If a SystemLink client with the store and forward feature enabled has not been connected to the server in a long time, a large backlog may accrue in its local store. Once the client is reconnected to the server, it will begin to forward the stored transactions to the server at a high rate. For a single client reconnecting, this is typically not an issue. However, if several clients are forwarding their backlog at the same time, it may cause performance issues on the server. This would likely only be a problem when there are many megabytes of data on 6 or more clients.

This article provides a guide on how to re-establish the connection of many clients with large backlogs. The inflow of data to the server may cause it to become unresponsive. Care should be taken with how you re-establish the connection of many clients.

1. Disable the forwarding service on all clients (they will continue to store data during this time). This can be done manually by disabling the “NI SystemLink Forwarding Service” in Windows Services. Or it can be done remotely using the following salt state. 

disable_forwarding:
   cmd.run:
      - name: "Stop-Service nisystemlinkforwarding -Force"
      - shell: powershell

2. Install the Store and Forward Salt Beacon on the clients if not already installed.

3. Identify 3-5 clients to resume the forwarding service. You might select clients based on priorities or try to get systems with smaller backlogs online first.

4. Enable forwarding service on those clients, either manually in Windows Services or with the following salt state:

enable_forwarding:
   cmd.run:
      - name: "Start-Service nisystemlinkforwarding"
      - shell: powershell

5. Monitor the performance and responsiveness of the server while data is forwarded. Also monitor the store and forward beacon’s tags to track the forwarding progress.

6. Continue enabling more clients in batches until all clients are enabled. If performance is stable while forwarding 3-5 clients, consider enabling more at a time.