High Availability / Configuring Clusters |
You can configure your system to send email alerts to notify you when significant events affect your cluster.
Vocera Platform sends audit events to the audit logs. When audit events are configured to send emails via SMTP, you will receive email notifications. You can configure cluster specific event codes to get notified on cluster related events, see Working with Notifications Settings for more information. For a list of cluster related event codes that you can configure, see Clustering Event Codes.
You can use the SMTP settings in the Destinations tab to configure email address information for alerts. For more information, see Configuring SMTP Settings.
The Vocera Platform voice server provides the following cluster-related email messages:
"Warning: Failover occurred on Vocera cluster. New active server has host name <IP address>."
The new active cluster node sends this message to notify you that a failover has occurred.
"Standby cluster member <IP address> is no longer active. Reported by active server <IP address>."
The active cluster node sends this message to notify you that it has lost contact with a standby node.
"Warning: Your Vocera Platform cluster had multiple active nodes. The server that was active the longest [<IP address>] is still active. The other one [<IP address>] has automatically reverted to standby mode."
The active cluster node sends this message to notify you that Vocera has automatically healed a split brain state. If a split brain occurs, you will receive other email messages before this one, as described in Troubleshooting Network Problems and Clusters.
In environments with an unstable network, these email messages may be symptoms of underlying problems you need to address. See Network Problems and Clustering for additional information on interpreting these messages.
You can choose the event codes related to clustering to get notified via SMTP (in addition to Syslog and SNMP notifications).
The following is a list of Event Codes related to clustering:
668=Failed to sync files to standby node {ipAddress} 673=A backup node has been added to the cluster with different versions of software than are installed on the current active node 695=Node state has changed from {prevNodeState} to {newNodeState} 696=Node has entered a FAILED state 699=Failed to process node message 700=Failed to process node action message 702=Failed to process node status message from {sourceNode} 703=Failed to receive node status message from node {sourceNode} for {duration} seconds 704=Failed to send node status message 710=Failed database replication for database {database} 744=Node {ipAddress} has been stuck in {nodeState} for {minutes} minutes 788=Voice cluster failover occurred. The new active server ip address is '{activeIpAddress}' 789=Voice cluster active node at ip address '{activeIpAddress}' restarting itself. Maximum service time for standby machine request was exceeded 790=Voice cluster standby node at ip address '{standbyIpAddress}' is no longer active. Reported by active server at ip address '{activeIpAddress}' 800=Voice cluster had multiple active nodes. The server that was active the longest '{activeIpAddress}' (or is non-empty) is still active. The other '{standbyIpAddress}' has automatically reverted to standby mode