fix: building base image -> update child images

additional user in slurm base docker image
labels, some cleanup
2021-06-23 14:29:32 +02:00 · 2021-06-23 14:28:50 +02:00 · 2021-06-22 19:09:52 +02:00 · 2021-06-22 16:48:56 +02:00 · 2021-06-22 00:26:00 +02:00 · 2021-06-21 21:34:24 +02:00
41 changed files with 4553 additions and 129 deletions
--- a/.gitignore
+++ b/.gitignore
@@ -1,2 +1,3 @@
 .*.swp
 *.retry
+vars_auth.yml
--- a/install.sh
+++ b/install.sh
@@ -0,0 +1,4 @@
+mkdir -p ./collections/ansible_collections/community
+# git clone --depth=1 -b 1.2.1 https://github.com/ansible-collections/community.grafana.git ./collections/ansible_collections/community/grafana
+git clone --depth=1 -b 1.2.1-extended https://github.com/ansible_community.grafana.git ./collections/ansible_collections/community/grafana
+
--- a/inv.yml
+++ b/inv.yml
@@ -6,6 +6,12 @@ all:
      ssh_args: -o ControlMaster=auto -o ControlPersist=60s
      # ansible_host: 192.168.122.139
      unpriv_user: thoto
+    ed-c7-2:
+      ansible_user: root
+      ansible_host: ed-c7-2.virt.uller.thoto.net
+      # ansible_host: 192.168.123.60 # + jumphost
+      ssh_args: -o ControlMaster=auto -o ControlPersist=60s
+      unpriv_user: thoto
  children:
    htcondor:
      hosts:
@@ -13,3 +19,8 @@ all:
    slurm:
      hosts:
        ed-c7-1:
+        ed-c7-2:
+    cobald:
+      hosts:
+        ed-c7-1:
+        ed-c7-2:
--- a/play.yml
+++ b/play.yml
@@ -1,11 +1,25 @@
 ---
 - hosts: all
+  vars_files:
+  - vars-auth.yml
+  - vars-influx.yml
  tasks:
+  - name: "install epel repo" # for htop etc.
+    yum:
+      name: epel-release
+      state: present
+
  - name: "install tools"
    yum:
      name: [ vim-enhanced, htop, screen, bind-utils, nmap-ncat, net-tools ]
      state: present

+  - name: "install ssh-key"
+    authorized_key:
+      user: thoto
+      key: "{{ssh_key}}"
+      state: present
+
 - hosts: htcondor
  pre_tasks:
  - name: "install htcondor repo"
@@ -41,16 +55,42 @@
      - htcondor-containered
      - htcondor

- hosts: slurm
+- hosts: slurm, cobald
  vars:
    container_privileged: True
-    num_nodes: 3
-  roles:
+    slurm_num_nodes: 10
+  tasks:
  - name: "setup docker"
-    role: docker
+    import_role: name=docker
    tags: docker
+
+  - name: "get facts from existing cobald instance (i.e. hostname)"
+    import_role:
+      name: cobald
+      tasks_from: facts
+    vars:
+      container_name: cobald
+    tags: [ slurm, cobald ]
+
  - name: "setup slurm test environment in docker containers"
-    role: slurm
+    import_role: name=slurm
    vars:
      slurm_user: slurm # or root
+      slurm_user_accounts:
+      - name: cobald
+        dir: /var/lib/cobald
+      num_nodes: "{{slurm_num_nodes}}"
+      extra_nodes:
+      - "{{cobald_container_hostname}}" # from cobald_facts, read or generated
+      docker_network: slurm
+    when: '"slurm" in group_names'
    tags: slurm
+
+  - name: "install cobald"
+    import_role: name=cobald
+    vars:
+      cobald_slurm: True
+      container_name: cobald
+      # docker_network: slurm # overriden by vars/slurm.yml
+    when: '"cobald" in group_names'
+    tags: cobald
--- a/roles/cobald/defaults/main.yml
+++ b/roles/cobald/defaults/main.yml
@@ -0,0 +1,6 @@
+cobald_domainname: cobald.local
+influx_admin_user: my-user
+influx_admin_pw: my-password
+influx_org: my-org
+influx_pubport: 28086
+influx_bucket: batleth
--- a/roles/cobald/files/cobald-entrypoint.sh
+++ b/roles/cobald/files/cobald-entrypoint.sh
@@ -0,0 +1,7 @@
+#!/bin/sh
+
+for i in /usr/local/lib/entrypoints.d/* ; do
+    [ -f $i ] && /bin/sh $i || break
+done
+
+exec "${@:-/bin/bash}"
--- a/roles/cobald/files/grafana-dashboard.json
+++ b/roles/cobald/files/grafana-dashboard.json
@@ -0,0 +1,589 @@
+{
+  "annotations": {
+    "list": [
+      {
+        "builtIn": 1,
+        "datasource": "-- Grafana --",
+        "enable": true,
+        "hide": true,
+        "iconColor": "rgba(0, 211, 255, 1)",
+        "name": "Annotations & Alerts",
+        "type": "dashboard"
+      }
+    ]
+  },
+  "editable": true,
+  "gnetId": null,
+  "graphTooltip": 0,
+  "id": 1,
+  "iteration": 1623317629899,
+  "links": [],
+  "panels": [
+    {
+      "collapsed": false,
+      "datasource": null,
+      "gridPos": {
+        "h": 1,
+        "w": 24,
+        "x": 0,
+        "y": 0
+      },
+      "id": 4,
+      "panels": [],
+      "title": "Row title",
+      "type": "row"
+    },
+    {
+      "aliasColors": {},
+      "bars": false,
+      "dashLength": 10,
+      "dashes": false,
+      "datasource": "InfluxDB",
+      "fieldConfig": {
+        "defaults": {
+          "unit": "none"
+        },
+        "overrides": []
+      },
+      "fill": 1,
+      "fillGradient": 0,
+      "gridPos": {
+        "h": 9,
+        "w": 12,
+        "x": 0,
+        "y": 1
+      },
+      "hiddenSeries": false,
+      "id": 2,
+      "interval": "1s",
+      "legend": {
+        "avg": false,
+        "current": false,
+        "max": true,
+        "min": true,
+        "show": true,
+        "total": false,
+        "values": true
+      },
+      "lines": true,
+      "linewidth": 1,
+      "maxDataPoints": 200,
+      "nullPointMode": "null",
+      "options": {
+        "alertThreshold": true
+      },
+      "percentage": false,
+      "pluginVersion": "7.5.7",
+      "pointradius": 2,
+      "points": false,
+      "renderer": "flot",
+      "seriesOverrides": [],
+      "spaceLength": 10,
+      "stack": false,
+      "steppedLine": false,
+      "targets": [
+        {
+          "groupBy": [
+            {
+              "params": [
+                "$__interval"
+              ],
+              "type": "time"
+            },
+            {
+              "params": [
+                "null"
+              ],
+              "type": "fill"
+            }
+          ],
+          "orderByTime": "ASC",
+          "policy": "default",
+          "query": "f_r = (r, accumulator) => ({\n    _value: accumulator._value + (\n      if r._value == \"AvailableState\" then 1\n      else if r._value == \"DownState\" then -1\n      else 0)\n    })\n\nnodes = from(bucket: \"batleth\")\n  |> range(start: v.timeRangeStart, stop: v.timeRangeStop)\n  |> keep(columns: [\"tardis_machine_name\", \"_time\"])\n  |> sort(columns: [\"_time\"], desc: true)\n  |> unique(column: \"tardis_machine_name\")\n//  |> yield()\n\noffset = from(bucket: \"batleth\")\n  |> range(start: 0, stop: v.timeRangeStart)\n  |> filter(fn: (r) => r._field == \"state\")\n  |> group(columns: [\"tardis_machine_name\", \"machine_type\"])\n  |> reduce(fn: f_r, identity: {_value: 0})\n  |> duplicate(column: \"_stop\", as: \"_time\")\n\nnew = from(bucket: \"batleth\")\n  |> range(start: v.timeRangeStart, stop: v.timeRangeStop)\n  |> filter(fn: (r) => r._field == \"state\")\n//  |> filter(fn: (r) => r.tardis_machine_name == \"${machine}\")\n  |> group(columns: [\"tardis_machine_name\", \"machine_type\"])\n  |> window(every: $__interval)\n  |> reduce(fn: f_r, identity: {_value: 0})\n  |> duplicate(column: \"_stop\", as: \"_time\")\n\nunion(tables: [offset, new])\n  |> window(every: inf)\n  |> cumulativeSum()\n  |> range(start: v.timeRangeStart, stop: v.timeRangeStop)\n  |> yield()\n",
+          "refId": "A",
+          "resultFormat": "time_series",
+          "select": [
+            [
+              {
+                "params": [
+                  "value"
+                ],
+                "type": "field"
+              },
+              {
+                "params": [],
+                "type": "mean"
+              }
+            ]
+          ],
+          "tags": []
+        }
+      ],
+      "thresholds": [],
+      "timeFrom": null,
+      "timeRegions": [],
+      "timeShift": null,
+      "title": "nodes running",
+      "tooltip": {
+        "shared": true,
+        "sort": 0,
+        "value_type": "individual"
+      },
+      "type": "graph",
+      "xaxis": {
+        "buckets": null,
+        "mode": "time",
+        "name": null,
+        "show": true,
+        "values": []
+      },
+      "yaxes": [
+        {
+          "format": "none",
+          "label": null,
+          "logBase": 1,
+          "max": null,
+          "min": null,
+          "show": true
+        },
+        {
+          "format": "short",
+          "label": null,
+          "logBase": 1,
+          "max": null,
+          "min": null,
+          "show": true
+        }
+      ],
+      "yaxis": {
+        "align": false,
+        "alignLevel": null
+      }
+    },
+    {
+      "aliasColors": {},
+      "bars": false,
+      "dashLength": 10,
+      "dashes": false,
+      "datasource": "InfluxDB",
+      "fieldConfig": {
+        "defaults": {},
+        "overrides": []
+      },
+      "fill": 1,
+      "fillGradient": 0,
+      "gridPos": {
+        "h": 9,
+        "w": 12,
+        "x": 12,
+        "y": 1
+      },
+      "hiddenSeries": false,
+      "id": 8,
+      "interval": "1s",
+      "legend": {
+        "avg": false,
+        "current": false,
+        "max": false,
+        "min": false,
+        "show": true,
+        "total": false,
+        "values": false
+      },
+      "lines": true,
+      "linewidth": 1,
+      "maxDataPoints": null,
+      "nullPointMode": "null",
+      "options": {
+        "alertThreshold": true
+      },
+      "percentage": false,
+      "pluginVersion": "7.5.7",
+      "pointradius": 2,
+      "points": false,
+      "renderer": "flot",
+      "seriesOverrides": [],
+      "spaceLength": 10,
+      "stack": false,
+      "steppedLine": false,
+      "targets": [
+        {
+          "groupBy": [
+            {
+              "params": [
+                "$__interval"
+              ],
+              "type": "time"
+            },
+            {
+              "params": [
+                "null"
+              ],
+              "type": "fill"
+            }
+          ],
