Add gitlab-health (fix #670)

CHANGELOG.md

@@ -31,6 +31,7 @@ Icinga Director:
 
 Monitoring Plugins:
 
+* gitlab-health (fix #670)
 * gitlab-version
 * openstack-nova-list
 * postgresql-version

assets/icingaweb2-module-director/all-the-rest.json

@@ -9086,6 +9086,49 @@
             "object_name": "GitLab Service Set",
             "object_type": "template",
             "services": {
+                "GitLab Health": {
+                    "action_url": null,
+                    "apply_for": null,
+                    "assign_filter": null,
+                    "check_command": null,
+                    "check_interval": null,
+                    "check_period": null,
+                    "check_timeout": null,
+                    "command_endpoint": null,
+                    "disabled": false,
+                    "display_name": null,
+                    "enable_active_checks": null,
+                    "enable_event_handler": null,
+                    "enable_flapping": null,
+                    "enable_notifications": null,
+                    "enable_passive_checks": null,
+                    "enable_perfdata": null,
+                    "event_command": null,
+                    "fields": [],
+                    "flapping_threshold_high": null,
+                    "flapping_threshold_low": null,
+                    "groups": [],
+                    "host": null,
+                    "icon_image": null,
+                    "icon_image_alt": null,
+                    "imports": [
+                        "tpl-service-gitlab-health"
+                    ],
+                    "max_check_attempts": null,
+                    "notes": null,
+                    "notes_url": null,
+                    "object_name": "GitLab Health",
+                    "object_type": "object",
+                    "retry_interval": null,
+                    "service_set": null,
+                    "template_choice": null,
+                    "use_agent": null,
+                    "use_var_overrides": null,
+                    "uuid": "161fb5d8-24ed-4bdf-b990-1818e9ab57c2",
+                    "vars": {},
+                    "volatile": null,
+                    "zone": null
+                },
                 "GitLab Version": {
                     "action_url": null,
                     "apply_for": null,

check-plugins/gitlab-health/README.rst

@@ -0,0 +1,86 @@
+Check example
+=============
+
+Overview
+--------
+
+Checks whether the GitLab application server is running. It does not hit the database or verifies other services are running. Its purpose is to notify that the application server is handling requests, but a STATE_OK response does not signify that the database or other services are ready.
+
+Hints:
+
+* To access monitoring resources, the requesting client IP needs to be included in the allowlist. For details, see `how to add IPs to the allowlist for the monitoring endpoints <https://docs.gitlab.com/ee/administration/monitoring/ip_allowlist.html>`.
+* GitLab Health Checks: https://docs.gitlab.com/ee/administration/monitoring/health_check.html
+
+
+Fact Sheet
+----------
+
+.. csv-table::
+    :widths: 30, 70
+
+    "Check Plugin Download",                "https://github.com/Linuxfabrik/monitoring-plugins/tree/main/check-plugins/gitlab-health"
+    "Check Interval Recommendation",        "Once a minute"
+    "Can be called without parameters",     "Yes"
+    "Compiled for",                         "Linux"
+
+
+Help
+----
+
+.. code-block:: text
+
+    usage: gitlab-health [-h] [-V] [--always-ok] [--severity {warn,crit}]
+                         [--test TEST] [--timeout TIMEOUT] [--url URL]
+
+    Checks whether the GitLab application server is running. It does not hit the
+    database or verifies other services are running.
+
+    options:
+      -h, --help            show this help message and exit
+      -V, --version         show program's version number and exit
+      --always-ok           Always returns OK.
+      --severity {warn,crit}
+                            Severity for alerting. Default: warn
+      --test TEST           For unit tests. Needs "path-to-stdout-file,path-to-
+                            stderr-file,expected-retc".
+      --timeout TIMEOUT     Network timeout in seconds. Default: 3 (seconds)
+      --url URL             GitLab health URL endpoint. Default:
+                            http://localhost/-/health
+
+
+Usage Examples
+--------------
+
+.. code-block:: bash
+
+    ./gitlab-health --severity warn --timeout 3 --url http://localhost/-/health
+
+Output:
+
+.. code-block:: text
+
+    The GitLab application server is processing requests, but this does not mean that the database or other services are ready.
+
+
+States
+------
+
+* Depending on the given ``--severity``, returns WARN (default) or CRIT if liveness and readiness probes to indicate service health and reachability to required services fail.
+
+
+Perfdata / Metrics
+------------------
+
+.. csv-table::
+    :widths: 25, 15, 60
+    :header-rows: 1
+
+    Name,                                       Type,               Description                                           
+    gitlab-health-state,                        Number,             "The current state (0 = OK, 1 = WARN, 2 = CRIT, 3 = UNKNOWN)."
+
+
+Credits, License
+----------------
+
+* Authors: `Linuxfabrik GmbH, Zurich <https://www.linuxfabrik.ch>`_
+* License: The Unlicense, see `LICENSE file <https://unlicense.org/>`_.

check-plugins/gitlab-health/gitlab-health

@@ -0,0 +1,124 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8; py-indent-offset: 4 -*-
+#
+# Author:  Linuxfabrik GmbH, Zurich, Switzerland
+# Contact: info (at) linuxfabrik (dot) ch
+#          https://www.linuxfabrik.ch/
+# License: The Unlicense, see LICENSE file.
+
+# https://github.com/Linuxfabrik/monitoring-plugins/blob/main/CONTRIBUTING.rst
+
+"""See the check's README for more details.
+"""
+
+import argparse  # pylint: disable=C0413
+import sys  # pylint: disable=C0413
+
+import lib.args  # pylint: disable=C0413
+import lib.base  # pylint: disable=C0413
+import lib.test  # pylint: disable=C0413
+import lib.url  # pylint: disable=C0413
+from lib.globals import (STATE_CRIT, STATE_OK,  # pylint: disable=C0413
+                          STATE_UNKNOWN, STATE_WARN)
+
+
+__author__ = 'Linuxfabrik GmbH, Zurich/Switzerland'
+__version__ = '2023082401'
+
+DESCRIPTION = """Checks whether the GitLab application server is running. It does not hit
+                the database or verifies other services are running."""
+
+DEFAULT_SEVERITY = 'warn'
+DEFAULT_TIMEOUT = 3
+DEFAULT_URL = 'http://localhost/-/health'
+
+
+def parse_args():
+    """Parse command line arguments using argparse.
+    """
+    parser = argparse.ArgumentParser(description=DESCRIPTION)
+
+    parser.add_argument(
+        '-V', '--version',
+        action='version',
+        version='%(prog)s: v{} by {}'.format(__version__, __author__)
+    )
+
+    parser.add_argument(
+        '--always-ok',
+        help='Always returns OK.',
+        dest='ALWAYS_OK',
+        action='store_true',
+        default=False,
+    )
+
+    parser.add_argument(
+        '--severity',
+        help='Severity for alerting. Default: %(default)s',
+        dest='SEVERITY',
+        default=DEFAULT_SEVERITY,
+        choices=['warn', 'crit'],
+    )
+
+    parser.add_argument(
+        '--test',
+        help='For unit tests. Needs "path-to-stdout-file,path-to-stderr-file,expected-retc".',
+        dest='TEST',
+        type=lib.args.csv,
+    )
+
+    parser.add_argument(
+        '--timeout',
+        help='Network timeout in seconds. Default: %(default)s (seconds)',
+        dest='TIMEOUT',
+        type=int,
+        default=DEFAULT_TIMEOUT,
+    )
+
+    parser.add_argument(
+        '--url',
+        help='GitLab health URL endpoint. Default: %(default)s',
+        dest='URL',
+        default=DEFAULT_URL,
+    )
+
+    return parser.parse_args()
+
+
+def main():
+    """The main function. Hier spielt die Musik.
+    """
+
+    # parse the command line, exit with UNKNOWN if it fails
+    try:
+        args = parse_args()
+    except SystemExit:
+        sys.exit(STATE_UNKNOWN)
+
+    # init some vars
+    state = STATE_OK
+
+    # fetch and analyze data
+    if args.TEST is None:
+        result = lib.base.coe(lib.url.fetch(args.URL, timeout=args.TIMEOUT))
+    else:
+        # do not call the command, put in test data
+        result, stderr, retc = lib.test.test(args.TEST)
+
+    if result == 'GitLab OK':
+        msg = 'The GitLab application server is processing requests, but this does not mean ' \
+              'that the database or other services are ready.'
+    else:
+        msg = 'The GitLab application server seems to have a problem.'
+        state = lib.base.str2state(args.SEVERITY)
+    perfdata = lib.base.get_perfdata('gitlab-health', state, None, None, None, 0, STATE_UNKNOWN)
+
+    # over and out
+    lib.base.oao(msg, state, perfdata, always_ok=args.ALWAYS_OK)
+
+
+if __name__ == '__main__':
+    try:
+        main()
+    except Exception:   # pylint: disable=W0703
+        lib.base.cu()

