Compute node (HCI) Replacement Guide
Remove Compute node from cluster
Stop the compute host
Goto Admin > Compute > All Hypervisors and click Compute host tab. Look for the hostname you wish to remove and click on button DISABLE SERVICE
Disable service
Leave a reason and click DISABLE SERVICE
Service Disabled
Make sure the service is disabled
Migrate Instances from disabled host
Goto Admin > Compute > Instances, look for the instances which is hosted on DISABLED Host and start the LIVE MIGRATE INSTANCE from the action menu
Allocate the instance to new host
You can choose a preferred host to migrate or Automatically and SUBMIT
All instances has been migrated
Repeat the steps and migrate the rest of the Instances
Connect to control node
$ ssh [email protected]
Warning: Permanently added '192.168.1.x' (ECDSA) to the list of known hosts.
Password:
Remove the compute node
control01> cluster
control01:cluster> remove_node
1: compute01
2: control03
3: compute02
4: control01
5: control02
Enter index: 3
this command is only applicable for compute or storage nodes
make sure its running instances have been properly terminated or migrated
shutdown the target host before proceeding
Enter 'YES' to confirm: YES
Results
Goto Admin > Compute > All Hypervisors and click Compute host tab. Host compute02 has been removed from the Cluster
Backup Compute02 policies
From your local pc terminal
$ scp -r root@Compute02_IPADDRESS:/etc/policies Downloads/compute02_policy
Shutdown the compute02
Adding Compute Host
Prepare a new node with CubeOS installed
Configuration
Reconfigure a new compute02 node by following any options of the list below:
Connect to controller
$ ssh [email protected]
Warning: Permanently added '192.168.1.x' (ECDSA) to the list of known hosts.
Password:
Check & Repair services
Welcome to the Cube Appliance
Enter "help" for a list of available commands
control01> cluster
control01:cluster> check_repair
Service Status Report
ClusterLink ok [ link(v) ]
ClusterSettings ok [ etcd(v) ]
HaCluster ok [ hacluster(v) ]
MsgQueue ok [ rabbitmq(v) ]
IaasDb ok [ mysql(v) ]
VirtualIp ok [ vip(v) haproxy_ha(v) ]
Storage FIXING [ ceph(v) ceph_mon(v) ceph_mgr(v) ceph_mds(v) ceph_osd(1) ceph_rgw(v) ]
ok [ ceph_osd(v) ]
ApiService ok [ haproxy(v) apache2(v) lmi(v) memcache(v) ]
Compute ok [ nova(v) ]
Network ok [ neutron(v) ]
Image ok [ glance(v) ]
BlockStor ok [ cinder(v) ]
FileStor ok [ manila(v) ]
ObjectStor ok [ swift(v) ]
Orchestration ok [ heat(v) ]
LBaaS ok [ octavia(v) ]
DNSaaS ok [ designate(v) ]
InstanceHa ok [ masakari(v) ]
DisasterRecovery ok [ freezer(v) es235(v) ]
DataPipe ok [ zookeeper(v) kafka(v) ]
Metrics ok [ ceilometer(v) monasca(v) telegraf(v) grafana(v) ]
LogAnalytics ok [ filebeat(v) auditbeat(v) logstash(v) es(v) kibana(v) ]
Notifications ok [ influxdb(v) kapacitor(v) ]
control01:cluster>
Results
Goto Admin > Compute > All Hypervisors and click Compute host tab. Host compute02 has been added back to the Cluster
