Check fault tolerance during initial indexing

Review the serverStatus web page to determine whether any exceptions occurred while building an initial index.

The Search service builds an initial index, and copies the index to each node in the cluster. This functionality works even if all of the Connections services cannot be fully crawled. The index is made available for users while any issues encountered during the initial index build are resolved.

1. After IBM Connections is installed, an indexing task runs and an initial index is built.

2. If an exception is found when crawling content from one or more services, the information needed to debug the problem is saved in the log file in SEARCH_INDEX_DIR. The name of this file starts with "InitialIndexBuildLog" and the file extension is .json. The information from this file is displayed on the serverStatus web page.

3. When index building completes, the index is copied automatically to each node in the cluster.

4. Review the serverStatus page to identify any issues encountered during the index build process.

   On the serverStatus webpage, the crawling exceptions encountered during an initial index build are displayed under the Initial Indexing Exceptions tab.

5. When the crawling exceptions are resolved for a service, the Search service attempts to re-crawl the service.

6. If the service is crawled and indexed successfully, an updated index is copied to each node in the cluster.

Parent topic:
The indexing process

Related:
Add a service to the search index
serverStatus web page