mgr/dashboard: monitoring: replace Grafana JSON with Grafonnet based Code

This PR intends to add grafonnet to generate grafana JSON files

Fixes: https://tracker.ceph.com/issues/45184
Signed-off-by: Aashish Sharma <aasharma@redhat.com>

diff --git a/ceph.spec.in b/ceph.spec.in
index 4d4082394d469d86a123d9a51eccd0d02eb56d1c..8843aefe6689da5b2e821b08085f03b052277a5b 100644 (file)
--- a/ceph.spec.in
+++ b/ceph.spec.in
@@ -353,6 +353,7 @@ BuildRequires:      lz4-devel >= 1.7
 %if 0%{with make_check}
 %if 0%{?fedora} || 0%{?rhel}
 BuildRequires: golang-github-prometheus
+BuildRequires: golang-github-google-jsonnet
 BuildRequires: libtool-ltdl-devel
 BuildRequires: ninja-build
 BuildRequires: xmlsec1
@@ -371,6 +372,7 @@ BuildRequires:      python%{python3_pkgversion}-pyOpenSSL
 %endif
 %if 0%{?suse_version}
 BuildRequires: golang-github-prometheus-prometheus
+BuildRequires: jsonnet
 BuildRequires: libxmlsec1-1
 BuildRequires: libxmlsec1-nss1
 BuildRequires: libxmlsec1-openssl1

diff --git a/debian/control b/debian/control
index a3b6d524e41e8444d0056716d36e97d07d256a07..14d479c6c72abd1fb03950eeb1c0b6309957376b 100644 (file)
--- a/debian/control
+++ b/debian/control
@@ -24,6 +24,7 @@ Build-Depends: automake,
                g++ (>= 7),
                javahelper,
                jq <pkg.ceph.check>,
+               jsonnet <pkg.ceph.check>,
                junit4,
                libaio-dev,
                libbabeltrace-ctf-dev,

diff --git a/doc/dev/developer_guide/dash-devel.rst b/doc/dev/developer_guide/dash-devel.rst
index 5c616ade37c9450a2feaee34c4a98992a363bd41..1f8744bec1fb23033e13bdf5ddb26abd61ea5f07 100644 (file)
--- a/doc/dev/developer_guide/dash-devel.rst
+++ b/doc/dev/developer_guide/dash-devel.rst
@@ -1652,6 +1652,58 @@ load the controllers that we want to test. In the above example we are only
 loading the ``Ping`` controller. We can also disable authentication of a
 controller at this stage, as depicted in the example.
 
+How to update or create new dashboards in grafana?
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+We are using ``jsonnet`` and ``grafonnet-lib`` to write code for the grafana dashboards.
+All the dashboards are written inside ``grafana_dashboards.jsonnet`` file in the
+monitoring/grafana/dashboards/jsonnet directory.
+
+We generate the dashboard json files directly from this jsonnet file by running this
+command in the grafana/dashboards directory:
+``jsonnet -m . jsonnet/grafana_dashboards.jsonnet``.
+(For the above command to succeed we need ``jsonnet`` package installed and ``grafonnet-lib``
+directory cloned in our machine. Please refer - 
+``https://grafana.github.io/grafonnet-lib/getting-started/`` in case you have some trouble.)
+
+To update an existing grafana dashboard or to create a new one, we need to update
+the ``grafana_dashboards.jsonnet`` file and generate the new/updated json files using the
+above mentioned command. For people who are not familiar with grafonnet or jsonnet implementation
+can follow this doc - ``https://grafana.github.io/grafonnet-lib/``.
+
+Example grafana dashboard in jsonnet format:
+
+To specify the grafana dashboard properties such as title, uid etc we can create a local function -
+
+::
+
+    local dashboardSchema(title, uid, time_from, refresh, schemaVersion, tags,timezone, timepicker)
+
+To add a graph panel we can spcify the graph schema in a local function such as -
+
+::
+
+    local graphPanelSchema(title, nullPointMode, stack, formatY1, formatY2, labelY1, labelY2, min, fill, datasource)
+
+and then use these functions inside the dashboard definition like -
+
+::
+
+    {
+        radosgw-sync-overview.json: //json file name to be generated
+
+        dashboardSchema(
+          'RGW Sync Overview', 'rgw-sync-overview', 'now-1h', '15s', .., .., ..
+        )
+
+        .addPanels([
+          graphPanelSchema(
+            'Replication (throughput) from Source Zone', 'Bps', null, .., .., ..)
+        ])
+    }
+
+The valid grafonnet-lib attributes can be found here - ``https://grafana.github.io/grafonnet-lib/api-docs/``.
+  
 
