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
Is your feature request related to a problem? Please describe.
The operator does not allow me to deploy a daemonset collector with any other allocation strategy than per-node. This is very reasonable, but if I add a prometheus scrape job that scrapes a single non-node-specific target it will not be allocated as that is currently the expected functionality of the "per-node" allocation strategy.
See below example of failure to deploy as daemonset with "consistent-hashing" allocation strategy.
Error: UPGRADE FAILED: an error occurred while rolling back the release. original upgrade error: cannot patch "opentelemetry" with kind OpenTelemetryCollector: admission webhook "vopentelemetrycollectorcreateupdatebeta.kb.io" denied the request: the OpenTelemetry Collector mode is set to daemonset, which must be used with target allocation strategy per-node
Describe the solution you'd like
I think there should be a fallback, ideally just using "consistent-hashing" or "least-weighted" for the leftover targets after "per-node" allocation is done.
Describe alternatives you've considered
No response
Additional context
The possibility of adding a fallback was talked about in this issue, but never acted upon.
The text was updated successfully, but these errors were encountered:
Component(s)
target allocator
Is your feature request related to a problem? Please describe.
The operator does not allow me to deploy a daemonset collector with any other allocation strategy than per-node. This is very reasonable, but if I add a prometheus scrape job that scrapes a single non-node-specific target it will not be allocated as that is currently the expected functionality of the "per-node" allocation strategy.
See below example of failure to deploy as daemonset with "consistent-hashing" allocation strategy.
Error: UPGRADE FAILED: an error occurred while rolling back the release. original upgrade error: cannot patch "opentelemetry" with kind OpenTelemetryCollector: admission webhook "vopentelemetrycollectorcreateupdatebeta.kb.io" denied the request: the OpenTelemetry Collector mode is set to daemonset, which must be used with target allocation strategy per-node
Describe the solution you'd like
I think there should be a fallback, ideally just using "consistent-hashing" or "least-weighted" for the leftover targets after "per-node" allocation is done.
Describe alternatives you've considered
No response
Additional context
The possibility of adding a fallback was talked about in this issue, but never acted upon.
The text was updated successfully, but these errors were encountered: