0.7.0

Features:

• Add mtr_report_path_id metric and path_id label to make path changes
  easier to detect. The path_id is a checksum over the concatenated
  hops of a specific mtr run.
• Switch internal logging to structured logging (log/slog).
  Log line format changes - downstream log parsers may need adjustment.
• Add -timeshift flag and support for random delays in the jobs file.
  Useful to spread load when running many jobs.
• Provide a Grafana dashboard alongside the documentation.

Bug Fixes:

• Fix mtr_report_min_loss: when mtr-exporter was used with multiple jobs,
  mtr_report_min_loss carried the minimum loss of the last job.

Maintenance:

• Add -show-license flag to print the bundled license text.
• Bump base container image to Alpine:3.23 (mtr:0.96).
• Bump Go toolchain to 1.26.3.
• Container images now published for linux/amd64 and linux/arm64.

0.6.0

Features:

• Add mtr_report_min_loss metric: observed minimal packet loss for
  a given run, defaults to 100.

  Improves the situation for GH#33.

Maintenance:

• Bump base container image to Alpine:3.22
  (no update of mtr, still on 0.95)

0.5.1

Bug Fixes:

• Fix improper placing of label "error" into the metrics. Fixes GH#32

0.5.0

Features

• Add mtr_runs_total metric: successful runs and failed runs
  (which get label "error") are exposed. As a result, one can
  check intermediate (or permanent) failures.

  sum by(mtr_exporter_job)(mtr_runs_total{}) - provides the
  absolute number of mtr runs for each job

  sum by(mtr_exporter, error)(mtr_runs_total{error!=""}) - provides the
  number of failed runs

  The diff between these two is the amount of successful runs.

  This implements GH#29, GH#30

• Add small information page on "/", linking to /metrics and the project page.

Bug Fixes:

• Fix -watch-jobs - allow launching with zero given jobs, they might be
  added later (fixes GH#22)
• Fix picking up job file in a container / pod scenario (GH#20, thanks Clavin)
• Remove -flag.deprecatedMetrics from "-h" output (fixes GH#26)
• Fix logic bug for job(s) from command line
• Fix printing the version when -version is given.

Maintenance:

• Improve documentation (thanks Guillaume)

Contributors:

• Guillaume Berche - https://github.com/gberche-orange
• Clavianus Juneardo - https://github.com/clavinjune

0.4.0

Features:

• add label 'hop' for the 'first', 'last' and
  'intermediate' hops towards the traced destination

Maintenance:

As announced with mtr-exporter 0.3.0, the following deprecated
metrics are removed:

deprecated
mtr_report_duration_ms_gauge
mtr_report_count_hubs_gauge
mtr_report_snt_gauge
mtr_report_loss_gauge
mtr_report_best_gauge
mtr_report_wrst_gauge
mtr_report_avg_gauge
mtr_report_last_gauge
mtr_report_stdev_gauge

Also, -flag.deprecatedMetrics is removed.

0.3.0

Features:

• allow tracing multiple mtr destinations at once
  via mtr jobs file

• improve robustness on parsing mtr JSON output
  (<=mtr:0.93, >=mtr:0.94 differ)

• add -flag.deprecatedMetrics to render deprecated metrics.
  helps with transition time until deprecated metrics are
  gone

Improvements:

• implemented Prometheus Best Practices, as a result, some metrics
  are renamed, the old names are marked deprecated.

deprecated	new
mtr_report_duration_ms_gauge	mtr_report_duration_seconds
mtr_report_count_hubs_gauge	mtr_report_count_hubs
mtr_report_snt_gauge	mtr_report_snt
mtr_report_loss_gauge	mtr_report_loss
mtr_report_best_gauge	mtr_report_best
mtr_report_wrst_gauge	mtr_report_wrst
mtr_report_avg_gauge	mtr_report_avg
mtr_report_last_gauge	mtr_report_last
mtr_report_stdev_gauge	mtr_report_stdev

Note: mtr_report_duration_seconds not only changed its name, but also
the unit: its seconds now, not milliseconds anymore.

With mtr-exporter:0.4.0 the deprecated ones will be removed.
Use -flag.deprecatedMetrics to have them exposed in the /metrics
response.

0.2.0

Features

• Add /health endpoint for quick & cheap checking healthiness
  (Thanks Jakub Sokołowski)

0.1.0

Initial Release