x/browbeat
Code Issues Proposed changes

Ignore sending null values from shaker output to Elasticsearch

Browse Source 
Shaker reports throughput/ping values as None for the first and last
few seconds of the test for synchronization. Currently we are pushing
null values also to ES, which most likely is affecting averages etc.
This commit prevents shaker from reporting these values to Elasticsearch.

Fixing linting errors in lib/Elastic.py

Change-Id: Ib8aaf425d68fae1254c3692ebf39e885888e497a
This commit is contained in:
Sai Sindhur Malleni 2016-09-29 17:14:48 -04:00
parent b04de2d9d3
commit 5e8267ec18
2 changed files with 7 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
lib/Elastic.py
View File

@ -96,8 +96,8 @@ class Elastic(object):
doc_type=_type,
refresh=True)
self.logger.info("Pushed data to Elasticsearch to index {}"
" and browbeat UUID {}" .format(self.index,
result['browbeat_uuid']))
" and browbeat UUID {}" .format(self.index,
result['browbeat_uuid']))
break
except Exception:
self.logger.error("Error pushing data to Elasticsearch, going to retry"

7
lib/Shaker.py
View File

@ -172,8 +172,11 @@ class Shaker(WorkloadBase.WorkloadBase):
self.grafana.grafana_urls()],
'shaker_uuid': str(shaker_uuid)}
# Ship Data to ES when record status is ok
result = self.elastic.combine_metadata(shaker_stats)
self.elastic.index_result(result, test_name)
if result['value'] is None:
self.logger.debug("Ignoring sending null values to ES")
else:
result = self.elastic.combine_metadata(shaker_stats)
self.elastic.index_result(result, test_name)
else:
# If the status of the record is not ok, ship minimal
# shaker_stats dictionary to ES