemaker_restart.yaml @@ -1,4 +1,4 @@ -heat_template_version: 2014-10-16 +heat_template_version: ocata description: 'Post-Puppet restart config for Pacemaker deployments' parameters: diff --git a/extraconfig/tasks/pre_puppet_pacemaker.yaml b/extraconfig/tasks/pre_puppet_pacemaker.yaml index 82546588..aa7514f9 100644 --- a/extraconfig/tasks/pre_puppet_pacemaker.yaml +++ b/extraconfig/tasks/pre_puppet_pacemaker.yaml @@ -1,4 +1,4 @@ -heat_template_version: 2014-10-16 +heat_template_version: ocata description: 'Pre-Puppet Config for Pacemaker deployments' parameters: diff --git a/extraconfig/tasks/run_puppet.sh b/extraconfig/tasks/run_puppet.sh new file mode 100755 index 00000000..b7771e33 --- /dev/null +++ b/extraconfig/tasks/run_puppet.sh @@ -0,0 +1,27 @@ +#!/bin/bash + +function run_puppet { + set -eux + local manifest="$1" + local role="$2" + local step="$3" + local rc=0 + + export FACTER_deploy_config_name="${role}Deployment_Step${step}" + if [ -e "/etc/puppet/hieradata/heat_config_${FACTER_deploy_config_name}.json" ]; then + set +e + puppet apply --detailed-exitcodes "${manifest}" + rc=$? + echo "puppet apply exited with exit code $rc" + else + echo "Step${step} doesn't exist for ${role}" + fi + set -e + + if [ $rc -eq 2 -o $rc -eq 0 ]; then + set +xu + return 0 + fi + set +xu + return $rc +} diff --git a/extraconfig/tasks/swift-ring-deploy.yaml b/extraconfig/tasks/swift-ring-deploy.yaml new file mode 100644 index 00000000..d17f78ae --- /dev/null +++ b/extraconfig/tasks/swift-ring-deploy.yaml @@ -0,0 +1,31 @@ +heat_template_version: ocata + +parameters: + servers: + type: json + SwiftRingGetTempurl: + default: '' + description: A temporary Swift URL to download rings from. + type: string + +resources: + SwiftRingDeployConfig: + type: OS::Heat::SoftwareConfig + properties: + group: script + inputs: + - name: swift_ring_get_tempurl + config: | + #!/bin/sh + pushd / + curl --insecure --silent "${swift_ring_get_tempurl}" | tar xz || true + popd + + SwiftRingDeploy: + type: OS::Heat::SoftwareDeployments + properties: + name: SwiftRingDeploy + config: {get_resource: SwiftRingDeployConfig} + servers: {get_param: servers} + input_values: + swift_ring_get_tempurl: {get_param: SwiftRingGetTempurl} diff --git a/extraconfig/tasks/swift-ring-update.yaml b/extraconfig/tasks/swift-ring-update.yaml new file mode 100644 index 00000000..440c6883 --- /dev/null +++ b/extraconfig/tasks/swift-ring-update.yaml @@ -0,0 +1,42 @@ +heat_template_version: ocata + +parameters: + servers: + type: json + SwiftRingPutTempurl: + default: '' + description: A temporary Swift URL to upload rings to. + type: string + +resources: + SwiftRingUpdateConfig: + type: OS::Heat::SoftwareConfig + properties: + group: script + inputs: + - name: swift_ring_put_tempurl + config: | + #!/bin/sh + TMP_DATA=$(mktemp -d) + function cleanup { + rm -Rf "$TMP_DATA" + } + trap cleanup EXIT + # sanity check in case rings are not consistent within cluster + swift-recon --md5 | grep -q "doesn't match" && exit 1 + pushd ${TMP_DATA} + tar -cvzf swift-rings.tar.gz /etc/swift/*.builder /etc/swift/*.ring.gz /etc/swift/backups/* + resp=`curl --insecure --silent -X PUT "${swift_ring_put_tempurl}" --write-out "%{http_code}" --data-binary @swift-rings.tar.gz` + popd + if [ "$resp" != "201" ]; then + exit 1 + fi + + SwiftRingUpdate: + type: OS::Heat::SoftwareDeployments + properties: + name: SwiftRingUpdate + config: {get_resource: SwiftRingUpdateConfig} + servers: {get_param: servers} + input_values: + swift_ring_put_tempurl: {get_param: SwiftRingPutTempurl} diff --git a/extraconfig/tasks/tripleo_upgrade_node.sh b/extraconfig/tasks/tripleo_upgrade_node.sh new file mode 100644 index 00000000..c2565410 --- /dev/null +++ b/extraconfig/tasks/tripleo_upgrade_node.sh @@ -0,0 +1,66 @@ +#!/bin/bash +# +# This delivers the operator driven upgrade script to be invoked as part of +# the tripleo major upgrade workflow. The utility 'upgrade-non-controller.sh' +# is used from the undercloud to invoke the /root/tripleo_upgrade_node.sh +# +set -eu + +UPGRADE_SCRIPT=/root/tripleo_upgrade_node.sh + +cat > $UPGRADE_SCRIPT << ENDOFCAT +### DO NOT MODIFY THIS FILE +### This file is automatically delivered to those nodes where the +### disable_upgrade_deployment flag is set in roles_data.yaml. + +set -eu +NOVA_COMPUTE="" +if hiera -c /etc/puppet/hiera.yaml service_names | grep nova_compute ; then + NOVA_COMPUTE="true" +fi +SWIFT_STORAGE="" +if hiera -c /etc/puppet/hiera.yaml service_names | grep swift_storage ; then + SWIFT_STORAGE="true" +fi + +DEBUG="true" +SCRIPT_NAME=$(basename $0) +$(declare -f log_debug) +$(declare -f manage_systemd_service) +$(declare -f systemctl_swift) + +# pin nova messaging +-1 for the nova-compute service +if [[ -n \$NOVA_COMPUTE ]]; then + crudini --set /etc/nova/nova.conf upgrade_levels compute auto +fi + +$(declare -f special_case_ovs_upgrade_if_needed) +special_case_ovs_upgrade_if_needed + +yum -y install python-zaqarclient # needed for os-collect-config +if [[ -n \$SWIFT_STORAGE ]]; then + systemctl_swift stop +fi +yum -y update +if [[ -n \$SWIFT_STORAGE ]]; then + systemctl_swift start +fi +# Due to bug#1640177 we need to restart compute agent +if [[ -n \$NOVA_COMPUTE ]]; then + echo "Restarting openstack ceilometer agent compute" + systemctl restart openstack-ceilometer-compute +fi + +# Apply puppet manifest to converge just right after the ${ROLE} upgrade +$(declare -f run_puppet) +for step in 1 2 3 4 5 6; do + if ! run_puppet /root/${ROLE}_puppet_config.pp ${ROLE} \${step}; then + echo "Puppet failure at step \${step}" + exit 1 + fi +done +ENDOFCAT + +# ensure the permissions are OK +chmod 0755 $UPGRADE_SCRIPT + diff --git a/extraconfig/tasks/yum_update.sh b/extraconfig/tasks/yum_update.sh index 8a88ee64..c66dd01f 100755 --- a/extraconfig/tasks/yum_update.sh +++ b/extraconfig/tasks/yum_update.sh @@ -10,6 +10,11 @@ echo "Started yum_update.sh on server $deploy_server_id at `date`" echo -n "false" > $heat_outputs_path.update_managed_packages +if [ -f /.dockerenv ]; then + echo "Not running due to running inside a container" + exit 0 +fi + if [[ -z "$update_identifier" ]]; then echo "Not running due to unset update_identifier" exit 0 @@ -42,7 +47,7 @@ if [[ "$list_updates" == "" ]]; then exit 0 fi -pacemaker_status=$(systemctl is-active pacemaker) +pacemaker_status=$(systemctl is-active pacemaker || :) # Fix the redis/rabbit resource start/stop timeouts. See https://bugs.launchpad.net/tripleo/+bug/1633455 # and https://bugs.launchpad.net/tripleo/+bug/1634851 @@ -63,18 +68,7 @@ if [[ "$pacemaker_status" == "active" && \ fi # Special-case OVS for https://bugs.launchpad.net/tripleo/+bug/1635205 -if [[ -n $(rpm -q --scripts openvswitch | awk '/postuninstall/,/*/' | grep "systemctl.*try-restart") ]]; then - echo "Manual upgrade of openvswitch - restart in postun detected" - mkdir OVS_UPGRADE || true - pushd OVS_UPGRADE - echo "Attempting to downloading latest openvswitch with yumdownloader" - yumdownloader --resolve openvswitch - echo "Updating openvswitch with nopostun option" - rpm -U --replacepkgs --nopostun ./*.rpm - popd -else - echo "Skipping manual upgrade of openvswitch - no restart in postun detected" -fi +special_case_ovs_upgrade_if_needed if [[ "$pacemaker_status" == "active" ]] ; then echo "Pacemaker running, stopping cluster node and doing full package update" diff --git a/extraconfig/tasks/yum_update.yaml b/extraconfig/tasks/yum_update.yaml index d313ca9f..8cff838e 100644 --- a/extraconfig/tasks/yum_update.yaml +++ b/extraconfig/tasks/yum_update.yaml @@ -1,4 +1,4 @@ -heat_template_version: 2014-10-16 +heat_template_version: ocata description: > Software-config for performing package updates using yum @@ -9,7 +9,12 @@ resources: type: OS::Heat::SoftwareConfig properties: group: script - config: {get_file: yum_update.sh} + config: + list_join: + - '' + - - get_file: pacemaker_common_functions.sh + - get_file: yum_update.sh + inputs: - name: update_identifier description: yum will only run for previously unused values of update_identifier diff --git a/extraconfig/tasks/yum_update_noop.yaml b/extraconfig/tasks/yum_update_noop.yaml index b759d9c5..9400c1d2 100644 --- a/extraconfig/tasks/yum_update_noop.yaml +++ b/extraconfig/tasks/yum_update_noop.yaml @@ -1,4 +1,4 @@ -heat_template_version: 2014-10-16 +heat_template_version: ocata description: 'No-op yum update task' resources: |