display build URL for changepoint (#15)

* display build URL for changepoint Signed-off-by: Shashank Reddy Boyapally <[email protected]> * added tabulate Signed-off-by: Shashank Reddy Boyapally <[email protected]> * added buildUrls to hunter table Signed-off-by: Shashank Reddy Boyapally <[email protected]> * added buildUrls to same table, made changes for generic indexes, rebases Signed-off-by: Shashank Reddy Boyapally <[email protected]> * fixed trivial index error Signed-off-by: Shashank Reddy Boyapally <[email protected]> * moved index to config to avoid ES instance index collision Signed-off-by: Shashank Reddy Boyapally <[email protected]> * fixed timestamp bug Signed-off-by: Shashank Reddy Boyapally <[email protected]> --------- Signed-off-by: Shashank Reddy Boyapally <[email protected]>
cloud-bulldozer · May 9, 2024 · d3513fe · d3513fe
1 parent 186fab7
commit d3513fe
Show file tree

Hide file tree

Showing 5 changed files with 39 additions and 20 deletions.
diff --git a/examples/small-scale-cluster-density.yaml b/examples/small-scale-cluster-density.yaml
@@ -1,5 +1,7 @@
 tests :
   - name : aws-small-scale-cluster-density-v2
+    index: ospst-perf-scale-ci-*
+    benchmarkIndex: ospst-ripsaw-kube-burner*
     metadata:
       platform: AWS
       masterNodesType: m6a.xlarge

diff --git a/examples/small-scale-node-density-cni.yaml b/examples/small-scale-node-density-cni.yaml
@@ -1,5 +1,7 @@
 tests :
   - name : aws-small-scale-node-density-cni
+    index: ospst-perf-scale-ci-*
+    benchmarkIndex: ospst-ripsaw-kube-burner*
     metadata:
       platform: AWS
       masterNodesType: m6a.xlarge

diff --git a/orion.py b/orion.py
@@ -1,19 +1,23 @@
 """
 This is the cli file for orion, tool to detect regressions using hunter
 """
+
 # pylint: disable = import-error
 import sys
+import warnings
 from functools import reduce
 import logging
 import os
 import re
+import pyshorteners
 
 import click
 import pandas as pd
 
 from fmatch.matcher import Matcher
 from utils import orion_funcs
 
+warnings.filterwarnings("ignore", message="Unverified HTTPS request.*")
 
 @click.group()
 # pylint: disable=unused-argument
@@ -22,7 +26,8 @@ def cli(max_content_width=120):
     cli function to group commands
     """
 
-# pylint: disable=too-many-locals
+
+# pylint: disable=too-many-locals, too-many-statements
 @click.command()
 @click.option("--uuid", default="", help="UUID to use as base for comparisons")
 @click.option("--baseline", default="", help="Baseline UUID(s) to to compare against uuid")
@@ -50,61 +55,68 @@ def orion(**kwargs):
     ES_URL=None
 
     if "ES_SERVER" in data.keys():
-        ES_URL = data['ES_SERVER']
+        ES_URL = data["ES_SERVER"]
     else:
-        if 'ES_SERVER' in os.environ:
-            ES_URL=os.environ.get("ES_SERVER")
+        if "ES_SERVER" in os.environ:
+            ES_URL = os.environ.get("ES_SERVER")
         else:
             logger.error("ES_SERVER environment variable/config variable not set")
             sys.exit(1)
-
+    shortener = pyshorteners.Shortener()
     for test in data["tests"]:
+        benchmarkIndex=test['benchmarkIndex']
         uuid = kwargs["uuid"]
         baseline = kwargs["baseline"]
-        match = Matcher(index="perf_scale_ci", level=level, ES_URL=ES_URL)
+        match = Matcher(index="ospst-perf-scale-ci-*",
+                        level=level, ES_URL=ES_URL, verify_certs=False)
         if uuid == "":
             metadata = orion_funcs.get_metadata(test, logger)
         else:
             metadata = orion_funcs.filter_metadata(uuid,match,logger)
 