 How to listen for manager notifications in a controller?
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

diff --git a/monitoring/grafana/dashboards/CMakeLists.txt b/monitoring/grafana/dashboards/CMakeLists.txt
index 41ca947a6dc8d150f007f82920eaf83712de3aea..49386d951abf1bf23cdedcfc0046348c180c49b1 100644 (file)
--- a/monitoring/grafana/dashboards/CMakeLists.txt
+++ b/monitoring/grafana/dashboards/CMakeLists.txt
@@ -1,8 +1,26 @@
-set(CEPH_GRAFANA_DASHBOARDS_DIR "${CMAKE_INSTALL_SYSCONFDIR}/grafana/dashboards/ceph-dashboard"
-  CACHE PATH "Location for grafana dashboards")
+set(CEPH_BUILD_VIRTUALENV $ENV{TMPDIR})
+if(NOT CEPH_BUILD_VIRTUALENV)
+  set(CEPH_BUILD_VIRTUALENV ${CMAKE_BINARY_DIR})
+endif()
 
-FILE(GLOB CEPH_GRAFANA_DASHBOARDS "*.json")
-
-install(FILES
-  ${CEPH_GRAFANA_DASHBOARDS}
-  DESTINATION ${CEPH_GRAFANA_DASHBOARDS_DIR})
+if(WITH_GRAFANA)
+  include(AddCephTest)
+  add_tox_test(grafana TOX_ENVS grafonnet-check)
+  set(ver 0.1.0)
+  set(name grafonnet-lib)
+  include(ExternalProject)
+  ExternalProject_Add(${name}
+    URL https://github.com/grafana/${name}/archive/v${ver}/${name}-${ver}.tar.gz
+    URL_MD5 0798752ed40864fa8b3db40a3c970642
+    BUILD_COMMAND ""
+    CONFIGURE_COMMAND ""
+    INSTALL_COMMAND "")
+  add_dependencies(tests
+    ${name})
+  ExternalProject_Get_Property(${name} SOURCE_DIR)
+  set_property(
+    TEST run-tox-grafana
+    APPEND
+    PROPERTY ENVIRONMENT
+    GRAFONNET_PATH=${SOURCE_DIR}/grafonnet)
+endif()

diff --git a/monitoring/grafana/dashboards/jsonnet/grafana_dashboards.jsonnet b/monitoring/grafana/dashboards/jsonnet/grafana_dashboards.jsonnet
new file mode 100644 (file)
index 0000000..11c89b2
--- /dev/null
+++ b/monitoring/grafana/dashboards/jsonnet/grafana_dashboards.jsonnet
@@ -0,0 +1,54 @@
+local g = import 'grafana.libsonnet';
+
+local dashboardSchema(title, uid, time_from, refresh, schemaVersion, tags,timezone, timepicker) =
+  g.dashboard.new(title=title, uid=uid, time_from=time_from, refresh=refresh, schemaVersion=schemaVersion, tags=tags, timezone=timezone, timepicker=timepicker);
+
+local graphPanelSchema(title, nullPointMode, stack, formatY1, formatY2, labelY1, labelY2, min, fill, datasource) =
+  g.graphPanel.new(title=title, nullPointMode=nullPointMode, stack=stack, formatY1=formatY1, formatY2=formatY2, labelY1=labelY1, labelY2=labelY2, min=min, fill=fill, datasource=datasource);
+
+local addTargetSchema(expr, intervalFactor, format, legendFormat) =
+  g.prometheus.target(expr=expr, intervalFactor=intervalFactor, format=format, legendFormat=legendFormat);
+
+local addTemplateSchema(name, datasource, query, refresh, hide, includeAll, sort) =
+  g.template.new(name=name, datasource=datasource, query=query, refresh=refresh, hide=hide, includeAll=includeAll, sort=sort);
+
+local addAnnotationSchema(builtIn, datasource, enable, hide, iconColor, name, type) =
+  g.annotation.datasource(builtIn=builtIn, datasource=datasource, enable=enable, hide=hide, iconColor=iconColor, name=name, type=type);
+
+{
+  "radosgw-sync-overview.json":
+    local RgwSyncOverviewPanel(title, formatY1, labelY1, rgwMetric, x, y, w, h) =
+      graphPanelSchema(title, 'null as zero', true, formatY1, 'short', labelY1, null, 0, 1, '$datasource')
+      .addTargets(
+        [addTargetSchema('sum by (source_zone) (rate(%s[30s]))' % rgwMetric, 1, 'time_series', '{{source_zone}}')]) + {gridPos: {x: x, y: y, w: w, h: h}};
+
+    dashboardSchema(
+      'RGW Sync Overview', 'rgw-sync-overview', 'now-1h', '15s', 16, ["overview"], '', {refresh_intervals:['5s','10s','15s','30s','1m','5m','15m','30m','1h','2h','1d'],time_options:['5m','15m','1h','6h','12h','24h','2d','7d','30d']}
+    )
+    .addAnnotation(
+      addAnnotationSchema(
+        1, '-- Grafana --', true, true, 'rgba(0, 211, 255, 1)', 'Annotations & Alerts', 'dashboard')
+    )
+    .addRequired(
+       type='grafana', id='grafana', name='Grafana', version='5.0.0'
+    )
+    .addRequired(
+       type='panel', id='graph', name='Graph', version='5.0.0'
+    )
+    .addTemplate(
+       addTemplateSchema('rgw_servers', '$datasource', 'prometehus', 1, 2, true, 1)
+    )
+    .addTemplate(
+       g.template.datasource('datasource', 'prometheus', 'default', label='Data Source')
+    )
+    .addPanels([
+      RgwSyncOverviewPanel(
+        'Replication (throughput) from Source Zone', 'Bps', null, 'ceph_data_sync_from_zone_fetch_bytes_sum', 0, 0, 8, 7),
+      RgwSyncOverviewPanel(
+        'Replication (objects) from Source Zone', 'short', 'Objects/s', 'ceph_data_sync_from_zone_fetch_bytes_count', 8, 0, 8, 7),
+      RgwSyncOverviewPanel(
+        'Polling Request Latency from Source Zone', 'ms', null, 'ceph_data_sync_from_zone_poll_latency_sum', 16, 0, 8, 7),
+      RgwSyncOverviewPanel(
+        'Unsuccessful Object Replications from Source Zone', 'short', 'Count/s', 'ceph_data_sync_from_zone_fetch_errors', 0, 7, 8, 7)
+    ])
+}

