feat: set --max-poll-interval-ms flags to Sentry/Snuba consumers (#4376)

* feat: set --max-poll-interval-ms flags to Sentry/Snuba consumers

Separated from #4313

Also introducint a healthcheck for snuba-replacer

* fix: health-check args for snuba-replacer

Author: aldy505

.env

@@ -30,16 +30,23 @@ HEALTHCHECK_FILE_RETRIES=3
 HEALTHCHECK_FILE_START_PERIOD=600s
 # Set SETUP_JS_SDK_ASSETS to 1 to enable the setup of JS SDK assets
 # SETUP_JS_SDK_ASSETS=1
-#
+
 # Set SETUP_CUSTOM_CA_CERTIFICATE to 1 to generate trust stores that inject
 # your custom CA certificates (from ./certificates/*.crt) into non-Sentry
 # services (relay, symbolicator, vroom, snuba, taskbroker, uptime-checker).
 # Re-run ./install.sh after adding or changing certificates — no Docker image
 # rebuilds are required. Follow the printed instructions to update your
 # docker-compose.override.yml on the first run.
 # SETUP_CUSTOM_CA_CERTIFICATE=1
-#
+
 # Sentry (and its' surrounding services) uses Statsd for metrics collection.
 # It's also the proper way to monitor self-hosted Sentry systems.
 # Set STATSD_ADDR to a valid IP:PORT combination to enable Statsd metrics collection.
 # STATSD_ADDR=127.0.0.1:8125
+
+# Set SENTRY_KAFKA_MAX_POLL_INTERVAL_MS to a value (in milliseconds) that is
+# higher than the expected maximum processing time of a batch of messages for
+# the Kafka consumers. This will prevent Kafka from considering the consumer
+# as failed and triggering a rebalance while it is still processing messages.
+# The default value is 30000 ms (30 seconds).
+# SENTRY_KAFKA_MAX_POLL_INTERVAL_MS=30000