+          "orderByTime": "ASC",
+          "policy": "default",
+          "query": "from(bucket: \"batleth\")\n  |> range(start: v.timeRangeStart, stop: v.timeRangeStop)\n  |> filter(fn: (r) => r._measurement == \"tardis_pipeline\")\n  |> filter(fn: (r) => r._field == \"demand\" or r._field == \"supply\")\n  |> drop(columns: [\"host\"])\n  |> aggregateWindow(every: $__interval, fn: mean)\n  |> yield()",
+          "queryType": "randomWalk",
+          "refId": "A",
+          "resultFormat": "time_series",
+          "select": [
+            [
+              {
+                "params": [
+                  "value"
+                ],
+                "type": "field"
+              },
+              {
+                "params": [],
+                "type": "mean"
+              }
+            ]
+          ],
+          "tags": []
+        }
+      ],
+      "thresholds": [],
+      "timeFrom": null,
+      "timeRegions": [],
+      "timeShift": null,
+      "title": "pipeline demand/supply (mean)",
+      "tooltip": {
+        "shared": true,
+        "sort": 0,
+        "value_type": "individual"
+      },
+      "type": "graph",
+      "xaxis": {
+        "buckets": null,
+        "mode": "time",
+        "name": null,
+        "show": true,
+        "values": []
+      },
+      "yaxes": [
+        {
+          "$$hashKey": "object:115",
+          "decimals": null,
+          "format": "short",
+          "label": "cpus",
+          "logBase": 1,
+          "max": null,
+          "min": null,
+          "show": true
+        },
+        {
+          "$$hashKey": "object:116",
+          "format": "short",
+          "label": null,
+          "logBase": 1,
+          "max": null,
+          "min": null,
+          "show": true
+        }
+      ],
+      "yaxis": {
+        "align": false,
+        "alignLevel": null
+      }
+    },
+    {
+      "aliasColors": {},
+      "bars": false,
+      "dashLength": 10,
+      "dashes": false,
+      "datasource": "InfluxDB",
+      "fieldConfig": {
+        "defaults": {},
+        "overrides": []
+      },
+      "fill": 1,
+      "fillGradient": 0,
+      "gridPos": {
+        "h": 8,
+        "w": 12,
+        "x": 0,
+        "y": 10
+      },
+      "hiddenSeries": false,
+      "id": 6,
+      "interval": "1s",
+      "legend": {
+        "avg": false,
+        "current": false,
+        "max": false,
+        "min": false,
+        "show": true,
+        "total": false,
+        "values": false
+      },
+      "lines": true,
+      "linewidth": 1,
+      "nullPointMode": "null",
+      "options": {
+        "alertThreshold": true
+      },
+      "percentage": false,
+      "pluginVersion": "7.5.7",
+      "pointradius": 2,
+      "points": false,
+      "renderer": "flot",
+      "seriesOverrides": [],
+      "spaceLength": 10,
+      "stack": false,
+      "steppedLine": false,
+      "targets": [
+        {
+          "groupBy": [
+            {
+              "params": [
+                "$__interval"
+              ],
+              "type": "time"
+            },
+            {
+              "params": [
+                "null"
+              ],
+              "type": "fill"
+            }
+          ],
+          "orderByTime": "ASC",
+          "policy": "default",
+          "query": "f = (r, accumulator) => ({\n    _value: accumulator._value + (if r._value == \"AvailableState\" then 1 else if r._value == \"DownState\" then -1 else 0)\n  })\n\nfrom(bucket: \"batleth\")\n  |> range(start: v.timeRangeStart, stop: v.timeRangeStop)\n  |> filter(fn: (r) => r._field == \"state\")\n  |> group(columns: [\"tardis_machine_name\"])\n  |> window(every: $__interval)\n  |> reduce(fn: f, identity: {_value: 0})\n  |> duplicate(column: \"_stop\", as: \"_time\")\n  |> window(every: inf, timeColumn: \"_time\")\n  |> yield()",
+          "queryType": "randomWalk",
+          "refId": "A",
+          "resultFormat": "time_series",
+          "select": [
+            [
+              {
+                "params": [
+                  "value"
+                ],
+                "type": "field"
+              },
+              {
+                "params": [],
+                "type": "mean"
+              }
+            ]
+          ],
+          "tags": []
+        }
+      ],
+      "thresholds": [],
+      "timeFrom": null,
+      "timeRegions": [],
+      "timeShift": null,
+      "title": "node fluctuation",
+      "tooltip": {
+        "shared": true,
+        "sort": 0,
+        "value_type": "individual"
+      },
+      "type": "graph",
+      "xaxis": {
+        "buckets": null,
+        "mode": "time",
+        "name": null,
+        "show": true,
+        "values": []
+      },
+      "yaxes": [
+        {
+          "format": "short",
+          "label": null,
+          "logBase": 1,
+          "max": null,
+          "min": null,
+          "show": true
+        },
+        {
+          "format": "short",
+          "label": null,
