Model not loaded when using Docker Compose

[LondonAppDev]

Hello,

I am running OpenSearch using Docker Compose for a local development environment.

I have mapped a volume on my container to /usr/share/opensearch/data in order to persist data when I stop/start the OpenSearch service.

I believe this works, since I can see the data stored in that directory when I index things in OpenSearch, and when I restart the service the data is still there.

However, when I load a model using the ML plugin, the model does not persist, and when I restart the service I get exceptions saying: Model not loaded.

My question is: Is there a different path which the model data is stored, outside of /usr/share/opensearch/data? Basically I'm wondering if there are any other paths which need to be included in my volume mapping in order to persist the index and the loaded ML models?

Thanks in advance,
Mark

[LondonAppDev]

I figured out the issue.

Basically, every time I restart OpenSearch, I need to load the model gain. I expect this is because the model needs to be "loaded" into memory in order to use it.