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
First off, nice project!
I tried a sample deployment of it and it works nicely for one instance. Would it be possible to adjust the operator so it works/calls the api of multiple ververica instances depending on the namespace the custom resources are deployed in?
Describe the solution you'd like
Currently the deployment pattern for this operator assumes only one instance is going to be managed by it. The --vvp-url expects one value and the api token env vars will most likely point to only one value as well.
Would it be possible to add a new parameter/change the existing one so multiple instances can be managed by this operator?
I imagine you'd have to 1) accept multiple urls to manage as an argument, 2) accept their version type (community/enterprise) and 3) an optional token per instance configured? This could be easily parsed from a config file perhaps.
The other bit that might be confusing is, if a VpDeployment shows up in a namespace, which ververica instance would that be deployed too? The easiest path forwards could be just deploy to the instance that lives in the same namespace as the custom resource?
Describe alternatives you've considered
I've also considered deploying this operator multiple times (once per instance of ververica) but that would entail modifying the CRD group.name per instance so they don't conflict and set the webhook.servicename to the name of each operator deployed. And then recompiling the operator each time to change the hardcoded group name in the code. OR set that group name as a flag that can be changed? It seems like a hack though and having one operator for all instances would be ideal.
What do you think?
The text was updated successfully, but these errors were encountered:
Hello,
First off, nice project!
I tried a sample deployment of it and it works nicely for one instance. Would it be possible to adjust the operator so it works/calls the api of multiple ververica instances depending on the namespace the custom resources are deployed in?
Describe the solution you'd like
Currently the deployment pattern for this operator assumes only one instance is going to be managed by it. The
--vvp-url
expects one value and the api token env vars will most likely point to only one value as well.Would it be possible to add a new parameter/change the existing one so multiple instances can be managed by this operator?
I imagine you'd have to 1) accept multiple urls to manage as an argument, 2) accept their version type (community/enterprise) and 3) an optional token per instance configured? This could be easily parsed from a config file perhaps.
The other bit that might be confusing is, if a VpDeployment shows up in a namespace, which ververica instance would that be deployed too? The easiest path forwards could be just deploy to the instance that lives in the same namespace as the custom resource?
Describe alternatives you've considered
I've also considered deploying this operator multiple times (once per instance of ververica) but that would entail modifying the CRD group.name per instance so they don't conflict and set the webhook.servicename to the name of each operator deployed. And then recompiling the operator each time to change the hardcoded group name in the code. OR set that group name as a flag that can be changed? It seems like a hack though and having one operator for all instances would be ideal.
What do you think?
The text was updated successfully, but these errors were encountered: