Cluster
Subsections of Cluster
cluster status
Options
| option | description | Allowed values |
|---|---|---|
| output-format | format the output to one of table, json or yaml | string |
Examples
- Show cluster status
proxcli cluster statusCluster log
Options
| option | description | Allowed values |
|---|---|---|
| output-format | format the output to one of table, json or yaml | string |
| max-items | max number of log lines we grab from nodes | integer |
| proxmox-nodes | comma separated list of nodes issuing the log lines | string |
| severities | filter logs by severities. this option is a coma separated list of severities labels (panic,alert,critical,error,warning,notice,info,debug) | string |
Examples
- Show cluster log
proxcli cluster log- Show errors searching in the last 1000 logs lines
proxcli cluster log --severities critical,error,warning --max-items 1000High availibility
Definition
High Availability ensures that a VM will stay running even if an individual node is shut down. This means that if the device is either powered off or has any sort of issue, the VM will automatically migrate to another node and start there. This will use all nodes to ensure the VMs configured will stay running as close to 100% of the time as possible.
source: https://www.wundertech.net/how-to-set-up-a-cluster-in-proxmox/
Virtual machine HA in proxmox depend on two concepts:
- groups
- resources
An HA group is a logical configuration specifying on which nodes the vm can be migrated when a proxmox node fail. HA resouces is a virtual machine associated with an HA group. Several resources can be created and associated with a group.
Subsections of High availibility
Cluster ha groups
Subsections of Cluster ha groups
List cluster HA group
Options
| option | description | Allowed values |
|---|
Examples
- list cluster ha groups
proxcli cluster ha groups listCreate an HA cluster group
Options
| option | description | Allowed values |
|---|---|---|
| group | Cluster ha group name to be created | string |
| proxmox-nodes | On which proxmox nodes the group should apply | string |
| restricted | The CRM tries to run services on the node with the highest priority. If a node with higher priority comes online, the CRM migrates the service to that node. Enabling nofailback prevents that behavior. | N/A |
| nofailback | Resources bound to restricted groups may only run on nodes defined by the group. The resource will be placed in the stopped state if no group node member is online. Resources on unrestricted groups may run on any cluster node if all group members are offline, but they will migrate back as soon as a group member comes online. One can implement a preferred node behavior using an unrestricted group with only one member. | N/A |
Examples
- create an HA group for application powerdns. the group will use proxmox nodes pve1 and pve2 and make sure the group resources will stay on those 2 nodes
proxcli cluster ha groups create --group powerdns --nofailback --proxmox-nodes "pve1,pve2"Delete an HA cluster group
Options
| option | description | Allowed values |
|---|---|---|
| group | Cluster ha group name to be deleted | string |
Note
It is impossible to delete a group with associated resources. You must first delete resources from group and then delete the group
Examples
- delete a cluster HA group
proxcli cluster ha groups delete --group powerdnsCluster HA resources
Subsections of Cluster HA resources
List cluster ha resource
Options
| option | description | Allowed values |
|---|
Examples
- List cluster resources
proxcli cluster ha resources listAdd cluster ha resource
Options
| option | description | Allowed values |
|---|---|---|
| group | the cluster ha group this resource belong to | string |
| vmid | the virtual machine id you want to assign to the resource | string |
| name | the cluster ha resource name | string |
| comment | additional information for this resource | N/A |
| state | Requested resource state. The CRM reads this state and acts accordingly. Please note that enabled is just an alias for started. (disabled | enabled |
| max-relocate | Maximal number of service relocate tries when a service failes to start | integrer |
| max-restart | Maximal number of tries to restart the service on a node after its start failed. | integer |
Examples
- create a cluster ha resource with minimal informations
proxcli cluster ha resources add --group gitlab --vmid 105Delete cluster ha resource
Options
| option | description | Allowed values |
|---|---|---|
| vmid | the virtual machine id you want to remove from resources | string |
| filter-name | A regex applied on virtual machine name used to select matching virtual machines to remove from resources | string |
Examples
- Delete a single cluster ha resource by virtual machine id
proxcli cluster ha resources delete --vmid 105- Delete all cluster ha resources (not recomended)
proxcli cluster ha resources delete --filter-name "^.*$"Migrate cluster ha resource
Options
| option | description | Allowed values |
|---|---|---|
| vmid | The virtual machine id to migrate | integer |
| filter-name | a regex on virtual machines name used to select multiples virtual machines to be migrated | string |
| proxmox-node | the target node to migrate the virtual machines to | string |
| block | wait for each resources to finish migration before starting another one (sequential mode) | string |
Note
vmid and filter-name are mutualy exclusive
Examples
- migrate a cluster ha resource to another node
proxcli cluster ha resources migrate --vmid 125 --proxmox-node pve1- migrate multiple cluster ha resources to another node
proxcli cluster ha resources migrate --filter-name "^b4p-powerdns" --proxmox-node pve1- migrate multiple cluster ha resources to another node waiting for each resource to finish migration
proxcli cluster ha resources migrate --filter-name "^b4p-powerdns" --proxmox-node pve1 --blockRelocate cluster ha resource
Options
| option | description | Allowed values |
|---|---|---|
| vmid | The virtual machine id to be relocated | integer |
| filter-name | a regex on virtual machines name used to select multiples virtual machines to be relocated | string |
| proxmox-node | the target node to relocate the virtual machines to | string |
| block | wait for each resources to finish relocation before starting another one (sequential mode) | string |
Note
vmid and filter-name are mutualy exclusive
Examples
- relocate a cluster ha resource to another node
proxcli cluster ha resources relocate --vmid 125 --proxmox-node pve1- relocate multiple cluster ha resources to another node
proxcli cluster ha resources relocate --filter-name "^b4p-powerdns" --proxmox-node pve1- relocate multiple cluster ha resources to another node waiting for each resource to finish relocation
proxcli cluster ha resources relocate --filter-name "^b4p-powerdns" --proxmox-node pve1 --block