Rasa NLU server dying every few hours

Hi, I have a standalone RASA NLU server that dies every few hours. I don’t know the reason behind this or how to fix it. I am using the following command to start the server.

rasa run --enable-api -m PATH/TO/MODEL


language: en
  - name: ConveRTTokenizer
  - name: ConveRTFeaturizer
  - name: RegexFeaturizer
  - name: LexicalSyntacticFeaturizer
  - name: CRFEntityExtractor
  # added new
  - name: CountVectorsFeaturizer
  - name: CountVectorsFeaturizer
    analyzer: "char_wb"
    min_ngram: 1
    max_ngram: 4
  - name: DIETClassifier
    epochs: 50
    entity_recognition: False
  - name: EntitySynonymMapper
  - name: ResponseSelector
    epochs: 100

Hm, where are you hosting it? Is your VM restarting?

It is running is in a Docker container. Earlier I thought that the server may be running in the background in the container and the absence of any foreground process might be killing the docker. So, I scheduled a random lightweight process in the foreground. But later I found that the NLU server is being shut down automatically in the docker after a few hours. I have no idea why this is happening.

As a temporary fix, right now, I have scheduled a crontab job that checks(in every 3 minutes) if the NLU server is running. If it is not then the docker is restarted. But, I don’t want to keep production servers like that.

any luck with this? I am facing the same issue. I have hosted rasa server on EC2 instance