         logger.info("The test %s has started", test["name"])
         if baseline == "":
-            uuids = match.get_uuid_by_metadata(metadata)
+            runs = match.get_uuid_by_metadata(metadata)
+            uuids = [run["uuid"] for run in runs]
+            buildUrls = {run["uuid"]: run["buildUrl"] for run in runs}
             if len(uuids) == 0:
                 logging.info("No UUID present for given metadata")
                 sys.exit()
         else:
             uuids = [uuid for uuid in re.split(' |,',baseline) if uuid]
             uuids.append(uuid)
-        if metadata["benchmark.keyword"] == "k8s-netperf" :
-            index = "k8s-netperf"
-            ids = uuids
-        elif metadata["benchmark.keyword"] == "ingress-perf" :
-            index = "ingress-performance"
+        index=benchmarkIndex
+        if metadata["benchmark.keyword"] in ["ingress-perf","k8s-netperf"] :
             ids = uuids
         else:
-            index = "ripsaw-kube-burner"
             if baseline == "":
-                runs = match.match_kube_burner(uuids)
+                runs = match.match_kube_burner(uuids, index)
                 ids = match.filter_runs(runs, runs)
             else:
                 ids = uuids
 
         metrics = test["metrics"]
         dataframe_list = orion_funcs.get_metric_data(ids, index, metrics, match, logger)
 
+        for i, df in enumerate(dataframe_list):
+            if i != 0:
+                dataframe_list[i] = df.drop(columns=['timestamp'])
+
         merged_df = reduce(
             lambda left, right: pd.merge(left, right, on="uuid", how="inner"),
             dataframe_list,
         )
 
+        shortener = pyshorteners.Shortener()
+        merged_df["buildUrl"] = merged_df["uuid"].apply(
+            lambda uuid: shortener.tinyurl.short(buildUrls[uuid])) #pylint: disable = cell-var-from-loop
         csv_name = kwargs["output"].split(".")[0]+"-"+test['name']+".csv"
         match.save_results(
             merged_df, csv_file_path=csv_name
         )
 
         if kwargs["hunter_analyze"]:
-            orion_funcs.run_hunter_analyze(merged_df,test)
+            _ = orion_funcs.run_hunter_analyze(merged_df,test)
 
 
 if __name__ == "__main__":

diff --git a/requirements.txt b/requirements.txt
@@ -3,10 +3,11 @@ certifi==2023.11.17
 click==8.1.7
 elastic-transport==8.11.0
 elasticsearch==7.13.0
-fmatch==0.0.5
+fmatch==0.0.7
 python-dateutil==2.8.2
 pytz==2023.3.post1
 PyYAML==6.0.1
 six==1.16.0
 tzdata==2023.4
 urllib3==1.26.18
+pyshorteners==1.0.1
diff --git a/utils/orion_funcs.py b/utils/orion_funcs.py
@@ -25,11 +25,12 @@ def run_hunter_analyze(merged_df,test):
     merged_df["timestamp"] = merged_df["timestamp"].astype(int) // 10**9
     metrics = {column: Metric(1, 1.0)
                for column in merged_df.columns
-               if column not in ["uuid","timestamp"]}
+               if column not in ["uuid","timestamp","buildUrl"]}
     data = {column: merged_df[column]
             for column in merged_df.columns
-            if column not in ["uuid","timestamp"]}
-    attributes={column: merged_df[column] for column in merged_df.columns if column in ["uuid"]}
+            if column not in ["uuid","timestamp","buildUrl"]}
+    attributes={column: merged_df[column]
+                for column in merged_df.columns if column in ["uuid","buildUrl"]}
     series=Series(
         test_name=test["name"],
         branch=None,
@@ -42,6 +43,7 @@ def run_hunter_analyze(merged_df,test):
     report=Report(series,change_points)
     output = report.produce_report(test_name="test",report_type=ReportType.LOG)
     print(output)
+    return change_points
 
 # pylint: disable=too-many-locals
 def get_metric_data(ids, index, metrics, match, logger):
@@ -71,7 +73,7 @@ def get_metric_data(ids, index, metrics, match, logger):
                 agg_value = metric['agg']['value']
                 agg_type = metric['agg']['agg_type']
                 agg_name = agg_value + "_" + agg_type
-                cpu_df = match.convert_to_df(cpu, columns=["uuid", agg_name])
+                cpu_df = match.convert_to_df(cpu, columns=["uuid","timestamp", agg_name])
                 cpu_df = cpu_df.rename(
                     columns={agg_name: metric_name+ "_" +  agg_name}
                 )