Google Search Appliance Configuring GSA Mirroring Google Search Appliance software version 7.
Google, Inc. 1600 Amphitheatre Parkway Mountain View, CA 94043 www.google.com September 2012 © Copyright 2012 Google, Inc. All rights reserved. Google and the Google logo are registered trademarks or service marks of Google, Inc. All other trademarks are the property of their respective owners. Use of any Google solution is governed by the license agreement included in your original contract.
Contents Configuring GSA Mirroring ...........................................................................................
Configuring GSA Mirroring This guide contains the information you need to use GSA mirroring, a feature of the Google Search Appliance. GSA mirroring is a feature that enables the index and other configuration information from one search appliance, called the master node, to be replicated to another search appliance, called a replica. You can set up either an active-active or active-passive mirroring configuration.
Requirements for GSA Mirroring GSA mirroring requires a sustainable 1 MB per second file transfer rate between the master Google Search Appliance and each replica search appliance. To determine whether a network can provide the required file transfer rate, we recommend that you measure the rate by transferring files on your network between the subnets where the search appliances are located. If the file transfer requirement is not met by the network, the mirroring feature might not work as expected.
In scenario two, both search appliances are freshly installed and the content is being crawled for the first time. The master search appliance produces checkpoint files with newly-crawled content. The checkpoint files will be of a moderate size and will take somewhat more time to transfer than in scenario one.
• To the master search appliance only, in an active-passive configuration. In this configuration, you can manually fail over search traffic to a replica node if the master search appliance fails. • To all nodes, in an active-active configuration. In this configuration, all nodes can serve results. You use a load balancer to distribute search requests among the nodes. Ensure that you use a load balancer that supports sticky user sessions, so that the user receives consistent search results.
• Connectors (Connector Manager definitions and configurations are copied) • Existing feeds • Certificates About Security The Google Search Appliance uses secret tokens and private IP addresses to enforce security within GSA mirroring configurations. The search appliances in a GSA mirroring configuration authenticate each other using shared secret tokens that you provide during configuration. The shared secret tokens must consist only of printable ASCII characters.
Using Kerberos with GSA Mirroring You must configure Kerberos on the master and all mirrored search appliances by using the Serving > Universal Login Auth Mechanisms > Kerberos page in the Admin Console. Kerberos keytabs are unique, so ensure that you generate and import different Kerberos keytabs for the master and mirrored search appliances. When you configure Kerberos on a replica search appliance, use a different Mechanism Name from the one used for the master.
Task Description Determine the appliance IDs of the participating search appliances You view the appliance IDs on the Admin Console by right-clicking the About link at the bottom of any page and choosing Open in New Tab, or by navigating to Administration > License. Confirm that all search appliances are running the same software version. You view the software version on the Admin Console by right-clicking the About link at the bottom of any page and choosing Open in New Tab.
6. Check Enable (if not checked by default). 7. Click Done or Add Another Replica. 8. Optionally, click Export and save the mirroring configuration file. 9. On your replica (mirror) search appliance, log in and perform the following steps. 10. Right-click the About link, choose Open in New Tab, and confirm that the replica node is on the same software version as the master search appliance. 11. Click Status and Reports. The Crawl Status page is displayed. 12. Click Pause Crawl. 13.
15. Right-click the About link, choose Open in New Tab, and confirm that the replica node is on the same software version as the master search appliance. 16. Click Status and Reports. The Crawl Status page is displayed. 17. Click Pause Crawl. 18. Click Administration > Reset Index. 19. Click Reset Index Now. 20. On the GSAn > Configuration page for the master node, click Apply Configuration. This broadcasts the configuration data to all search appliances in the GSAn network.
• Configuring the IP address and other tasks that must be performed from the Network Configuration wizard • Configuring Kerberos settings. For complete information, see “Using Kerberos with GSA Mirroring” on page 9.
2. In a text editor, open the file server.xml. 3. Locate the Valve class and add the IP address of the replica search appliance to the RemoteAddressValve property. This enables the connector manager to respond to requests coming from the replica node. For example: Valve className="org.apache.catalina.valves.RemoteAddrValve" allow="127\.0\.0\.1|172\.25\.17\.88|172\.25\.16\.144|pattern_for_replica_IP_a ddress" For more information, see the Apache Tomcat Valve documentation (http://tomcat.apache.
If the master Google Search Appliance fails and a replica search appliance is promoted to be the master, do not directly add the former master node back as the master, because this will cause problems in the mirroring configuration. If you need to use the former master search appliance as the master, add it as a replica of the new master first. Wait until all index and configuration data are fully synchronized with the new master node, and then you can add the search appliance as the master again.
4. Click the Remove link that corresponds to the replica you are deleting from the configuration. 5. Click Apply Changes. 6. Log in to the Admin Console of the replica node. 7. Click GSAn > Configuration. 8. Click Disable GSAn. The Crawl Diagnostics page is not available for approximately 15 minutes after you disable GSAn. Disabling Mirroring on an Existing Replica Node Use these instructions to disable mirroring from the master node to a replica node.
Label Explanation Serving Index Sync Completion Time The date and time when the currently-serving index on a node was synchronized; for example, 2010/01/08 16:42. Serving Index Sync Time The amount of time it took for the currently-serving index to be completely synchronized from the master to the replica nodes; for example, 0h 30m. Syncing Index Creation Time The date and time when the index that is currently being synchronized was created; for example, 2010/01/10/17:14.
