Building Lucene index should not load all messages in memory #409
Comments
guusdk
added a commit
to guusdk/openfire-monitoring-plugin
that referenced
this issue
Dec 10, 2024
…mory when indexing When iterating over all rows of a potentially large table, ensure that the database result set is configured to be 'linear' (forward-only and read-only). This gives a better chance of the database driver to release all rows that have been iterated over, which prevents the fetch buffer to eventually include all rows (and potentially cause out-of-memory issues).
guusdk
added a commit
to guusdk/openfire-monitoring-plugin
that referenced
this issue
Dec 10, 2024
…mory when indexing When iterating over all rows of a potentially large table, ensure that the database result set is configured to be 'linear' (forward-only and read-only). This gives a better chance of the database driver to release all rows that have been iterated over, which prevents the fetch buffer to eventually include all rows (and potentially cause out-of-memory issues).
guusdk
added a commit
to guusdk/openfire-monitoring-plugin
that referenced
this issue
Jan 7, 2025
…mory when indexing When iterating over all rows of a potentially large table, ensure that the database result set is configured to be 'linear' (forward-only and read-only). This gives a better chance of the database driver to release all rows that have been iterated over, which prevents the fetch buffer to eventually include all rows (and potentially cause out-of-memory issues).
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Out of Memory Errors have been reported when the Lucene index is being built (which typically happens at startup). This occurs with some database driver (MSSQL for example), that, while processing a large SQL result set (which contains all rows of a the table that contains archived messages), keeps all rows that it has iterated over in memory.
The implementation should cause the database driver to discard rows that have been processed instead.
As a workaround, the property
conversation.search.index-enabledcan be set tofalsein or after release 2.6.0 of the plugin. This prevents the Lucene index from being created. As a result, full text search will yield no results.The text was updated successfully, but these errors were encountered: