daemon: expose prometheus in new global.monitoring config section + document it

refs #67

cmd/config_job_prometheus.go

@@ -12,6 +12,7 @@ import (
 )
 
 type PrometheusJob struct {
+	Name   string
 	Listen string
 }
 
@@ -55,12 +56,10 @@ func parsePrometheusJob(c JobParsingContext, name string, i map[string]interface
 	if s.Listen == "" {
 		return nil, errors.New("must specify 'listen' attribute")
 	}
-	return &PrometheusJob{s.Listen}, nil
+	return &PrometheusJob{name, s.Listen}, nil
 }
 
-func (*PrometheusJob) JobName() string {
+func (j *PrometheusJob) JobName() string { return j.Name }
-	return "prometheus"
-}
 
 func (j *PrometheusJob) JobType() JobType { return JobTypePrometheus }
 

cmd/config_parse.go

@@ -93,9 +93,30 @@ func parseConfig(i interface{}) (c *Config, err error) {
 
 	cpc := ConfigParsingContext{&c.Global}
 	jpc := JobParsingContext{cpc}
-
-	// Jobs
 	c.Jobs = make(map[string]Job, len(asMap.Jobs))
+
+	// FIXME internal jobs should not be mixed with user jobs
+	// Monitoring Jobs
+	var monJobs []map[string]interface{}
+	if err := mapstructure.Decode(asMap.Global["monitoring"], &monJobs); err != nil {
+		return nil, errors.Wrap(err, "cannot parse monitoring section")
+	}
+	for i, jc := range monJobs {
+		if jc["name"] == "" || jc["name"] == nil {
+			// FIXME internal jobs should not require a name...
+			jc["name"] = fmt.Sprintf("prometheus-%d", i)
+		}
+		job, err := parseJob(jpc, jc)
+		if err != nil {
+			return nil, errors.Wrapf(err, "cannot parse monitoring job #%d", i)
+		}
+		if job.JobType() != JobTypePrometheus {
+			return nil, errors.Errorf("monitoring job #%d has invalid job type", i)
+		}
+		c.Jobs[job.JobName()] = job
+	}
+
+	// Regular Jobs
 	for i := range asMap.Jobs {
 		job, err := parseJob(jpc, asMap.Jobs[i])
 		if err != nil {
@@ -109,7 +130,7 @@ func parseConfig(i interface{}) (c *Config, err error) {
 		}
 		jn := job.JobName()
 		if _, ok := c.Jobs[jn]; ok {
-			err = errors.Errorf("duplicate job name: %s", jn)
+			err = errors.Errorf("duplicate or invalid job name: %s", jn)
 			return nil, err
 		}
 		c.Jobs[job.JobName()] = job

cmd/config_test.go

@@ -17,7 +17,7 @@ func TestSampleConfigsAreParsedWithoutErrors(t *testing.T) {
 		"./sampleconf/pullbackup/backuphost.yml",
 		"./sampleconf/pullbackup/productionhost.yml",
 		"./sampleconf/random/debugging.yml",
-		"./sampleconf/random/logging.yml",
+		"./sampleconf/random/logging_and_monitoring.yml",
 	}
 
 	for _, p := range paths {

cmd/sampleconf/random/logging_and_monitoring.yml

@@ -20,5 +20,9 @@ global:
       level: debug
       format: logfmt
 
+  monitoring:
+
+    - type: prometheus
+      listen: ':9090'
 
 jobs: []

docs/changelog.rst

@@ -10,6 +10,11 @@ Changelog
 The changelog summarized bugfixes that are deemed relevant for users.
 Developers should consult the git commit log or GitHub issue tracker.
 
+0.0.4 (unreleased)
+------------------
+
+* |feature| :issue:`67`: Expose `Prometheus <https://prometheus.io>`_ metrics via HTTP (:ref:`config docs <monitoring-prometheus>`)
+
 0.0.3
 -----
 

docs/configuration.rst

@@ -13,4 +13,5 @@ Configuration
     configuration/map_filter_syntax
     configuration/prune
     configuration/logging
+    configuration/monitoring
     configuration/misc

docs/configuration/logging.rst

@@ -8,7 +8,7 @@ Logging
 zrepl uses structured logging to provide users with easily processable log messages.
 
 Logging outlets are configured in the ``global`` section of the |mainconfig|.
-Check out :sampleconf:`random/logging.yml` for an example on how to configure multiple outlets:
+Check out :sampleconf:`random/logging_and_monitoring.yml` for an example on how to configure multiple outlets:
 
 ::
 

docs/configuration/monitoring.rst

@@ -0,0 +1,30 @@
+.. include:: ../global.rst.inc
+
+.. _monitoring:
+
+Monitoring
+==========
+
+Monitoring endpoints are configured in the ``global.monitoring`` section of the |mainconfig|.
+Check out :sampleconf:`random/logging_and_monitoring.yml` for examples.
+
+.. _monitoring-prometheus:
+
+Prometheus
+----------
+
+zrepl can expose `Prometheus metrics <https://prometheus.io/docs/instrumenting/exposition_formats/>`_ via HTTP.
+The ``listen`` attribute is a `net.Listen <https://golang.org/pkg/net/#Listen>`_  string for tcp, e.g. ``:9091`` or ``127.0.0.1:9091``.
+
+The Prometheues monitoring job appears in the ``zrepl control`` job list and may be specified **at most once**.
+There is no stability guarantee on the exported metrics.
+
+::
+
+    global:
+      monitoring:
+        - type: prometheus
+          listen: ':9091'
+
+
+

docs/tutorial.rst

@@ -193,6 +193,7 @@ Summary
 Congratulations, you have a working pull backup. Where to go next?
 
 * Read more about :ref:`configuration format, options & job types <configuration_toc>`
+* Configure :ref:`logging <logging>` \& :ref:`monitoring <monitoring>`.
 * Learn about :ref:`implementation details <implementation_toc>` of zrepl.
 
 

docs/usage.rst

@@ -37,7 +37,7 @@ zrepl daemon
 ============
 
 All actual work zrepl does is performed by a daemon process.
-Logging is configurable via the config file. Please refer to the :ref:`logging documention <logging>`.
+The daemon supports structured :ref:`logging <logging>` and provides :ref:`monitoring endpoints <monitoring>`.
 
 When installating from a package, the package maintainer should have provided an init script / systemd.service file.
 You should thus be able to start zrepl daemon using your init system.