You can improve your system availability and reliability using high availability options that combine multiple servers, external load balancing, and failover procedures.
Implementing a high availability configuration means setting up your CLIP servers so that service is continuous despite power outages, machine downtime, and heavy load.
Load balancing and high availability is implemented in a distributed deployment. You configure load balancing and high availability by adding an additional CLIP server.
High availability is implemented in two layers:

  • Hardware infrastructure. This layer includes redundant servers, networks, power supplies, and so forth.
  • Application. This layer has two components:
    • Load balancing. Load balancing divides the work load among several computers. As a result, system performance and availability increases. External load balancing is a software and hardware unit supplied by an outside vendor. This unit must be installed and configured to work with CLIP.
    • Automatic Failover. Work performed by one CLIP server is taken over by a redundant CLIP server if a CLIP server or component fails or becomes temporarily unavailable. The failover is automatic, which means that no manual intervention is required.

Implementation of load balancing and failover is discussed in detail throughout this chapter.

High Availability Concept 

In a CLIP HA scenario, all requests to CLIP are distributed to the CLIP instances via a Load Balancer.

Normal Operation

During normal operation, each CLIP instance maintains a local database to store the relationships between events and incidents along with all attributes received during create- and update operations of events and incidents for later reference and lookup via the CLIP Web UI. However, if a single CLIP instance handles a request, it stores all relevant information in all databases of CLIP instances available in the cluster.

Automatic failover

In case one CLIP instance fails ( or is down because of scheduled maintenance ), the Load Balancer does detect the "Failed" status and automatically sends all requests to the remaining instance, which is capable to handle all requests, as all information about event / incident relationships of the failed instance were also logged in the other CLIP instance's database during normal operation.

Automatic recovery

In case a failed CLIP instance becomes available again, the Load Balancer does detect this and immediately starts to also send requests to this instance again.

The restarted instance is available for operations immediately without any delay:

  • should the restarted instance process requests for updates related to event/incident pairs that were handled by the other instance during its downtime, i.e. the relevant data is not contained in the local DB, the restarted instance synchronizes all relevant data related to this event / incident pair ( sharding ).
  • A background process does update the DB of the restarted instance with all data of other instances that were not logged into the local DB during downtime or failure, so when other instances fail, the instance can continue to process all update requests.

Setup for High Availability

To install and configure CLIP for HA and Scalability, we recommend a phased approach:

  1. Phase 1: Install, Configure and Test the first instance of CLIP
    1. start with the primary CLIP server and install the CLIP software
    2. configure the primary CLIP server. At this stage, use the primary's CLIP server's physical hostname / IP Address for the connections configured in the "OMi connected server", in the ServiceNow properties for the backsync URL and/or Remedy filters for backsync.
    3. Test your integrations until you have a final configuration ( ClipConfig.xml ).
  2. Phase 2: Install, Configure and Test a second instance of CLIP
    1. install the CLIP software
    2. configure the second CLIP server so that its configuration matches the configuration of the primary CLIP server by copying "ClipConfig.xml" to the second CLIP server. At this stage, use the secondary's CLIP server's physical hostname / IP Address for the connections configured in the "OMi connected server", in the ServiceNow properties for the backsync URL and/or Remedy filters for backsync.
    3. Verify your integrations work properly
  3. Phase 3: Configure all CLIP servers to work in a CLIP HA cluster
    1. Stop all CLIP servers
    2. configure the load balancer according to ...
    3. configure the virtual host name to access the CLIP cluster in the
      1. OMi "Connected Server" configuration as the "Fully Qualified DNS Name" of the "Target Server" in the "Server Properties" Tab.
      2. ServiceNow "HP BSM Integration Properties" field named "URL for the CLIP WebService for backwards synchronization" as part of this URL
      3. BMC Remedy filters to backsync incident changes to CLIP
    4. on all CLIP servers, edit ClipConfig.xml to configure the list of CLIP servers in the cluster by adding the database tag's attribute "clusterServer" according to section 4.11, "Configuring Clip database". Example:
    5. Start all CLIP servers
    6. Verify your integrations work properly

Load Balancer Configuration

When you install multiple CLIP servers, CLIP can utilize external load balancing mechanisms to help ensure the even distribution of processing and communication activities across the network. This is particularly important in cases of high load, to avoid overwhelming any single server.

  1. Create a virtual host name. The virtual host name is for accessing the CLIP server to submit new or changed events and incidents. The virtual host name must be a fully qualified domain name (FQDN), in the format <servername>.<domainname>.This virtual host name must be used in the
    1. OMi "Connected Server" configuration as the "Fully Qualified DNS Name" of the "Target Server" in the "Server Properties" Tab.
    2. ServiceNow "HP BSM Integration Properties" field named "URL for the CLIP WebService for backwards synchronization" as part of this URL
    3. BMC Remedy filters to backsync incident changes to CLIP
  2. Configure the load balancer for CLIP access. The OMi gateway servers, ServiceNow / ServiceNow mid tier and/or BMC Remedy must be able to access the virtual IP of the load balancer. Use the standard settings for the load balancer, but set the following:
    1. The load balancing method should be set to "Fixed weighting" or "Round robin"
      1. Fixed weighting – this method is used for redundancy rather than load balancing, all connections will go to the CLIP server with the highest weight. In the event this server fails then the server with the next highest weight takes over. The advantage of this option is that all requests are sent to the same CLIP server as long it is available which can make logfile analysis and troubleshooting easier.
      2. Round robin – this method tells the load balancer to direct requests to all real servers in a round robin order. The advantage of this option is that the work is divided among the CLIP servers.
    2. Set the real CLIP servers parameters:
      1. Set the "check parameters" to "TCP connection only"
      2. For each real CLIP server, configure the physical IP address and the port ( 9090 for HTTP / 9443 for HTTPS by default )
      3. If the "fixed weighting" option is used as the load balancing method is used, configure a higher weight for the primary and a lower weight for the secondary or failover CLIP server.
    3. CLIP supports hardware and virtual appliance-based load balancers. A hardware load balancer solution is preferred for performance reasons.