Hi,
One month ago approximately, we changed our Decipher configuration from a single server to a multi-service deployment. The following is the multi-device deployment distribution we are currently using:
Server 1: Decipher Licensing Service, Decipher Web Client and RabbitMQ.
Server 2: Decipher Automated Clients
Server 3: Decipher Server and Decipher Server Plugin
Server 4: SQL server.
However, we realized that the Capture models we defined (they were defined and trained before the new installation) are not being trained automatically, apparently because they do not recognize that they have documents available for training. As you can see in the images attached, the column corresponding to "New Documents Available" is always empty (before that it says Loading...), and we know that more than 50 documents have been processed by the model since the new installation.
Could you please help us with this issue or have you seen any similar cases? Which could be the point that may be causing this problem?
Please let me know if you need more details.
Thank you in advance
Ander
Hi Ander,
The installation process for a multi-device environment has some slight differences to ensure Decipher can operate across the separate servers.
Please can you confirm if the service account being used to write to the Decipher Db has the correct folder permissions as outlined below?
And please confirm that machine learning has been enabled in the config file.
If these are both completed successfully, I would recommend raising a support ticket.
Thanks
Ben
