Posted by Zachary Wick (M-Files) on 24 January 2019 09:34 PM
NOTE: IDOL installation and configuration is currently a complex setup and it is generally suggested to have the assistance of M-Files or other skilled staff in setting it up and configuration.
Q: What's the difference between the regular M-Files search engine and IDOL?
A: More information can be found in this document from our knowledge base https://kb.cloudvault.m-files.com/Default.aspx?#3ECA226F-7B54-428B-B539-DE443E6134EC/object/5C96D90F-2620-4913-90C7-ED99D7F8DA16/latest
A: IDOL has a very strict start up procedure. The DIH-DAH component should always be started last during initial startup. Check the application.log and content_application.log in the IDOL\logs folder as they will have more information about what is wrong and will also assist support in troubleshooting the issue.
Q: I'm seeing the following in my content_application.log whenever I try to start IDOL. "14/11/2018 14:38:11  00-Always: Error loading unstemmed index (Invalid)"
A: The unstemmed index can be caused when the cluster was not started in the proper order the first time. It can also be caused by virus protection scanning index files.
Q: I'm seeing the following error in my logs "13/11/2018 22:02:06  70-Error: Action failed - unencrypted communication is not allowed (22.214.171.124): "ACTION=GETSTATUS""
A: This error usually means an error in the configuration or an error in the firewall blocking communication.
Q: How do I know if IDOL is indexing files from the M-Files Desktop?
A: If IDOL is successfully indexing files, you can search for a text string in the Desktop and check the "Score" column. If the scores are larger than 4000, IDOL is working. If it's around 20,000 or larger, it's still using dtSearch.
Q: How do we handle redundancy?
A: IDOL redundancy can be achieved in a few different ways. It depends on what the goal is. If you want a full hot site, you just stand up another cluster and have it index the vault at the same time as the other cluster. Just add registry entries for the second IDOL with a different GUID. Set one (registry …/VaultOptions/<GUID>/) ActiveCombinedIndex to point to one of them, set both of them enabled. Now both will be generating indexes. In case of detected failure, just change the ActiveCombinedIndex to point to the GUID of the other IDOL cluster and perform a vault off-line/on-line.
Q: How do we handle fail overs?
A: If you want load balancing/fail over, it can be configured to have multiple DIH-DAH's in the same cluster. The fail over would be handled by third party hardware/software
Q: What are the best practices for backing up the IDOL configuration?
A: You should be backing up all of the IDOL configuration folders, but as a general best practice it would probably be best to back up the entire drive. You will want to backup index files. This way if it were to fail you don't have to wait for a complete re-index.