Node management

Node states

Nodes in Jackadi can exist in one of three states:

State	Description	Capabilities
Candidate	The initial state when an node first connects to the manager, not yet approved.	Cannot execute tasks, awaits approval.
Accepted	Nodes that have been approved and can execute tasks.	Can receive and execute tasks, full system access.
Rejected	Nodes that have been denied access to the system, e.g. duplicate nodes or manually rejected.	Cannot connect or execute tasks.

List nodes

command

jack nodes list

output

Accepted

 • node1
 • node2

 Candidates


 Rejected

Detailed view

command

jack nodes list --verbose

output

Accepted

 • node1 (172.18.0.3 MIICIjANBgkqhkiG9w...EJEX0CAwEAAQ==)
 • node2 (172.18.0.2 MIICIjANBgkqhkiG9w...HWGG0CAwEAAQ==)

 Candidates


 Rejected

JSON output

command

jack nodes list --json

output

{
   "Accepted": [
      {
         "Id": "node1",
         "Address": "172.18.0.3",
         "Certificate": "MIICIjANBgkqhkiG9w...EJEX0CAwEAAQ==",
         "IsConnected": true,
         "Since": {
            "Seconds": 1757578390,
            "Nanos": 564026075
         },
         "LastMsg": {
            "Seconds": 1757610077,
            "Nanos": 274820288
         },
         "IsActive": true
      },
      {
         "Id": "node2",
         "Address": "172.18.0.2",
         "Certificate": "MIICIjANBgkqhkiG9w...HWGG0CAwEAAQ==",
         "IsConnected": true,
         "Since": {
            "Seconds": 1757578390,
            "Nanos": 569209812
         },
         "LastMsg": {
            "Seconds": 1757610077,
            "Nanos": 274888616
         },
         "IsActive": true
      }
   ],
   "Candidates": null,
   "Rejected": null
}

Check node health

command

jack nodes health

output

Nodes

 • node1 (connected, active)
 • node2 (connected, active)

node1 is connected and had recent activity (e.g. a task has been executed).
node2 is connected and did not have any recent activity.
node3 is not connected to the manager anymore (e.g. due to service down, or network issue).

Detailed health view

command

jack nodes health --verbose

output

Nodes

 • node1
   • state: connected, active
   • connected since: September 11, 2025 at 08:13 UTC
   • last active: September 11, 2025 at 17:02 UTC
 • node2
   • state: connected, active
   • connected since: September 11, 2025 at 08:13 UTC
   • last active: September 11, 2025 at 17:02 UTC

Accept nodes

command

jack nodes accept <node_id>

example

# Accept a candidate node
$ jack nodes accept node2

output

node registered: id:"node2"  address:"172.18.0.2"  certificate:"MIICIjANBgkqhkiG9w...HWGG0CAwEAAQ=="

Accept with specific configuration

command

jack nodes accept <node_id> --address <address> --certificate <cert>

example

# Accept node with specific address and certificate
$ jack nodes accept node2 --address "192.168.1.101:8080" --certificate "node2.pem"

output

node registered: id:"node2"  address:"192.168.1.101:8080"  certificate:"MIICIjANBgkqhkiG9w...HWGG0CAwEAAQ=="

Reject nodes

command

jack nodes reject <node_id>

example

# Reject a candidate node
$ jack nodes reject node5

output

node rejected: node2

Remove nodes

command

jack nodes remove <node_id>

example

# Remove an accepted node
$ jack nodes remove node1

output

node removed: node2

Force accept rejected nodes

command

jack nodes accept <node_id> --force

example

# Force accept a previously rejected node
$ jack nodes accept node4 --force

output

node registered: id:"node2"  address:"172.18.0.2"  certificate:"MIICIjANBgkqhkiG9w...HWGG0CAwEAAQ=="

Auto-acceptance

For development or trusted environments, you can enable auto-acceptance on the manager:

command

manager --auto-accept-node

This automatically accepts new nodes when they connect without requiring manual approval.

Security considerations

Production environments: Disable auto-accept and manually review all node connection requests.
Rogue node protection: Jackadi prevents multiple nodes with the same ID - duplicates are automatically rejected.
TLS certificates: Use certificates for secure node authentication in production.
Regular auditing: Monitor node lists regularly to ensure only authorized nodes are connected.

Node workflow example

example

# 1. Start an node (from node machine)
$ node --id="web-server-01" --manager-address="manager.example.com:8080"

# 2. Check for new candidates (from manager machine)
$ jack nodes list

output

Accepted

 • node1

 Candidates

 • web-server-01

 Rejected

example

# 3. Accept the new node
$ jack nodes accept web-server-01

output

node registered: id:"web-server-01"  address:"172.18.0.4"  certificate:"MIICIjANBgkqhkiG9w...HWGG0CAwEAAQ=="

example

# 4. Verify the node is now accepted and connected
$ jack nodes health

output

Nodes

 • node1 (connected, active)
 • web-server-01 (connected, inactive)

Manual installation