IBM Control Center Monitor technical overview
IBM® Control Center Monitor consists of user interfaces, application servers, event processors (EPs), and databases. IBM Control Center Monitor supports dynamically discovered servers and manually defined servers, and collects events from these types of servers in different ways.
IBM Control Center Monitor user interfaces include the console and web console.
The web console is a web-based user interface that is hosted by the Jetty Web Application Servers. The web console provides functions that are used routinely, such as monitoring activity and status, and running reports. All web consoles, regardless of the application servers they are connected to in a cluster, display the same information.
The console is a user interface that can be installed locally and opened as a web application on the computer. The console provides similar user functions as the web console for monitoring server activity and status. In addition, the console allows administrators to configure IBM Control Center Monitor and the configuration objects that are necessary for monitoring servers, and define reports than can be run from the console or web console. Like web consoles, when multiple EPs are implemented, all consoles connected to EPs in a cluster display the same information.
IBM Control Center Monitor supports onboarding servers through dynamic discovery and manual definition.
Servers that post their events to Control Center using the Open Server Architecture (OSA) interface such as IBM Global High Availability Mailbox, are onboarded when the servers begin posting events to the event repository (ER) through RESTful Application Programming Interfaces (APIs). Jetty Web Application Servers host the ER. You do not need to configure IBM Control Center Monitor before the servers post events to the ER. The ER servlet dynamically discovers the new servers, assigns them to EPs, and writes their events to the unprocessed event table in the IBM Control Center Monitor database. EPs then collect events of assigned servers from the unprocessed event table and process those events for rules and alerts.
Manually defined servers, also known as "legacy" servers, such as IBM Sterling Connect:Direct® and IBM Sterling B2B Integrator, are onboarded after the servers are configured in IBM Control Center Monitor through the console. EPs need to collect events from manually defined servers through polling, listening on a queue, or receiving Simple Network Management Protocol (SNMP) traps. EPs process events from manually defined servers immediately after the events are received.
After the ER servlet writes events from the dynamically discovered servers to the unprocessed event table, EPs poll the unprocessed event table for events from the dynamically discovered servers they are monitoring. Events from the servers are then passed through a series of services on the IBM Control Center Monitor EP for processing, including the visibility service, metadata service, rule service, and service level criteria (SLCs) service. The events trigger rules and SLCs, and associated actions. After EPs process the events, EPs write these events to the events table and remove them from the unprocessed event table.
EPs communicate with manually defined servers through the node services. EPs handle events from these servers immediately after the events are received. Similar to dynamically discovered servers, events from manually defined servers are also passed through visibility service, metadata service, rule service, and SLC service. Events also trigger rules and SLCs, and are written to the events table in the database after they are processed.
Databases and the reporting engine
IBM Control Center Monitor uses a production database to store events and statistics that are gathered from the monitored servers. The unprocessed events table in the production database is used to record events that are received from dynamically discovered servers. The events table is used to record the processed events from both dynamically discovered servers and manually defined servers for historical purposes.
A reports database is used to generate IBM Control Center Monitor reports. Jasper Reports Library server is used as the reporting engine with the production and reports databases to access and report information about monitored servers.
The high availability architecture can use one or more EPs and Jetty Web Application Servers. When you install multiple EPs and Jetty Web Application Servers, you can implement IBM Control Center Monitor in a high availability environment. When an EP in the cluster is down, the servers that are assigned to the EP are temporarily distributed to other EPs for event processing according to the defined EP failover policy. Jetty Web Application Servers can host multiple web consoles and ERs. When a web console or an ER is down, the workload is failed over to other web consoles or ERs if a load balancer is configured.
In case of clustering (high availability setup – multiple EPs connect to the same database), reports functionality is available in all EPs. When a report is run from IBM Sterling Control Center Classic Console, the report output is generated by the EP to which Swing UI is connected to.
When a report is run from IBM Control Center Monitor Monitor Web Console, the report output is generated by the CEP. In case of Scheduled Reports, the report output is generated by the CEP.
IBM Control Center Monitor supports native database high availability solutions such as the database high availability solution provided by Oracle Real Application Clusters 12c.