Environments.yml

@nik202

I need some help, I installed rasax and rasa open source using kubernetes and helm.

I have followed the instructions as much as possible but I am getting this error regarding the envrionments, any idea how to deal with the environments.yml file? Where it is found? e.t.c

(base) admin:$ kubectl logs rasa-x-1651686780-rasa-x-65d6bc5765-r6rlc -n rasaxv1
Starting Rasa X server... ๐Ÿš€
[2022-05-10 10:40:33 +0000] [11] [INFO] Goin' Fast @ http://0.0.0.0:5002
[2022-05-10 10:40:33 +0000] [24] [INFO] Starting worker [24]
[2022-05-10 10:40:33 +0000] [22] [INFO] Starting worker [22]
[2022-05-10 10:40:33 +0000] [21] [INFO] Starting worker [21]
[2022-05-10 10:40:33 +0000] [26] [INFO] Starting worker [26]
WARNING:rasax.community.services.settings_service:Could not inject deployment environments configuration from file 'environments.yml'. Details:
Environment needs to contain entries for production and worker services.
WARNING:rasax.community.services.settings_service:Rasa X will continue to use your current configuration.Your current configuration is:
{
  "environments": {
    "rasa": {}
  }
}
``

@ChrisRahme any ideas on this please? I have been trying to find out but to no avail. This is stopping my model from training after attaching the rasax instance.

No, sorry :confused:

@ChrisRahme

anyone you know who would have an idea?

@atwine me neither. I m expertise in docker and working on Kub.

@nik202

Anyone you know who could assist here?

pinging @stephens for your help please and many thanks in advance.

I no longer use Rasa X.

@stephens Thanks Greg for the prompt reply :slight_smile:

@atwine Iโ€™m sorry I donโ€™t know anyone now and those I know they all left Rasa. Please google may be he will help.

@nik202 thank you for trying. I will see how to figure it out.

@stephens What does your set up look like on the front end?

Bot maintenance is done with an IDE and git workflow (branching, PRโ€™s, merging, etc). For conversation review, either use the event broker to forward the tracker to an aggregation and dashboard environment (like the ELK stack with Kibana dashboard) and I also have a simple Typescript UI to read the tracker db directly and view the conversations in a web browser (this is a view only UI). Have also looked at Botario.

I found the solution:

  • The issue was in the helm chart, I had not turned on one of the values that would issue the production pod and this caused errors. Once I updated this in the rasax helm chart, everything connected.

Next challenge is to connect external open source servers.

1 Like