diff options
| -rwxr-xr-x | environments/neutron-sriov.yaml | 3 | ||||
| -rw-r--r-- | extraconfig/tasks/major_upgrade_block_storage.sh | 11 | ||||
| -rwxr-xr-x | extraconfig/tasks/major_upgrade_ceph_mon.sh | 82 | ||||
| -rw-r--r-- | extraconfig/tasks/major_upgrade_ceph_storage.sh | 106 | ||||
| -rw-r--r-- | extraconfig/tasks/major_upgrade_pacemaker.yaml | 48 | ||||
| -rw-r--r-- | puppet/services/aodh-api.yaml | 3 | ||||
| -rw-r--r-- | puppet/services/neutron-plugin-ml2.yaml | 7 | ||||
| -rw-r--r-- | puppet/services/pacemaker.yaml | 15 |
8 files changed, 16 insertions, 259 deletions
diff --git a/environments/neutron-sriov.yaml b/environments/neutron-sriov.yaml index 9b7e51f9..5e9e15e3 100755 --- a/environments/neutron-sriov.yaml +++ b/environments/neutron-sriov.yaml @@ -9,9 +9,6 @@ parameter_defaults: #NovaSchedulerDefaultFilters: ['RetryFilter','AvailabilityZoneFilter','RamFilter','ComputeFilter','ComputeCapabilitiesFilter','ImagePropertiesFilter','ServerGroupAntiAffinityFilter','ServerGroupAffinityFilter','PciPassthroughFilter'] #NovaSchedulerAvailableFilters: ["nova.scheduler.filters.all_filters","nova.scheduler.filters.pci_passthrough_filter.PciPassthroughFilter"] - # Provide the vendorid:productid of the VFs - #NeutronSupportedPCIVendorDevs: ['8086:154c','8086:10ca','8086:1520'] - #NeutronPhysicalDevMappings: "datacentre:ens20f2" # Number of VFs that needs to be configured for a physical interface diff --git a/extraconfig/tasks/major_upgrade_block_storage.sh b/extraconfig/tasks/major_upgrade_block_storage.sh deleted file mode 100644 index 64c4457e..00000000 --- a/extraconfig/tasks/major_upgrade_block_storage.sh +++ /dev/null @@ -1,11 +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 -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_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 a745e723..00000000 --- a/extraconfig/tasks/major_upgrade_ceph_storage.sh +++ /dev/null @@ -1,106 +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 - -declare -f special_case_ovs_upgrade_if_needed > $UPGRADE_SCRIPT -# use >> here so we don't lose the declaration we added above -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 ${OSD_PIDS} 2> /dev/null; do - sleep 2; -done" - -special_case_ovs_upgrade_if_needed - -# Update (Ceph to Jewel) -yum -y install python-zaqarclient # needed for os-collect-config -yum -y update - -# Restart/Exit if not on Jewel, only in that case we need the changes -UPDATED_VERSION=$(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 $OSD_IDS; do - OSD_FS=$(df -l --output=fstype /var/lib/ceph/osd/ceph-${OSD_ID} | tail -n +2) - 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_pacemaker.yaml b/extraconfig/tasks/major_upgrade_pacemaker.yaml index b63aafbd..8c91027d 100644 --- a/extraconfig/tasks/major_upgrade_pacemaker.yaml +++ b/extraconfig/tasks/major_upgrade_pacemaker.yaml @@ -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: @@ -86,30 +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: - list_join: - - '' - - - get_file: pacemaker_common_functions.sh - - 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: @@ -136,7 +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/puppet/services/aodh-api.yaml b/puppet/services/aodh-api.yaml index 62c4b093..4bd9fc47 100644 --- a/puppet/services/aodh-api.yaml +++ b/puppet/services/aodh-api.yaml @@ -89,6 +89,3 @@ outputs: - name: Stop aodh_api service (running under httpd) tags: step2 service: name=httpd state=stopped - - name: Run aodh dbsync - tags: step5 - command: aodh-dbsync diff --git a/puppet/services/neutron-plugin-ml2.yaml b/puppet/services/neutron-plugin-ml2.yaml index 71a0076f..3abd04f3 100644 --- a/puppet/services/neutron-plugin-ml2.yaml +++ b/puppet/services/neutron-plugin-ml2.yaml @@ -60,12 +60,6 @@ parameters: default: 'vxlan' description: The tenant network type for Neutron. type: comma_delimited_list - NeutronSupportedPCIVendorDevs: - description: | - List of supported pci vendor devices in the format VendorID:ProductID. - By default Intel & Mellanox SR-IOV capable NICs are supported. - type: comma_delimited_list - default: ['15b3:1004','8086:10ca'] resources: NeutronBase: @@ -91,7 +85,6 @@ outputs: neutron::plugins::ml2::tunnel_id_ranges: {get_param: NeutronTunnelIdRanges} neutron::plugins::ml2::vni_ranges: {get_param: NeutronVniRanges} neutron::plugins::ml2::tenant_network_types: {get_param: NeutronNetworkType} - neutron::plugins::ml2::supported_pci_vendor_devs: {get_param: NeutronSupportedPCIVendorDevs} step_config: | include ::tripleo::profile::base::neutron::plugins::ml2 diff --git a/puppet/services/pacemaker.yaml b/puppet/services/pacemaker.yaml index c47229f1..2b1071a3 100644 --- a/puppet/services/pacemaker.yaml +++ b/puppet/services/pacemaker.yaml @@ -87,6 +87,10 @@ parameters: \[(?<pid>[^ ]*)\] (?<host>[^ ]*) (?<message>.*)$/ + PacemakerResources: + type: comma_delimited_list + description: List of resources managed by pacemaker + default: ['rabbitmq','haproxy'] outputs: role_data: @@ -127,3 +131,14 @@ outputs: tripleo::profile::base::pacemaker::remote_authkey: {get_param: PacemakerRemoteAuthkey} step_config: | include ::tripleo::profile::base::pacemaker + upgrade_tasks: + - name: Stop pacemaker cluster + tags: step1 + pacemaker_cluster: state=offline + - name: Start pacemaker cluster + tags: step4 + pacemaker_cluster: state=online + - name: Check pacemaker resource + tags: step4 + pacemaker_resource: state=started resource={{item}} check_mode=true wait_for_resource=true timeout=200 + with_items: {get_param: PacemakerResources} |