Repository navigation
Clickhouse and on cluster statements #3651
Replies: 2 comments 2 replies
|
Hi @electrical, it looks like you’re using a declarative workflow and the operator to apply changes. If so, you shouldn't use schema diff. Instead, you create a schema plan, push that plan to the Atlas registry, and then the operator will use the approved plan to apply changes to the target database. Check this: |
2 replies
|
Following up on this. |
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
Hi,
We are evaluating Atlas for our DB migrations and are testing it with Clickhouse and running into some issues blocking further progress.
We have a cluster of several nodes ( shards + replication ) which works fine when managing schemas manually.
For managing it with Atlas we have built the following:
atlas schema pushto the registry.From there we use the atlas kubernetes operator to pull in the changes and apply them to the database cluster.
The diff stage nicely shows the DDL's being planned with the
on clusterpart.When pushing this, we can't see any details of this
on clusterin there.Also, when its applied using the Operator, it seems to completely miss the
on clusterpart.In all the steps we provide dev-url to a clustered database setup that mimics our production setup at a smaller scale, but with the same config settings related to replication, sharding and naming.
In the atlas.hcl file we also have the following under the
envThis seems to do the right thing for the
diffcommand, but i couldn't find a way to ensure it does the same thing for anything else.I've gone through all the documentation I could find but at a loss after trying to make this work for a week.
I hope someone can point me in the right direction.
Happy to provide any further details if needed.
Thanks!
All reactions