[feature request] InfluxDB backpressure support

[SKuckluck]

Component

OpenTelemetry.Exporter.InfluxDB

Is your feature request related to a problem?

Yes. In production environments with sustained high metric volume, the InfluxDB metrics exporter cannot keep up with the incoming write rate. Pending metric batches accumulate faster than they are exported, which causes unbounded memory growth and can eventually lead to process instability or restarts.

What is the expected behavior?

The exporter should support configurable backpressure handling so that memory usage remains bounded when the write rate cannot keep up with the incoming metric rate.
When the pending queue reaches capacity, the exporter should be able to:

• block until space becomes available,
• drop the current batch, or
• evict the oldest queued batch and enqueue the current one instead.

The default behavior should remain unchanged when this feature is not enabled.

Which alternative solutions or features have you considered?

None

Additional context

This behavior was observed in a production deployment. The issue is not that telemetry should be blocked indefinitely, but that the exporter currently has no controlled way to handle overload, which allows memory usage to grow without bound.

Tip

_{React with 👍 to help prioritize this issue. Please use comments to provide useful context, avoiding +1 or me too, to help us triage it. Learn more here.}