Hi. starting yesterday I am getting

AuthenticationFailed

on backend connecting to mongodb. I am using this on kubernetes using this helm chart by @Jaegu Kim. I checked

MONGODB_ROOT_PASSWORD

env var for mongodb is same as what we use in

MONGODB_URI

. has there been any change in

growthbook/growthbook:latest

image?

Hello team, if my experiment query has an error that says "Precondition failed" it's a duplicate of another experiment that doesn't have the error, what could be the problem? Thank you in advance.

Just wanted to clarify a few things about general bucketing and `trackingCallback`: • What if weights are changed in the json file for an experiment? Can a user change buckets if they have been bucketed in a certain bucket, but then the weights change for an experiment? Can a user change buckets in any other circumstance? • Is

trackingCallback

called every time the experiment is viewed or only the first time an experiment group assigned? I believe from my tests it is the former.

Does growthbook support split url testing for websites?

Hi All. Just started doing a POC for growthbook with playwright test automation. 🙂 works really well from my local environment. However once I try to run the tests on CI (Bitbucket) seems like every flag is returned as disabled. Has anyone ever ran into something like this, where the exact same code works locally and flags return the expected value, but in CI they return always as disabled? Done some digging through the Github Issues and seems like I could not find anything.

Hey! 😄 I wanted to try self hosting Growthbook in a kubernetes cluster, is there any documentation available on how to achieve this? Should I convert the docker compose file to a kubernetes resource?

Are there any plans to add

OR

operator in experiment targeting conditions?

Hi there, I'm running into an issue getting sticky bucketing/fallback attributes to work as per the documentation and need some help identifying if this is a bug/implementation issue or a misunderstanding. For context, I want to ensure that a user who initially sees a variation when logged out will continue to see that variation when they log in. Based on this documentation I believe I can achieve that by supplying a fallbackAttribute when the user is logged out and then providing the same fallbackAttribute with a hashAttribute when the user logs in. But when I implement this users are still seeing a different variation when logged in versus logged out, here's a loom demoing the issue: https://www.loom.com/share/bdc9d2e0ae1341029238d421433f39b6 So, have I understood this correctly, or is a different implementation required to persist variations when a user goes from logged out to logged in? Our implementation details: • Vue2 using the GrowthBookVuePlugin and LocalStorageStickyBucketService • StickyBucketing & Fallback Attribute are enabled in settings and the Fallback attribute is enabled on the experiment (confirmed working in loom above) • hashAttribute is id and fallbackAttribute is device_id ◦ *id i*s only provided when users are logged in, device_id is always sent

Hello Growthbook team, I've setup growthbook with mixpanel. I'm trying to test a simple A/B experiment, the metrics are getting tracked correctly, the experiment event is also getting logged on the mixpanel dashboard, unfortunately there are no experiment results getting displayed as seen here, what could be my problem? (I've decreased the sample size to make sure that the results will get displayed after 3 conversions)

Hey there, I had a question around how Growthbook knows that a user has been assigned during a % based feature flag, as in how does Growthbook store the state so as to cap at 10% of users? Love using Growthbook so far, and I am self hosting if that’s useful info. Thanks!