diff --git a/monitoring/grafana/dashboards/radosgw-sync-overview.json b/monitoring/grafana/dashboards/radosgw-sync-overview.json
index e9136d78e3b7cb18b4deaad4e0a9a8dc2f5b67a6..d6cd5e98a5e3c1e9771a455f6924c23326225761 100644 (file)
--- a/monitoring/grafana/dashboards/radosgw-sync-overview.json
+++ b/monitoring/grafana/dashboards/radosgw-sync-overview.json
@@ -1,440 +1,455 @@
 {
-  "__requires": [
-    {
-      "type": "grafana",
-      "id": "grafana",
-      "name": "Grafana",
-      "version": "5.0.0"
-    },
-    {
-      "type": "panel",
-      "id": "graph",
-      "name": "Graph",
-      "version": "5.0.0"
-    }
-  ],
-  "annotations": {
-    "list": [
+   "__inputs": [ ],
+   "__requires": [
       {
-        "builtIn": 1,
-        "datasource": "-- Grafana --",
-        "enable": true,
-        "hide": true,
-        "iconColor": "rgba(0, 211, 255, 1)",
-        "name": "Annotations & Alerts",
-        "type": "dashboard"
-      }
-    ]
-  },
-  "editable": false,
-  "gnetId": null,
-  "graphTooltip": 0,
-  "id": null,
-  "iteration": 1534386107523,
-  "links": [],
-  "panels": [
-    {
-      "aliasColors": {},
-      "bars": false,
-      "dashLength": 10,
-      "dashes": false,
-      "datasource": "$datasource",
-      "fill": 1,
-      "gridPos": {
-        "h": 7,
-        "w": 8,
-        "x": 0,
-        "y": 0
-      },
-      "id": 1,
-      "legend": {
-        "avg": false,
-        "current": false,
-        "max": false,
-        "min": false,
-        "show": true,
-        "total": false,
-        "values": false
-      },
-      "lines": true,
-      "linewidth": 1,
-      "links": [],
-      "nullPointMode": "null as zero",
-      "percentage": false,
-      "pointradius": 5,
-      "points": false,
-      "renderer": "flot",
-      "seriesOverrides": [],
-      "spaceLength": 10,
-      "stack": true,
-      "steppedLine": false,
-      "targets": [
-        {
-          "expr": "sum by (source_zone) (rate(ceph_data_sync_from_zone_fetch_bytes_sum[30s]))",
-          "format": "time_series",
-          "intervalFactor": 1,
-          "legendFormat": "{{source_zone}}",
-          "refId": "A"
-        }
-      ],
-      "thresholds": [],
-      "timeFrom": null,
-      "timeShift": null,
-      "title": "Replication (throughput) from Source Zone",
-      "tooltip": {
-        "shared": true,
-        "sort": 0,
-        "value_type": "individual"
-      },
-      "type": "graph",
-      "xaxis": {
-        "buckets": null,
-        "mode": "time",
-        "name": null,
-        "show": true,
-        "values": []
-      },
-      "yaxes": [
-        {
-          "unit": "bytes",
-          "format": "Bps",
-          "decimals": null,
-          "logBase": 1,
-          "max": null,
-          "min": "0",
-          "show": true
-        },
-        {
-          "format": "short",
-          "label": null,
-          "logBase": 1,
-          "max": null,
-          "min": null,
-          "show": false
-        }
-      ]
-    },
-    {
-      "aliasColors": {},
-      "bars": false,
-      "dashLength": 10,
-      "dashes": false,
-      "datasource": "$datasource",
-      "fill": 1,
-      "gridPos": {
-        "h": 7,
-        "w": 7.4,
-        "x": 8.3,
-        "y": 0
-      },
-      "id": 2,
-      "legend": {
-        "avg": false,
-        "current": false,
-        "max": false,
-        "min": false,
-        "show": true,
-        "total": false,
-        "values": false
-      },
-      "lines": true,
-      "linewidth": 1,
-      "links": [],
-      "nullPointMode": "null as zero",
-      "percentage": false,
-      "pointradius": 5,
-      "points": false,
-      "renderer": "flot",
-      "seriesOverrides": [],
-      "spaceLength": 10,
-      "stack": true,
-      "steppedLine": false,
-      "targets": [
-        {
-          "expr": "sum by (source_zone) (rate(ceph_data_sync_from_zone_fetch_bytes_count[30s]))",
-          "format": "time_series",
-          "intervalFactor": 1,
-          "legendFormat": "{{source_zone}}",
-          "refId": "A"
-        }
-      ],
-      "thresholds": [],
-      "timeFrom": null,
-      "timeShift": null,
-      "title": "Replication (objects) from Source Zone",
-      "tooltip": {
-        "shared": true,
-        "sort": 0,
-        "value_type": "individual"
-      },
-      "type": "graph",
-      "xaxis": {
-        "buckets": null,
-        "mode": "time",
-        "name": null,
-        "show": true,
-        "values": []
-      },
-      "yaxes": [
-        {
-          "format": "short",
-          "decimals": null,
-          "label": "Objects/s",
-          "logBase": 1,
-          "max": null,
-          "min": "0",
-          "show": true
-        },
-        {
-          "format": "short",
-          "label": null,
-          "logBase": 1,
-          "max": null,
-          "min": null,
-          "show": false
-        }
-      ]
-    },
-    {
-      "aliasColors": {},
-      "bars": false,
-      "dashLength": 10,
-      "dashes": false,
-      "datasource": "$datasource",
-      "fill": 1,
-      "gridPos": {
-        "h": 7,
-        "w": 8,
-        "x": 16,
-        "y": 0
-      },
-      "id": 3,
-      "legend": {
-        "avg": false,
-        "current": false,
-        "max": false,
-        "min": false,
-        "show": true,
-        "total": false,
-        "values": false
+         "id": "grafana",
+         "name": "Grafana",
+         "type": "grafana",
+         "version": "5.0.0"
       },
-      "lines": true,
-      "linewidth": 1,
-      "links": [],
-      "nullPointMode": "null as zero",
-      "percentage": false,
-      "pointradius": 5,
-      "points": false,
-      "renderer": "flot",
-      "seriesOverrides": [],
-      "spaceLength": 10,
-      "stack": true,
-      "steppedLine": false,
-      "targets": [
-        {
-          "expr": "sum by (source_zone) (rate(ceph_data_sync_from_zone_poll_latency_sum[30s]) * 1000)",
-          "format": "time_series",
-          "intervalFactor": 1,
-          "legendFormat": "{{source_zone}}",
-          "refId": "A"
-        }
-      ],
-      "thresholds": [],
-      "timeFrom": null,
-      "timeShift": null,
-      "title": "Polling Request Latency from Source Zone",
-      "tooltip": {
-        "shared": true,
-        "sort": 0,
-        "value_type": "individual"
-      },
-      "type": "graph",
-      "xaxis": {
-        "buckets": null,
-        "mode": "time",
-        "name": null,
-        "show": true,
-        "values": []
-      },
-      "yaxes": [
-        {
-          "unit": "s",
-          "format": "ms",
-          "decimals": null,
-          "logBase": 1,
-          "max": null,
-          "min": "0",
-          "show": true
-        },
-        {
-          "format": "short",
-          "label": null,
-          "logBase": 1,
-          "max": null,
-          "min": null,
-          "show": false
-        }
+      {
+         "id": "graph",
+         "name": "Graph",
+         "type": "panel",
+         "version": "5.0.0"
+      }
+   ],
+   "annotations": {
+      "list": [
+         {
+            "builtIn": 1,
+            "datasource": "-- Grafana --",
+            "enable": true,
+            "hide": true,
+            "iconColor": "rgba(0, 211, 255, 1)",
+            "name": "Annotations & Alerts",
+            "showIn": 0,
+            "tags": [ ],
+            "type": "dashboard"
+         }
       ]
-    },
-    {
-      "aliasColors": {},
-      "bars": false,
-      "dashLength": 10,
-      "dashes": false,
-      "datasource": "$datasource",
-      "fill": 1,
-      "gridPos": {
-        "h": 7,
-        "w": 8,
-        "x": 0,
-        "y": 7
-      },
-      "id": 4,
-      "legend": {
-        "avg": false,
-        "current": false,
-        "max": false,
-        "min": false,
-        "show": true,
-        "total": false,
-        "values": false
-      },
-      "lines": true,
-      "linewidth": 1,
-      "links": [],
-      "nullPointMode": "null as zero",
-      "percentage": false,
-      "pointradius": 5,
-      "points": false,
-      "renderer": "flot",
-      "seriesOverrides": [],
-      "spaceLength": 10,
-      "stack": true,
-      "steppedLine": false,
-      "targets": [
-        {
-          "expr": "sum by (source_zone) (rate(ceph_data_sync_from_zone_fetch_errors[30s]))",
-          "format": "time_series",
-          "intervalFactor": 1,
-          "legendFormat": "{{source_zone}}",
-          "refId": "A"
-        }
-      ],
-      "thresholds": [],
-      "timeFrom": null,
-      "timeShift": null,
-      "title": "Unsuccessful Object Replications from Source Zone",
-      "tooltip": {
-        "shared": true,
-        "sort": 0,
-        "value_type": "individual"
+   },
+   "editable": false,
+   "gnetId": null,
+   "graphTooltip": 0,
+   "hideControls": false,
+   "id": null,
+   "links": [ ],
+   "panels": [
+      {
+         "aliasColors": { },
+         "bars": false,
+         "dashLength": 10,
+         "dashes": false,
+         "datasource": "$datasource",
+         "fill": 1,
+         "gridPos": {
+            "h": 7,
+            "w": 8,
+            "x": 0,
+            "y": 0
+         },
+         "id": 2,
+         "legend": {
+            "alignAsTable": false,
+            "avg": false,
+            "current": false,
+            "max": false,
+            "min": false,
+            "rightSide": false,
+            "show": true,
+            "sideWidth": null,
+            "total": false,
+            "values": false
+         },
+         "lines": true,
+         "linewidth": 1,
+         "links": [ ],
+         "nullPointMode": "null as zero",
+         "percentage": false,
+         "pointradius": 5,
+         "points": false,
+         "renderer": "flot",
+         "repeat": null,
+         "seriesOverrides": [ ],
+         "spaceLength": 10,
+         "stack": true,
+         "steppedLine": false,
+         "targets": [
+            {
+               "expr": "sum by (source_zone) (rate(ceph_data_sync_from_zone_fetch_bytes_sum[30s]))",
+               "format": "time_series",
+               "intervalFactor": 1,
+               "legendFormat": "{{source_zone}}",
+               "refId": "A"
+            }
+         ],
+         "thresholds": [ ],
+         "timeFrom": null,
+         "timeShift": null,
+         "title": "Replication (throughput) from Source Zone",
+         "tooltip": {
+            "shared": true,
+            "sort": 0,
+            "value_type": "individual"
+         },
+         "type": "graph",
+         "xaxis": {
+            "buckets": null,
+            "mode": "time",
+            "name": null,
+            "show": true,
+            "values": [ ]
+         },
+         "yaxes": [
+            {
+               "format": "Bps",
+               "label": null,
+               "logBase": 1,
+               "max": null,
+               "min": 0,
+               "show": true
+            },
+            {
+               "format": "short",
+               "label": null,
+               "logBase": 1,
+               "max": null,
+               "min": 0,
+               "show": true
+            }
+         ]
       },
-      "type": "graph",
-      "xaxis": {
-        "buckets": null,
-        "mode": "time",
-        "name": null,
-        "show": true,
-        "values": []
+      {
+         "aliasColors": { },
+         "bars": false,
+         "dashLength": 10,
+         "dashes": false,
+         "datasource": "$datasource",
+         "fill": 1,
+         "gridPos": {
+            "h": 7,
+            "w": 8,
+            "x": 8,
+            "y": 0
+         },
+         "id": 3,
+         "legend": {
+            "alignAsTable": false,
+            "avg": false,
+            "current": false,
+            "max": false,
+            "min": false,
+            "rightSide": false,
+            "show": true,
+            "sideWidth": null,
+            "total": false,
+            "values": false
+         },
+         "lines": true,
+         "linewidth": 1,
+         "links": [ ],
+         "nullPointMode": "null as zero",
+         "percentage": false,
+         "pointradius": 5,
+         "points": false,
+         "renderer": "flot",
+         "repeat": null,
+         "seriesOverrides": [ ],
+         "spaceLength": 10,
+         "stack": true,
+         "steppedLine": false,
+         "targets": [
+            {
+               "expr": "sum by (source_zone) (rate(ceph_data_sync_from_zone_fetch_bytes_count[30s]))",
+               "format": "time_series",
+               "intervalFactor": 1,
+               "legendFormat": "{{source_zone}}",
+               "refId": "A"
+            }
+         ],
+         "thresholds": [ ],
+         "timeFrom": null,
+         "timeShift": null,
+         "title": "Replication (objects) from Source Zone",
+         "tooltip": {
+            "shared": true,
+            "sort": 0,
+            "value_type": "individual"
+         },
+         "type": "graph",
+         "xaxis": {
+            "buckets": null,
+            "mode": "time",
+            "name": null,
+            "show": true,
+            "values": [ ]
+         },
+         "yaxes": [
+            {
+               "format": "short",
+               "label": "Objects/s",
+               "logBase": 1,
+               "max": null,
+               "min": 0,
+               "show": true
+            },
+            {
+               "format": "short",
+               "label": null,
+               "logBase": 1,
+               "max": null,
+               "min": 0,
+               "show": true
+            }
+         ]
       },
-      "yaxes": [
-        {
-          "format": "short",
-          "decimals": null,
-          "label": "Count/s",
-          "logBase": 1,
-          "max": null,
-          "min": "0",
-          "show": true
-        },
-        {
-          "format": "short",
-          "label": null,
-          "logBase": 1,
-          "max": null,
-          "min": null,
-          "show": false
-        }
-      ]
-    }
-  ],
-  "refresh": "15s",
-  "schemaVersion": 16,
-  "style": "dark",
-  "tags": [
-    "overview"
-  ],
-  "templating": {
-    "list": [
       {
-        "allValue": null,
-        "current": {},
-        "datasource": "$datasource",
-        "hide": 2,
-        "includeAll": true,
-        "label": null,
-        "multi": false,
-        "name": "rgw_servers",
-        "options": [],
-        "query": "prometheus",
-        "refresh": 1,
-        "regex": "",
-        "sort": 1,
-        "tagValuesQuery": "",
-        "tags": [],
-        "tagsQuery": "",
-        "type": "query",
-        "useTags": false
+         "aliasColors": { },
+         "bars": false,
+         "dashLength": 10,
+         "dashes": false,
+         "datasource": "$datasource",
+         "fill": 1,
+         "gridPos": {
+            "h": 7,
+            "w": 8,
+            "x": 16,
+            "y": 0
+         },
+         "id": 4,
+         "legend": {
+            "alignAsTable": false,
+            "avg": false,
+            "current": false,
+            "max": false,
+            "min": false,
+            "rightSide": false,
+            "show": true,
+            "sideWidth": null,
+            "total": false,
+            "values": false
+         },
+         "lines": true,
+         "linewidth": 1,
+         "links": [ ],
+         "nullPointMode": "null as zero",
+         "percentage": false,
+         "pointradius": 5,
+         "points": false,
+         "renderer": "flot",
+         "repeat": null,
+         "seriesOverrides": [ ],
+         "spaceLength": 10,
+         "stack": true,
+         "steppedLine": false,
+         "targets": [
+            {
+               "expr": "sum by (source_zone) (rate(ceph_data_sync_from_zone_poll_latency_sum[30s]))",
+               "format": "time_series",
+               "intervalFactor": 1,
+               "legendFormat": "{{source_zone}}",
+               "refId": "A"
+            }
+         ],
+         "thresholds": [ ],
+         "timeFrom": null,
+         "timeShift": null,
+         "title": "Polling Request Latency from Source Zone",
+         "tooltip": {
+            "shared": true,
+            "sort": 0,
+            "value_type": "individual"
+         },
+         "type": "graph",
+         "xaxis": {
+            "buckets": null,
+            "mode": "time",
+            "name": null,
+            "show": true,
+            "values": [ ]
+         },
+         "yaxes": [
+            {
+               "format": "ms",
+               "label": null,
+               "logBase": 1,
+               "max": null,
+               "min": 0,
+               "show": true
+            },
+            {
+               "format": "short",
+               "label": null,
+               "logBase": 1,
+               "max": null,
+               "min": 0,
+               "show": true
+            }
+         ]
       },
       {
-        "current": {
-        "tags": [],
-        "text": "default",
-        "value": "default"
-        },
-        "hide": 0,
-        "label": "Data Source",
-        "name": "datasource",
-        "options": [],
-        "query": "prometheus",
-        "refresh": 1,
-        "regex": "",
-        "type": "datasource"
+         "aliasColors": { },
+         "bars": false,
+         "dashLength": 10,
+         "dashes": false,
+         "datasource": "$datasource",
+         "fill": 1,
+         "gridPos": {
+            "h": 7,
+            "w": 8,
+            "x": 0,
+            "y": 7
+         },
+         "id": 5,
+         "legend": {
+            "alignAsTable": false,
+            "avg": false,
+            "current": false,
+            "max": false,
+            "min": false,
+            "rightSide": false,
+            "show": true,
+            "sideWidth": null,
+            "total": false,
+            "values": false
+         },
+         "lines": true,
+         "linewidth": 1,
+         "links": [ ],
+         "nullPointMode": "null as zero",
+         "percentage": false,
+         "pointradius": 5,
+         "points": false,
+         "renderer": "flot",
+         "repeat": null,
+         "seriesOverrides": [ ],
+         "spaceLength": 10,
+         "stack": true,
+         "steppedLine": false,
+         "targets": [
+            {
+               "expr": "sum by (source_zone) (rate(ceph_data_sync_from_zone_fetch_errors[30s]))",
+               "format": "time_series",
+               "intervalFactor": 1,
+               "legendFormat": "{{source_zone}}",
+               "refId": "A"
+            }
+         ],
+         "thresholds": [ ],
+         "timeFrom": null,
+         "timeShift": null,
+         "title": "Unsuccessful Object Replications from Source Zone",
+         "tooltip": {
+            "shared": true,
+            "sort": 0,
+            "value_type": "individual"
+         },
+         "type": "graph",
+         "xaxis": {
+            "buckets": null,
+            "mode": "time",
+            "name": null,
+            "show": true,
+            "values": [ ]
+         },
+         "yaxes": [
+            {
+               "format": "short",
+               "label": "Count/s",
+               "logBase": 1,
+               "max": null,
+               "min": 0,
+               "show": true
+            },
+            {
+               "format": "short",
+               "label": null,
+               "logBase": 1,
+               "max": null,
+               "min": 0,
+               "show": true
+            }
+         ]
       }
-    ]
-  },
-  "time": {
-    "from": "now-1h",
-    "to": "now"
-  },
-  "timepicker": {
-    "refresh_intervals": [
-      "5s",
-      "10s",
-      "15s",
-      "30s",
-      "1m",
-      "5m",
-      "15m",
-      "30m",
-      "1h",
-      "2h",
-      "1d"
-    ],
-    "time_options": [
-      "5m",
-      "15m",
-      "1h",
-      "6h",
-      "12h",
-      "24h",
-      "2d",
-      "7d",
-      "30d"
-    ]
-  },
-  "timezone": "",
-  "title": "RGW Sync Overview",
-  "uid": "rgw-sync-overview",
-  "version": 2
+   ],
+   "refresh": "15s",
+   "rows": [ ],
+   "schemaVersion": 16,
+   "style": "dark",
+   "tags": [
+      "overview"
+   ],
+   "templating": {
+      "list": [
+         {
+            "allValue": null,
+            "current": { },
+            "datasource": "$datasource",
+            "hide": 2,
+            "includeAll": true,
+            "label": null,
+            "multi": false,
+            "name": "rgw_servers",
+            "options": [ ],
+            "query": "prometehus",
+            "refresh": 1,
+            "regex": "",
+            "sort": 1,
+            "tagValuesQuery": "",
+            "tags": [ ],
+            "tagsQuery": "",
+            "type": "query",
+            "useTags": false
+         },
+         {
+            "current": {
+               "text": "default",
+               "value": "default"
+            },
+            "hide": 0,
+            "label": "Data Source",
+            "name": "datasource",
+            "options": [ ],
+            "query": "prometheus",
+            "refresh": 1,
+            "regex": "",
+            "type": "datasource"
+         }
+      ]
+   },
+   "time": {
+      "from": "now-1h",
+      "to": "now"
+   },
+   "timepicker": {
+      "refresh_intervals": [
+         "5s",
+         "10s",
+         "15s",
+         "30s",
+         "1m",
+         "5m",
+         "15m",
+         "30m",
+         "1h",
+         "2h",
+         "1d"
+      ],
+      "time_options": [
+         "5m",
+         "15m",
+         "1h",
+         "6h",
+         "12h",
+         "24h",
+         "2d",
+         "7d",
+         "30d"
+      ]
+   },
+   "timezone": "",
+   "title": "RGW Sync Overview",
+   "uid": "rgw-sync-overview",
+   "version": 0
 }

