We currently had the problem that the creation of the search index of IBM Connections 4.5 never came to an end. In the log file you saw the following error messages:
[codesyntax lang=”text”]
|
1 2 3 4 5 |
CLFRW0957W: The seedlist url https://server.acme.com/activities/seedlist/myserver?Action=GetDocuments&Range=100&Start=1&Source=com.ibm.ilel.seedlist.retriever.connections.activities.ActivitiesRetrieverFactory&Format=xml&Locale=en_US&State=MTM0NjkyMjc3NTI3MDsxMTA0NTM0MDAwMDAwOzEzODMwNDE3NjAzODY7ZmFsc2U7MDswOzA7dHJ1ZTs%3D&useLocalFS=true has been requested more than twice in the same crawl.The application will now attempt to delete the seedlist pages persisted during this crawl activities. If this warning repeats on successive crawling or indexing jobs, there is a problem with the seedlist for the activities application. In this scenario, please contact IBM support. CLFRW0303E: Search has encountered a problem while crawling activities. The final seedlist page could not be reached. The crawl of activities could not be completed successfully. CLFRW0283E: Search has encountered a problem while crawling. CLFRW0294I: Search has finished crawling the activities component. CLFRW0962E: Some of the entries for activities were not indexed successfully. |
[/codesyntax]
Because of these problems activites started to index again and again and the search index was never built correctly.
We solved the issue by changing the following line in “search-config.xml”:
OLD:
<seedlist name=”activitiesContentSource” pageSize=”100“/>
NEW:
<seedlist name=”activitiesContentSource” pageSize=”500“/>
After restarting the cluster running the search application the index was able to finish indexing activities and the search index completed successfully. You might need to recreate your search index after you made the change (however it worked here without doing that).