Support Center > Search Results > SecureKnowledge Details
Changing Cluster topology and performing policy installation under high load, results in a fail-over
Symptoms
  • Pushing policy in order to update the cluster topology during high load, causing the members to fail-over.
Cause

The current design is that while pushing policy, the cluster enters a "freeze" state, which is causing the Pnotes to not change their state, in order to prevent a fail-over.

In case of high load, once the freeze mechanism time is over, the Pnotes will try to change the Cluster state as well. In case there is high load on the Security Gateway, and the Cluster topology was not yet updated to all the Cluster members,  the member which was updated, detect CCP loss and trigger a fail-over.


Solution
Note: To view this solution you need to Sign In .