diff --git a/monitoring/grafana/dashboards/requirements-grafonnet.txt b/monitoring/grafana/dashboards/requirements-grafonnet.txt
new file mode 100644 (file)
index 0000000..9891d55
--- /dev/null
+++ b/monitoring/grafana/dashboards/requirements-grafonnet.txt
@@ -0,0 +1 @@
+jsondiff

diff --git a/monitoring/grafana/dashboards/test-jsonnet.sh b/monitoring/grafana/dashboards/test-jsonnet.sh
new file mode 100644 (file)
index 0000000..127992c
--- /dev/null
+++ b/monitoring/grafana/dashboards/test-jsonnet.sh
@@ -0,0 +1,30 @@
+#!/usr/bin/env bash
+
+set -e
+TEMPDIR=`mktemp -d`
+BASEDIR=$(dirname "$0")
+
+JSONNET_PATH="${GRAFONNET_PATH}" jsonnet -m ${TEMPDIR} $BASEDIR/jsonnet/grafana_dashboards.jsonnet
+
+truncate -s 0 ${TEMPDIR}/json_difference.log
+for json_files in $BASEDIR/*.json
+do
+    JSON_FILE_NAME=$(basename $json_files)
+    for generated_files in ${TEMPDIR}/*.json
+    do
+        GENERATED_FILE_NAME=$(basename $generated_files)
+        if [ $JSON_FILE_NAME == $GENERATED_FILE_NAME ]; then
+            jsondiff --indent 2 $generated_files $json_files | tee -a ${TEMPDIR}/json_difference.log
+        fi
+    done
+done
+
+if [[ $(wc -l < ${TEMPDIR}/json_difference.log) -eq 0 ]]
+then
+    rm -rf ${TEMPDIR}
+    echo "Congratulations! Grafonnet Check Passed"
+else
+    rm -rf ${TEMPDIR}
+    echo "Grafonnet Check Failed, failed comparing generated file with existing"
+    exit 1
+fi

