You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Jan 19, 2024. It is now read-only.
The new official helm chart is hard to manage and hard to use in production. Here are main reasons:
There is literally no way to specify nodeSelector, custom affinity and tolerations. This is critical for production deployments because we often want to run Neo4j on dedicated nodes and this is just impossible with the new chart.
We need to manage separate releases for each new node in the cluster with the new official helm chart. This is very strange and completely opposite approach compared with the charts for other clustered databases (see bitnami's Kafka, Postgres, etc.). That is fine to define cores using CRDs if we have an operator but separate releases for each node with just a statefulset is something really strange.
I understand that there may be multiple important reasons for deprecating this community developed helm chart but maybe comments and +1 for this issue may help developers to react and support us.
The text was updated successfully, but these errors were encountered:
Sign up for freeto subscribe to this conversation on GitHub.
Already have an account?
Sign in.
The new official helm chart is hard to manage and hard to use in production. Here are main reasons:
nodeSelector
, customaffinity
andtolerations
. This is critical for production deployments because we often want to run Neo4j on dedicated nodes and this is just impossible with the new chart.I understand that there may be multiple important reasons for deprecating this community developed helm chart but maybe comments and +1 for this issue may help developers to react and support us.
The text was updated successfully, but these errors were encountered: