Node List
In the Node List option you can check the list of nodes known to the HSM in question; it is with this list of nodes that this HSM will try to replicate all transactions where it acts as coordinator. A node can be placed on the list automatically, via auto-discovery with the SLP protocol, or manually, with the operator adding and removing nodes.
Info
All nodes in the pool must run the same firmware version. Running different versions is not recommended without prior consultation with the HSM manufacturer or distributor.
Dinamo - Local Management Console
┌─────────────────────┤ Node List ├──────────────────────┐
│ │
│ ┌────────────────────────────────────────────────────┐ │
│ │ M 172.17.0.2 ↑ │ │
│ │ ░ │ │
│ │ ▒ │ │
│ │ ▒ │ │
│ │ ▒ │ │
│ │ ▒ │ │
│ │ ▒ │ │
│ │ ↓ │ │
│ └────────────────────────────────────────────────────┘ │
│ │
│ A - Auto Discovered M - Manual Entry │
│ │
│ ┌──────────┐ ┌──────┐ ┌───────┐ ┌───────┐ ┌───────┐ │
│ │ Discover │ │ Test │ │ Add │ │ Del │ │ Close │ │
│ └──────────┘ └──────┘ └───────┘ └───────┘ └───────┘ │
│ │
└────────────────────────────────────────────────────────┘
Service running... Replication Domain: <list>On the Discover it is possible to trigger the node auto-discovery mechanism on demand via IP multicast with SLP protocol. The discovery process will overwrite all automatic entries in the list (marked with A), but does not touch the manual inputs (marked with M). If automatic discovery finds a node with an IP that already appears in the list marked as manual, this node is promoted to automatic, avoiding duplicate entries in the replication list. Before proceeding with the auto-discovery process, the operator is warned and must confirm the operation.
It is possible to run the auto-discovery function at any time, but it is recommended that it be done under controlled conditions, preferably outside the environment's production window and with all nodes in the Domain active and operational, so that replication services are not interrupted and the replication list is complete on all nodes. This will avoid impacts on the business application.
Dinamo - Local Management Console
┌─────────────────────┤ Node List ├──────────────────────┐
│ │
│ ┌────────────────────────────────────────────────────┐ │
│ │ M 172.17.0.2 ↑ │ │
│ │ ░ │ │
│ │ ▒ │ │
│ │ ┌───────────────────┤ ├───────────────────┐ ▒ │ │
│ │ │ │ ▒ │ │
│ │ │ Retrieve list from specific remote node? │ ▒ │ │
│ │ │ │ ▒ │ │
│ │ │ ┌────┐ ┌─────┐ │ ↓ │ │
│ └────│ │ No │ │ Yes │ │ ───┘ │
│ │ └────┘ └─────┘ │ │
│ │ │ │
│ │ │ │
│ ┌──└──────────────────────────────────────────┘ ──┐ │
│ │ Di e │ │
│ └──────────┘ └──────┘ └───────┘ └───────┘ └───────┘ │
│ │
└────────────────────────────────────────────────────────┘
Service running... Replication Domain: <list>The Test button performs connectivity tests and checks the conditions for operating replication between the two nodes.
The functions of the Test and Del buttons act on the node selected in the list.
The node auto-discovery function is only performed automatically during the configuration of the Replication Domain. After that, the list is only changed, with manual addition and removal of nodes and automatic re-discovery, on explicit demand from the operator. Cases of network partitioning preventing communication between HSMs, nodes that are permanently removed from the pool by default or for maintenance and new nodes added that were not recognized by the old nodes in the sensitization signal are examples of situations where the operator must act to recompose the replication list. These are all exception situations. In the CAP theory model, the HSM replication mechanism is designed to meet the Consistency and Partition Tolerance requirements.