Open source feature flagging and experimentation platform for stress free deployments, measured impact, and smarter decisions.

GrowthBook Users

Hi Anton, we don’t have a lot of documentation on that -but I do know it works. We have many customers on kubernetes

Hi :wave:,
I am glad to hear that it's possible to deploy on kubernetes. If you know someone who find out how to solve this matter, I would be happy :simple_smile:

I detailled everything in an issue
<https://github.com/growthbook/growthbook/issues/1303>

It seems the front application can't connect to the back

I am just doing it on a VM for now and not in a cluster so I sadly can’t help.

Oh, Adda and I sorted it out- we had to open a second port for the API and adjust the API_HOST and APP_ORIGIN to match the public urls

Exactly, I had the same problem. Thanks to Graham now it's solved.

image.png

hi <@U059YBR0JLR> <@U01T6HCHD0A>
In my case, the backend keeps getting killed.
I have set the `API_HOST` and `APP_ORIGIN` .
I have also `CORS_ORIGIN_REGEX` and `JWT_SECRET` suggested in the issue but it's still the same.

Any suggestions please?

often this can happen if you are not able to connect to Mongo

there's nothing specific in the logs, only what I shared in the screenshot.

I can also confirm that it was able to connect to Mongo as it created these collections

Can you DM me the env variables or docker compose yml file?

Oh, I'm not using a docker compose yml.
I'm deploying it to kubernetes. However I'm using the same configuration that worked fine when i used docker compose

Yhh, I checked this too.
Sometimes, it's the frontend and most times it's the back end but I can't seem to get both of them up at the same time

this is another common error when I restart deployments

Hi <@U01T6HCHD0A>
I figured out what my issue was. This may be useful to someone else.

I pre-configured memory requests and limits of about 200MiB whereas growthbook pods use about 400MiB

The configuration below works okay for me. This configuration can also be ignored.