diff --git a/monitoring/grafana/dashboards/tox.ini b/monitoring/grafana/dashboards/tox.ini
new file mode 100644 (file)
index 0000000..10aeb9f
--- /dev/null
+++ b/monitoring/grafana/dashboards/tox.ini
@@ -0,0 +1,22 @@
+[tox]
+envlist = grafonnet-{check,fix}
+skipsdist = true
+
+[grafonnet]
+deps =
+    -rrequirements-grafonnet.txt
+
+[testenv:grafonnet-{check,fix}]
+basepython = python3
+whitelist_externals =
+    jsonnet
+    bash
+description =
+    check: Ensure that auto-generated grafana dashboard files matches the current version
+    fix: generate dashboard json files from jsonnet file with latest changes
+deps =
+    {[grafonnet]deps}
+passenv = GRAFONNET_PATH
+commands =
+     check: bash test-jsonnet.sh
+     fix: jsonnet -m . jsonnet/grafana_dashboards.jsonnet

diff --git a/run-make-check.sh b/run-make-check.sh
index 0e07cfe48353d73df29ace5f54b831a43567f83d..f0d0397f7021bcfc0d2990a8031048d3e252d47c 100755 (executable)
--- a/run-make-check.sh
+++ b/run-make-check.sh
@@ -71,7 +71,7 @@ function main() {
     fi
     FOR_MAKE_CHECK=1 prepare
     # Init defaults after deps are installed.
-    local cmake_opts=" -DWITH_GTEST_PARALLEL=ON -DWITH_FIO=ON -DWITH_CEPHFS_SHELL=ON -DWITH_SPDK=ON -DENABLE_GIT_VERSION=OFF"
+    local cmake_opts=" -DWITH_GTEST_PARALLEL=ON -DWITH_FIO=ON -DWITH_CEPHFS_SHELL=ON -DWITH_GRAFANA=ON -DWITH_SPDK=ON -DENABLE_GIT_VERSION=OFF"
     if [ $WITH_SEASTAR ]; then
         cmake_opts+=" -DWITH_SEASTAR=ON"
     fi