README.md

Bigquery metrics

This metrics-bigquery job generates metrics that summarize data in our Bigquery test result database. Each metric is defined with a config file that is consumed by the metrics-bigquery periodic prow job. Each metric config is a yaml file like the following:

# Metric name
metric: failures
# BigQuery query
query: |
  #standardSQL
  select /* find the most recent time each job passed (may not be this week) */
    job,
    max(started) latest_pass
  from `k8s-gubernator.build.all`
  where
    result = 'SUCCESS'
  group by job

# JQ filter to make daily results from raw query results
jqfilter: |
  [(.[] | select((.latest_pass|length) > 0)
  | {(.job): {
      latest_pass: (.latest_pass)
  }})] | add

# JQ filter to make influxdb timeseries data points for Velodrome. (Optional)
jqmeasurements: |
  [(.[] | select((.latest_pass|length) > 0) | {
    measurement: "latest_pass_time",
    tags: {
      job: (.job)
    },
    fields: {
      job: (.job),
      latest_pass: (.latest_pass)
  }})]

Metrics

• build-stats - number of daily builds and pass rate
  • Config
  • build-stats-latest.json
• presubmit-health - presubmit failure rate and timing across PRs
  • Config
  • presubmit-health-latest.json
• failures - find jobs that have been failing the longest
  • Config
  • failures-latest.json
• flakes - find the flakiest jobs this week (and the flakiest tests in each job).
  • Config
  • flakes-latest.json
• flakes-daily - find flakes from the previous day. Similar to flakes, but creates more granular results for display in Velodrome.
  • Config
  • flakes-daily-latest.json
• job-flakes - compute consistency of all jobs
  • Config
  • job-flakes-latest.json
• pr-consistency - calculate PR flakiness for the previous day.
  • Config
  • pr-consistency-latest.json
• weekly-consistency - compute overall weekly consistency for PRs
  • Config
  • weekly-consistency-latest.json
• istio-job-flakes - compute overall weekly consistency for postsubmits
  • Config
  • istio-job-flakes-latest.json

Adding a new metric

To add a new metric, create a PR that adds a new yaml config file specifying the metric name (metric), the bigquery query to execute (query), and a jq filter to filter the data for the daily and latest files (jqfilter). Optionally: Include a jqfilter to extract influxdb timeseries measurements from the raw query results (jqmeasurements).

Run ./bigquery.py --config configs/my-new-config.yaml and verify that the output is what you expect.

Add the new metric to the list above.

After merging, find the new metric on GCS within 24 hours.

Details

Each query is run every 24 hours to produce a json file containing the complete raw query results named with the format raw-yyyy-mm-dd.json. The raw file is then filtered with the associated jq filter and the results are stored in daily-yyyy-mm-dd.json. These files are stored in the k8s-metrics GCS bucket in a directory named with the metric name and persist for a year after their creation. Additionally, the latest filtered results for a metric are stored in the root of the k8s-metrics bucket and named with the format METRICNAME-latest.json.

If a config specifies the optional jq filter used to create influxdb timeseries data points, then the job will use the filter to generate timeseries points from the raw query results. The points are uploaded to Velodrome's influxdb instance where they can be used to create graphs and tables.

Consistency

Consistency means the test, job, pr always produced the same answer. For example suppose we run a build of a job 5 times at the same commit:

• 5 passing runs, 0 failing runs: consistent
• 0 passing runs, 5 failing runs: consistent
• 1-4 passing runs, 1-4 failing runs: inconsistent aka flaked