check-plugins/gitlab-health/grafana/gitlab-health.yml

@@ -0,0 +1,115 @@
+apiVersion: grizzly.grafana.com/v1alpha1
+kind: Dashboard
+metadata:
+  folder: linuxfabrik-monitoring-plugins
+  name: gitlab-health
+spec:
+  schemaVersion: 2023041201
+  tags:
+    - Linuxfabrik
+    - Grizzly
+    - static
+  time:
+    from: now-90d
+    to: now
+  timepicker:
+    hidden: false
+    refresh_intervals:
+    - 1m
+  timezone: browser
+  title: GitLab Health
+  uid: linuxfabrik-monitoring-plugins-gitlab-health
+  editable: true
+  liveNow: true
+  refresh: 1m
+  templating:
+    list:
+    - hide: 2
+      label: Command
+      name: command
+      query: cmd-check-gitlab-health
+      type: constant
+    - label: Hostname
+      name: hostname
+      query: SHOW TAG VALUES FROM "cmd-check-gitlab-health" WITH KEY = "hostname"
+      refresh: 2
+      sort: 1
+      type: query
+
+  panels:
+
+  - title: GitLab Health
+    type: timeseries
+    gridPos:
+      h: 8
+      w: 12
+      x: 12
+      y: 8
+    fieldConfig:
+      defaults:
+        color:
+          mode: palette-classic
+        custom:
+          lineInterpolation: smooth
+          spanNulls: true
+        decimals: 0
+        max: 3
+        min: 0
+        unit: short
+      overrides:
+      - matcher:
+          id: byName
+          options: gitlab-health
+        properties:
+        - id: mappings
+          value:
+          - options:
+              '0':
+                text: OK
+              '1':
+                text: WARN
+              '2':
+                text: CRIT
+              '3':
+                text: UNKN
+            type: value
+    options:
+      legend:
+        calcs:
+        - min
+        - max
+        displayMode: table
+        placement: bottom
+        showLegend: true
+      tooltip:
+        mode: multi
+        sort: none
+
+    targets:
+
+    - alias: gitlab-health
+      refId: gitlab-health
+      groupBy:
+      - params:
+        - $interval
+        type: time
+      measurement: /^$command$/
+      resultFormat: time_series
+      select:
+      - - params:
+          - value
+          type: field
+        - params: []
+          type: mean
+      tags:
+      - key: hostname
+        operator: '=~'
+        value: /^$hostname$/
+      - condition: AND
+        key: service
+        operator: '='
+        value: GitLab Health
+      - condition: AND
+        key: metric
+        operator: '='
+        value: gitlab-health