This document explains the root cause and implications of error message: "Communications between clustering nodes might fail because of limited bandwidth, slow/stuck remote cluster nodes or similar causes"
When trying to join a Strategy Intelligence Server node to a cluster in a Strategy 9.x or later, the following error message is displayed:
Communications between clustering nodes might fail because of limited bandwidth, slow/stuck remote cluster nodes or similar causes
The process to join a node to a cluster includes the following tasks:
1) Send the join request from the origin node to the destination node
2) The destination node acknowledges it has received the join request.
3) The destination node proceeds to join the cluster
4) The destination node sends a broadcast message informing that it has successfully joined the cluster.
In case the destination node is unable to send the broadcast notification in step 4 within 20 seconds of the cluster join completion, this error message is automatically triggered.
Reasons for this broadcast taking longer than 20 seconds might be:
- Network issues,
- High Intelligence Server workload
- Additional system failures.
As this error message is triggered by a delay of step 4 but only after the completion of step 3, the cluster join should have already been completed by that time. Therefore, a simple refresh of the cluster monitor should show both origin node and destination node as members of the cluster.
Once an additional node has been added to the cluster, the web sessions and subscription jobs should be load balanced as per the configured settings. Nevertheless, it is strongly suggested to continue monitoring the available network and processing resources to ensure there are no issues external to Strategy that could be affecting the communication between the nodes (network issues or additional system failures).'
