Merge pull request redpanda-data#17615 from redpanda-data/PESDLC-1063…

…-htt-test-consume-miss-cache

PESDLC-1063 Refactor test_consume_miss_cache

tests/rptest/redpanda_cloud_tests/high_throughput_test.py

@@ -1441,7 +1441,6 @@ def consumer_saw_msgs(consumer):
     # The testcase occasionally fails on various parts:
     # - toing on `_consume_from_offset(self.topic, 1, p_id, "newest", 30)`
     # - failing to ensure all manifests are in the cloud in `stop_and_scrub_object_storage`
-    @ignore
     @cluster(num_nodes=7, log_allow_list=RESTART_LOG_ALLOW_LIST)
     def test_consume_miss_cache(self):
         # create default topics
@@ -1498,14 +1497,11 @@ def stage_consume_miss_cache(self, producer: KgoVerifierProducer):
         partition_size_check: list[MetricCheck] = []
         partition_size_metric = "vectorized_storage_log_partition_size"
 
-        # https://github.com/redpanda-data/cloudv2/issues/10685#issuecomment-1893009486
-        raise NotImplementedError('partition_size_check not implemented')
-
-        for node in self.redpanda.nodes:
+        for pod in self.redpanda.pods:
             partition_size_check.append(
                 MetricCheck(self.logger,
                             self.redpanda,
-                            node,
+                            pod,
                             partition_size_metric,
                             reduce=sum))
 
@@ -1598,11 +1594,11 @@ def check_partition_size(old_size, new_size):
             "vectorized_storage_log_cached_read_bytes_total",
         ]
 
-        for node in self.redpanda.nodes:
+        for pod in self.redpanda.pods:
             check_batch_cache_reads.append(
                 MetricCheck(self.logger,
                             self.redpanda,
-                            node,
+                            pod,
                             cache_metrics,
                             reduce=sum))
 

tests/rptest/services/metrics_check.py

@@ -97,12 +97,18 @@ def _capture(self, check_metrics):
             self.logger.info(f"  Captured {k}={v}")
 
         if len(samples) == 0:
-            metrics_endpoint = ("/metrics" if self._metrics_endpoint
-                                == MetricsEndpoint.METRICS else
-                                "/public_metrics")
-            url = f"http://{self.node.account.hostname}:9644{metrics_endpoint}"
-            import requests
-            dump = requests.get(url).text
+            # Announce
+            dump = "No metrics extracted"
+            # handle cloud cluster separately
+            if getattr(self.redpanda, "_cloud_cluster", None) is None:
+                metrics_endpoint = ("/metrics" if self._metrics_endpoint
+                                    == MetricsEndpoint.METRICS else
+                                    "/public_metrics")
+                url = f"http://{self.node.account.hostname}:9644{metrics_endpoint}"
+                import requests
+                dump = requests.get(url).text
+            else:
+                dump = self.redpanda._cloud_cluster.get_public_metrics()
             self.logger.warn(f"Metrics dump: {dump}")
             raise RuntimeError("Failed to capture metrics!")