bigtable: mutations_batcher() causes "Missing resource_project attribute for metric …" warnings every 60s #18546

Determine this is the right repository

• I determined this is the correct repository in which to report this bug.

Summary of the issue

Once a process has flushed a MutationsBatcher, the built-in client-side metrics exporter logs this warning on every export (every 60s), for each affected metric, until the process exits:

WARNING google.cloud.bigtable.data._metrics.handlers.gcp_exporter: Missing resource_project attribute for metric operation_latencies

Other operations such as bulk_mutate_rows, read_rows and mutate_row don't trigger it. In our production service this produced about 60k warnings in two days, and the batcher's metrics are never exported.

API client name and version

google-cloud-bigtable 2.48.0 (also present in 2.49.0)

Reproduction steps: code

import logging
import time

from google.cloud.bigtable.data import BigtableDataClient, RowMutationEntry, SetCell

logging.basicConfig(level=logging.WARNING)

client = BigtableDataClient(project="my-project")
table = client.get_table("my-instance", "my-table")

with table.mutations_batcher() as batcher:
    batcher.append(RowMutationEntry(b"row", [SetCell("family", b"q", b"v")]))

time.sleep(130)

Reproduction steps: actual results

WARNING:google.cloud.bigtable.data._metrics.handlers.gcp_exporter:Missing resource_project attribute for metric operation_latencies
...

The warning appears once per affected metric on every 60s export.

Reproduction steps: expected results

No warnings. The batcher's metrics are exported with the same resource labels as bulk_mutate_rows.

OS & version + platform

Linux (GKE, python:3.14-slim)

Python environment

Python 3.14.8

Python dependencies

google-cloud-bigtable 2.48.0, google-api-core 2.40.0, grpcio 1.84.0, opentelemetry-sdk 1.44.0