The error is fixed in the subsequent maintenance LTS release (version 5.0.9), therefore customers should upgrade their installations to version 5.0.9 or later, in order to get their new sessions indexed. However, the upgrade doesn't attempt to reindex the sessions for which the indexing failed previously. Please read below for a resolution.
If you cannot upgrade your SPS deployment immediately but you can use external indexers, you may want to upgrade your external indexers to an updated version 5.0.8b, which is attached to this article.
Please refer to the Configuring external indexers in the Administration Guide about how to set up and configure the external indexer, and also disable the integrated indexers by setting the number of workers to 0 at Basic Settings > Local Services > Indexer service.
Unfortunately, it is not possible to correct the error on a running SPS deployment, therefore either the SPS or the external indexers must be upgraded.
Unfortunately upgrading to the fixed version doesn't attempt to work off the failed indexing backlog, this must be performed by the administrator manually via the command line.
Please upload the attached shell script to the SPS to the /mnt/firmware/root directory. Please log in to the core shell of the appliance and issue the following commands:
chmod +x /root/reindex-failed-sessions.sh
The script removes the sessions for which indexing failed with a specific error message from the indexing queue and inserts them to the active indexing queue. Depending on the number and the size of the failed to index sessions, reindexing may take considerable time.