+          "logBase": 1,
+          "max": null,
+          "min": null,
+          "show": true
+        }
+      ],
+      "yaxis": {
+        "align": false,
+        "alignLevel": null
+      }
+    },
+    {
+      "aliasColors": {},
+      "bars": false,
+      "dashLength": 10,
+      "dashes": false,
+      "datasource": "InfluxDB",
+      "fieldConfig": {
+        "defaults": {},
+        "overrides": []
+      },
+      "fill": 1,
+      "fillGradient": 0,
+      "gridPos": {
+        "h": 8,
+        "w": 12,
+        "x": 12,
+        "y": 10
+      },
+      "hiddenSeries": false,
+      "id": 10,
+      "interval": "1s",
+      "legend": {
+        "avg": false,
+        "current": false,
+        "max": false,
+        "min": false,
+        "show": true,
+        "total": false,
+        "values": false
+      },
+      "lines": true,
+      "linewidth": 1,
+      "maxDataPoints": null,
+      "nullPointMode": "null",
+      "options": {
+        "alertThreshold": true
+      },
+      "percentage": false,
+      "pluginVersion": "7.5.7",
+      "pointradius": 2,
+      "points": false,
+      "renderer": "flot",
+      "seriesOverrides": [],
+      "spaceLength": 10,
+      "stack": false,
+      "steppedLine": false,
+      "targets": [
+        {
+          "groupBy": [
+            {
+              "params": [
+                "$__interval"
+              ],
+              "type": "time"
+            },
+            {
+              "params": [
+                "null"
+              ],
+              "type": "fill"
+            }
+          ],
+          "orderByTime": "ASC",
+          "policy": "default",
+          "query": "from(bucket: \"batleth\")\n  |> range(start: v.timeRangeStart, stop: v.timeRangeStop)\n  |> filter(fn: (r) => r._measurement == \"tardis_pipeline\")\n  |> filter(fn: (r) => r._field == \"utilisation\" or r._field == \"allocation\")\n  |> keep(columns: [\"_time\", \"_measurement\", \"_field\", \"_value\", \"tardis_machine_name\"])\n  |> aggregateWindow(every: $__interval, fn: mean)\n  |> yield()",
+          "queryType": "randomWalk",
+          "refId": "A",
+          "resultFormat": "time_series",
+          "select": [
+            [
+              {
+                "params": [
+                  "value"
+                ],
+                "type": "field"
+              },
+              {
+                "params": [],
+                "type": "mean"
+              }
+            ]
+          ],
+          "tags": []
+        }
+      ],
+      "thresholds": [],
+      "timeFrom": null,
+      "timeRegions": [],
+      "timeShift": null,
+      "title": "pipeline (utilization/allocation)",
+      "tooltip": {
+        "shared": true,
+        "sort": 0,
+        "value_type": "individual"
+      },
+      "type": "graph",
+      "xaxis": {
+        "buckets": null,
+        "mode": "time",
+        "name": null,
+        "show": true,
+        "values": []
+      },
+      "yaxes": [
+        {
+          "format": "short",
+          "label": null,
+          "logBase": 1,
+          "max": null,
+          "min": null,
+          "show": true
+        },
+        {
+          "format": "short",
+          "label": null,
+          "logBase": 1,
+          "max": null,
+          "min": null,
+          "show": true
+        }
+      ],
+      "yaxis": {
+        "align": false,
+        "alignLevel": null
+      }
+    }
+  ],
+  "refresh": "5s",
+  "schemaVersion": 27,
+  "style": "dark",
+  "tags": [],
+  "templating": {
+    "list": [
+      {
+        "allValue": null,
+        "current": {
+          "selected": false,
+          "text": "cobald-xvmcqc",
+          "value": "cobald-xvmcqc"
+        },
+        "datasource": "InfluxDB",
+        "definition": "from(bucket: \"batleth\")\n  |> range(start: v.timeRangeStart, stop: v.timeRangeStop)\n  |> keep(columns: [\"tardis_machine_name\", \"_time\"])\n  |> sort(columns: [\"_time\"], desc: true)\n  |> unique(column: \"tardis_machine_name\")",
+        "description": null,
+        "error": null,
+        "hide": 0,
+        "includeAll": false,
+        "label": null,
+        "multi": false,
+        "name": "machine",
+        "options": [],
+        "query": "from(bucket: \"batleth\")\n  |> range(start: v.timeRangeStart, stop: v.timeRangeStop)\n  |> keep(columns: [\"tardis_machine_name\", \"_time\"])\n  |> sort(columns: [\"_time\"], desc: true)\n  |> unique(column: \"tardis_machine_name\")",
+        "refresh": 2,
+        "regex": "",
+        "skipUrlSync": false,
+        "sort": 0,
+        "tagValuesQuery": "",
+        "tags": [],
+        "tagsQuery": "",
+        "type": "query",
+        "useTags": false
+      }
+    ]
+  },
+  "time": {
+    "from": "now-5m",
+    "to": "now"
+  },
+  "timepicker": {},
+  "timezone": "",
+  "title": "cobald",
+  "uid": "urDuvE6Gk",
+  "version": 2
+}
--- a/roles/cobald/files/influxdb-dashboard-cobald.json
+++ b/roles/cobald/files/influxdb-dashboard-cobald.json
@@ -0,0 +1,794 @@
+{
+ "meta": {
+  "version": "1",
+  "type": "dashboard",
+  "name": "cobald-Template",
+  "description": "template created from dashboard: cobald"
+ },
+ "content": {
+  "data": {
+   "type": "dashboard",
+   "attributes": {
+    "name": "cobald",
+    "description": ""
+   },
+   "relationships": {
+    "label": {
+     "data": []
+    },
+    "cell": {
+     "data": [
+      {
+       "type": "cell",
+       "id": "07900a722c363000"
+      },
+      {
+       "type": "cell",
+       "id": "07900a7236f63000"
+      },
+      {
+       "type": "cell",
+       "id": "07900a723cf63000"
+      },
+      {
+       "type": "cell",
+       "id": "07900a7243f63000"
+      },
+      {
+       "type": "cell",
+       "id": "079e694f29581000"
+      },
+      {
+       "type": "cell",
+       "id": "079e6e037c181000"
+      }
+     ]
+    },
+    "variable": {
+     "data": []
+    }
+   }
+  },
+  "included": [
+   {
+    "id": "07900a722c363000",
+    "type": "cell",
+    "attributes": {
+     "x": 4,
+     "y": 0,
+     "w": 4,
+     "h": 4
+    },
+    "relationships": {
+     "view": {
+      "data": {
+       "type": "view",
+       "id": "07900a722c363000"
+      }
+     }
+    }
+   },
+   {
+    "id": "07900a7236f63000",
+    "type": "cell",
+    "attributes": {
+     "x": 0,
+     "y": 0,
+     "w": 4,
+     "h": 4
+    },
+    "relationships": {
+     "view": {
+      "data": {
+       "type": "view",
+       "id": "07900a7236f63000"
+      }
+     }
+    }
+   },
+   {
+    "id": "07900a723cf63000",
+    "type": "cell",
+    "attributes": {
+     "x": 4,
+     "y": 4,
+     "w": 4,
+     "h": 4
+    },
+    "relationships": {
+     "view": {
+      "data": {
+       "type": "view",
+       "id": "07900a723cf63000"
+      }
+     }
+    }
+   },
+   {
+    "id": "07900a7243f63000",
+    "type": "cell",
+    "attributes": {
+     "x": 0,
+     "y": 4,
+     "w": 4,
+     "h": 4
+    },
+    "relationships": {
+     "view": {
+      "data": {
+       "type": "view",
+       "id": "07900a7243f63000"
+      }
+     }
+    }
+   },
+   {
+    "id": "079e694f29581000",
+    "type": "cell",
+    "attributes": {
+     "x": 8,
+     "y": 0,
+     "w": 4,
+     "h": 4
+    },
+    "relationships": {
+     "view": {
+      "data": {
+       "type": "view",
+       "id": "079e694f29581000"
+      }
+     }
+    }
+   },
+   {
+    "id": "079e6e037c181000",
+    "type": "cell",
+    "attributes": {
+     "x": 8,
+     "y": 4,
+     "w": 4,
+     "h": 4
+    },
+    "relationships": {
+     "view": {
+      "data": {
+       "type": "view",
+       "id": "079e6e037c181000"
+      }
+     }
+    }
+   },
+   {
+    "type": "view",
+    "id": "07900a722c363000",
+    "attributes": {
+     "name": "nodes running",
+     "properties": {
+      "shape": "chronograf-v2",
+      "queries": [
+       {
+        "text": "from(bucket: \"batleth\")\n  |> range(start: v.timeRangeStart, stop: v.timeRangeStop)\n  |> filter(fn: (r) => r._field == \"state\")\n  |> group()\n  |> window(every: 10s)\n  |> reduce(fn: (r, accumulator) => ({\n      _value: accumulator._value + (\n        if r._value == \"AvailableState\" then 1 \n        else if r._value == \"DownState\" then -1 \n        else 0)\n    }), identity: {_value: 0})\n  |> duplicate(column: \"_stop\", as: \"_time\")\n  |> window(every: inf) //, timeColumn: \"_time\")\n  |> cumulativeSum()\n//  |> reduce(fn: (r, accumulator) => ({r with x: r._value * 2}), identity: {x:0})\n//  |> map(fn: (r) => ({r with vnew: r._value*2}))\n//  |> integral(unit: 10s, timeColumn: \"_stop\")\n//  |> window(every: inf) //, timeColumn: \"_stop\")\n  |> yield()",
+        "editMode": "advanced",
+        "name": "",
+        "builderConfig": {
+         "buckets": [],
+         "tags": [
+          {
+           "key": "_measurement",
+           "values": [],
+           "aggregateFunctionType": "filter"
+          }
+         ],
+         "functions": [
+          {
+           "name": "mean"
+          }
+         ],
+         "aggregateWindow": {
+          "period": "auto",
+          "fillValues": false
+         }
+        }
+       }
+      ],
+      "axes": {
+       "x": {
+        "bounds": [
+         "",
+         ""
+        ],
+        "label": "",
+        "prefix": "",
+        "suffix": "",
+        "base": "10",
+        "scale": "linear"
+       },
+       "y": {
+        "bounds": [
+         "",
+         ""
+        ],
+        "label": "",
+        "prefix": "",
+        "suffix": "",
+        "base": "10",
+        "scale": "linear"
+       }
+      },
+      "type": "xy",
+      "legend": {},
+      "geom": "line",
+      "colors": [
+       {
+        "id": "9b960932-18d9-4f57-80ba-24998a06613d",
+        "type": "scale",
+        "hex": "#31C0F6",
+        "name": "Nineteen Eighty Four",
+        "value": 0
+       },
+       {
+        "id": "c1742651-0d5e-4148-b9c0-92beb642417a",
+        "type": "scale",
+        "hex": "#A500A5",
+        "name": "Nineteen Eighty Four",
+        "value": 0
+       },
+       {
+        "id": "756fa8cb-9d9c-4e45-9a4f-f2b106b0216a",
+        "type": "scale",
+        "hex": "#FF7E27",
+        "name": "Nineteen Eighty Four",
+        "value": 0
+       }
+      ],
+      "note": "",
+      "showNoteWhenEmpty": false,
+      "xColumn": "_time",
+      "generateXAxisTicks": [],
+      "xTotalTicks": 0,
+      "xTickStart": 0,
+      "xTickStep": 0,
+      "yColumn": "_value",
+      "generateYAxisTicks": [],
+      "yTotalTicks": 0,
+      "yTickStart": 0,
+      "yTickStep": 0,
+      "shadeBelow": false,
+      "position": "overlaid",
+      "timeFormat": "",
+      "hoverDimension": "auto",
+      "legendColorizeRows": true,
+      "legendOpacity": 1,
+      "legendOrientationThreshold": 100000000
+     }
+    }
+   },
+   {
+    "type": "view",
+    "id": "07900a7236f63000",
+    "attributes": {
+     "name": "Name this Cell",
+     "properties": {
+      "shape": "chronograf-v2",
+      "queries": [
+       {
+        "text": "from(bucket: \"batleth\")\n  |> range(start: v.timeRangeStart, stop: v.timeRangeStop)\n  |> filter(fn: (r) => r[\"tardis_machine_name\"] == \"678162c190d5\")\n  |> window(every: 10s)\n  |> count()\n  |> duplicate(column: \"_stop\", as: \"_time\")\n  |> window(every: inf)\n  |> yield()",
+        "editMode": "advanced",
+        "name": "",
+        "builderConfig": {
+         "buckets": [],
+         "tags": [
+          {
+           "key": "_measurement",
+           "values": [],
+           "aggregateFunctionType": "filter"
+          }
+         ],
+         "functions": [
+          {
+           "name": "mean"
+          }
+         ],
+         "aggregateWindow": {
+          "period": "auto",
+          "fillValues": false
+         }
+        }
+       }
+      ],
+      "axes": {
+       "x": {
+        "bounds": [
+         "",
+         ""
+        ],
+        "label": "",
+        "prefix": "",
+        "suffix": "",
+        "base": "10",
+        "scale": "linear"
+       },
+       "y": {
+        "bounds": [
+         "",
+         ""
+        ],
+        "label": "",
+        "prefix": "",
+        "suffix": "",
+        "base": "10",
+        "scale": "linear"
+       }
+      },
+      "type": "xy",
+      "legend": {},
+      "geom": "line",
+      "colors": [
+       {
+        "id": "2566435b-7ee0-4222-8ac0-b7f14ab783d9",
+        "type": "scale",
+        "hex": "#31C0F6",
+        "name": "Nineteen Eighty Four",
+        "value": 0
+       },
+       {
+        "id": "9263bcff-35a0-4025-bacd-68a1bef54784",
+        "type": "scale",
+        "hex": "#A500A5",
+        "name": "Nineteen Eighty Four",
+        "value": 0
+       },
+       {
+        "id": "2f04bd8c-2203-4be6-bc34-c25720d24379",
+        "type": "scale",
+        "hex": "#FF7E27",
+        "name": "Nineteen Eighty Four",
+        "value": 0
+       }
+      ],
+      "note": "",
+      "showNoteWhenEmpty": false,
+      "xColumn": "_time",
+      "generateXAxisTicks": [],
+      "xTotalTicks": 0,
+      "xTickStart": 0,
+      "xTickStep": 0,
+      "yColumn": "_value",
+      "generateYAxisTicks": [],
+      "yTotalTicks": 0,
+      "yTickStart": 0,
+      "yTickStep": 0,
+      "shadeBelow": false,
+      "position": "overlaid",
+      "timeFormat": "",
+      "hoverDimension": "auto",
+      "legendColorizeRows": true,
+      "legendOpacity": 1,
+      "legendOrientationThreshold": 100000000
+     }
+    }
+   },
+   {
+    "type": "view",
+    "id": "07900a723cf63000",
+    "attributes": {
+     "name": "node fluctuation",
+     "properties": {
+      "shape": "chronograf-v2",
+      "queries": [
+       {
+        "text": "f = (r, accumulator) => ({\n    _value: accumulator._value + (if r._value == \"AvailableState\" then 1 else if r._value == \"DownState\" then -1 else 0)\n  })\n\nfrom(bucket: \"batleth\")\n  |> range(start: v.timeRangeStart, stop: v.timeRangeStop)\n  |> filter(fn: (r) => r._field == \"state\")\n  |> group()\n  |> window(every: 10s)\n  |> reduce(fn: f, identity: {_value: 0})\n  |> duplicate(column: \"_stop\", as: \"_time\")\n  |> window(every: inf, timeColumn: \"_time\")\n  |> yield()",
+        "editMode": "advanced",
+        "name": "",
+        "builderConfig": {
+         "buckets": [],
+         "tags": [
+          {
+           "key": "_measurement",
+           "values": [],
+           "aggregateFunctionType": "filter"
+          }
+         ],
+         "functions": [
+          {
+           "name": "mean"
+          }
+         ],
+         "aggregateWindow": {
+          "period": "auto",
+          "fillValues": false
+         }
+        }
+       }
+      ],
+      "axes": {
+       "x": {
+        "bounds": [
+         "",
+         ""
+        ],
+        "label": "",
+        "prefix": "",
+        "suffix": "",
+        "base": "10",
+        "scale": "linear"
+       },
+       "y": {
+        "bounds": [
+         "",
+         ""
+        ],
+        "label": "",
+        "prefix": "",
+        "suffix": "",
+        "base": "10",
+        "scale": "linear"
+       }
+      },
+      "type": "xy",
+      "legend": {},
+      "geom": "line",
+      "colors": [
+       {
+        "id": "9b960932-18d9-4f57-80ba-24998a06613d",
+        "type": "scale",
+        "hex": "#31C0F6",
+        "name": "Nineteen Eighty Four",
+        "value": 0
+       },
+       {
+        "id": "c1742651-0d5e-4148-b9c0-92beb642417a",
+        "type": "scale",
+        "hex": "#A500A5",
+        "name": "Nineteen Eighty Four",
+        "value": 0
+       },
+       {
+        "id": "756fa8cb-9d9c-4e45-9a4f-f2b106b0216a",
+        "type": "scale",
+        "hex": "#FF7E27",
+        "name": "Nineteen Eighty Four",
+        "value": 0
+       }
+      ],
+      "note": "",
+      "showNoteWhenEmpty": false,
+      "xColumn": "_time",
+      "generateXAxisTicks": [],
+      "xTotalTicks": 0,
+      "xTickStart": 0,
+      "xTickStep": 0,
+      "yColumn": "_value",
+      "generateYAxisTicks": [],
+      "yTotalTicks": 0,
+      "yTickStart": 0,
+      "yTickStep": 0,
+      "shadeBelow": false,
+      "position": "overlaid",
+      "timeFormat": "",
+      "hoverDimension": "auto",
+      "legendColorizeRows": true,
+      "legendOpacity": 1,
+      "legendOrientationThreshold": 100000000
+     }
+    }
+   },
+   {
+    "type": "view",
+    "id": "07900a7243f63000",
+    "attributes": {
+     "name": "states",
+     "properties": {
+      "shape": "chronograf-v2",
+      "queries": [
+       {
+        "text": "from(bucket: \"batleth\")\n  |> range(start: v.timeRangeStart, stop: v.timeRangeStop)\n  |> filter(fn: (r) => r._field == \"state\")\n  |> group(columns: [\"_value\"], mode: \"by\")\n  |> duplicate(column: \"_value\", as: \"state\")\n  |> window(every: 10s)\n  |> count(column: \"state\")\n  |> rename(columns: {\"_value\": \"_field\", \"state\": \"_value\"})\n  |> group(columns: [\"_field\"])\n  |> yield()",
+        "editMode": "advanced",
+        "name": "",
+        "builderConfig": {
+         "buckets": [],
+         "tags": [
+          {
+           "key": "_measurement",
+           "values": [],
+           "aggregateFunctionType": "filter"
+          }
+         ],
+         "functions": [
+          {
+           "name": "mean"
+          }
+         ],
+         "aggregateWindow": {
+          "period": "auto",
+          "fillValues": false
+         }
+        }
+       }
+      ],
+      "axes": {
+       "x": {
+        "bounds": [
+         "",
+         ""
+        ],
+        "label": "",
+        "prefix": "",
+        "suffix": "",
+        "base": "10",
+        "scale": "linear"
+       },
+       "y": {
+        "bounds": [
+         "",
+         ""
+        ],
+        "label": "",
+        "prefix": "",
+        "suffix": "",
+        "base": "10",
+        "scale": "linear"
+       }
+      },
+      "type": "xy",
+      "legend": {},
+      "geom": "line",
+      "colors": [
+       {
+        "id": "9b960932-18d9-4f57-80ba-24998a06613d",
+        "type": "scale",
+        "hex": "#31C0F6",
+        "name": "Nineteen Eighty Four",
+        "value": 0
+       },
+       {
+        "id": "c1742651-0d5e-4148-b9c0-92beb642417a",
+        "type": "scale",
+        "hex": "#A500A5",
+        "name": "Nineteen Eighty Four",
+        "value": 0
+       },
+       {
+        "id": "756fa8cb-9d9c-4e45-9a4f-f2b106b0216a",
+        "type": "scale",
+        "hex": "#FF7E27",
+        "name": "Nineteen Eighty Four",
+        "value": 0
+       }
+      ],
+      "note": "",
+      "showNoteWhenEmpty": false,
+      "xColumn": "_stop",
+      "generateXAxisTicks": [],
+      "xTotalTicks": 0,
+      "xTickStart": 0,
+      "xTickStep": 0,
+      "yColumn": "_value",
+      "generateYAxisTicks": [],
+      "yTotalTicks": 0,
+      "yTickStart": 0,
+      "yTickStep": 0,
+      "shadeBelow": false,
+      "position": "overlaid",
+      "timeFormat": "",
+      "hoverDimension": "auto",
+      "legendColorizeRows": true,
+      "legendOpacity": 1,
+      "legendOrientationThreshold": 100000000
+     }
+    }
+   },
+   {
+    "type": "view",
+    "id": "079e694f29581000",
+    "attributes": {
+     "name": "pipeline (demand / supply)",
+     "properties": {
+      "shape": "chronograf-v2",
+      "queries": [
+       {
+        "text": "from(bucket: \"batleth\")\n  |> range(start: v.timeRangeStart, stop: v.timeRangeStop)\n  |> filter(fn: (r) => r._measurement == \"tardis_pipeline\")\n  |> filter(fn: (r) => r._field == \"demand\" or r._field == \"supply\")\n//  |> filter(fn: (r) => r._field == \"state\")\n//  |> group()\n//  |> window(every: 10s)\n//  |> duplicate(column: \"_stop\", as: \"_time\")\n//  |> window(every: inf) //, timeColumn: \"_time\")\n//  |> cumulativeSum()\n  |> yield()",
+        "editMode": "advanced",
+        "name": "",
+        "builderConfig": {
+         "buckets": [],
+         "tags": [
+          {
+           "key": "_measurement",
+           "values": [],
+           "aggregateFunctionType": "filter"
+          }
+         ],
+         "functions": [
+          {
+           "name": "mean"
+          }
+         ],
+         "aggregateWindow": {
+          "period": "auto",
+          "fillValues": false
+         }
+        }
+       }
+      ],
+      "axes": {
+       "x": {
+        "bounds": [
+         "",
+         ""
+        ],
+        "label": "",
+        "prefix": "",
+        "suffix": "",
+        "base": "10",
+        "scale": "linear"
+       },
+       "y": {
+        "bounds": [
+         "",
+         ""
+        ],
+        "label": "",
+        "prefix": "",
+        "suffix": "",
+        "base": "10",
+        "scale": "linear"
+       }
+      },
+      "type": "xy",
+      "legend": {},
+      "geom": "line",
+      "colors": [
+       {
+        "id": "4ef29481-ecf3-4a09-b0f5-e34e8d3e50b5",
+        "type": "scale",
+        "hex": "#31C0F6",
+        "name": "Nineteen Eighty Four",
+        "value": 0
+       },
+       {
+        "id": "719de04a-f70f-4c54-a1bb-982a9d13dbae",
+        "type": "scale",
+        "hex": "#A500A5",
+        "name": "Nineteen Eighty Four",
+        "value": 0
+       },
+       {
+        "id": "8d5c1f25-3801-4cdd-ad40-b8e2e78342a0",
+        "type": "scale",
+        "hex": "#FF7E27",
+        "name": "Nineteen Eighty Four",
+        "value": 0
+       }
+      ],
+      "note": "",
+      "showNoteWhenEmpty": false,
+      "xColumn": "_time",
+      "generateXAxisTicks": [],
+      "xTotalTicks": 0,
+      "xTickStart": 0,
+      "xTickStep": 0,
+      "yColumn": "_value",
+      "generateYAxisTicks": [],
+      "yTotalTicks": 0,
+      "yTickStart": 0,
+      "yTickStep": 0,
+      "shadeBelow": false,
+      "position": "overlaid",
+      "timeFormat": "",
+      "hoverDimension": "auto",
+      "legendColorizeRows": true,
+      "legendOpacity": 1,
+      "legendOrientationThreshold": 100000000
+     }
+    }
+   },
+   {
+    "type": "view",
+    "id": "079e6e037c181000",
+    "attributes": {
+     "name": "pipeline (Utilization / Allocation)",
+     "properties": {
+      "shape": "chronograf-v2",
+      "queries": [
+       {
+        "text": "from(bucket: \"batleth\")\n  |> range(start: v.timeRangeStart, stop: v.timeRangeStop)\n  |> filter(fn: (r) => r._measurement == \"tardis_pipeline\")\n  |> filter(fn: (r) => r._field == \"utilisation\" or r._field == \"allocation\")\n//  |> filter(fn: (r) => r._field == \"state\")\n//  |> group()\n//  |> window(every: 10s)\n//  |> duplicate(column: \"_stop\", as: \"_time\")\n//  |> window(every: inf) //, timeColumn: \"_time\")\n//  |> cumulativeSum()\n  |> yield()",
+        "editMode": "advanced",
+        "name": "",
+        "builderConfig": {
+         "buckets": [],
+         "tags": [
+          {
+           "key": "_measurement",
+           "values": [],
+           "aggregateFunctionType": "filter"
+          }
+         ],
+         "functions": [
+          {
+           "name": "mean"
+          }
+         ],
+         "aggregateWindow": {
+          "period": "auto",
+          "fillValues": false
+         }
+        }
+       }
+      ],
+      "axes": {
+       "x": {
+        "bounds": [
+         "",
+         ""
+        ],
+        "label": "",
+        "prefix": "",
+        "suffix": "",
+        "base": "10",
+        "scale": "linear"
+       },
+       "y": {
+        "bounds": [
+         "",
+         ""
+        ],
+        "label": "",
+        "prefix": "",
+        "suffix": "",
+        "base": "10",
+        "scale": "linear"
+       }
+      },
+      "type": "xy",
+      "legend": {},
+      "geom": "line",
+      "colors": [
+       {
+        "id": "4ef29481-ecf3-4a09-b0f5-e34e8d3e50b5",
+        "type": "scale",
+        "hex": "#31C0F6",
+        "name": "Nineteen Eighty Four",
+        "value": 0
+       },
+       {
+        "id": "719de04a-f70f-4c54-a1bb-982a9d13dbae",
+        "type": "scale",
+        "hex": "#A500A5",
+        "name": "Nineteen Eighty Four",
+        "value": 0
+       },
+       {
+        "id": "8d5c1f25-3801-4cdd-ad40-b8e2e78342a0",
+        "type": "scale",
+        "hex": "#FF7E27",
+        "name": "Nineteen Eighty Four",
+        "value": 0
+       }
+      ],
+      "note": "",
+      "showNoteWhenEmpty": false,
+      "xColumn": "_time",
+      "generateXAxisTicks": [],
+      "xTotalTicks": 0,
+      "xTickStart": 0,
+      "xTickStep": 0,
+      "yColumn": "_value",
+      "generateYAxisTicks": [],
+      "yTotalTicks": 0,
+      "yTickStart": 0,
+      "yTickStep": 0,
+      "shadeBelow": false,
+      "position": "overlaid",
+      "timeFormat": "",
+      "hoverDimension": "auto",
+      "legendColorizeRows": true,
+      "legendOpacity": 1,
+      "legendOrientationThreshold": 100000000
+     }
+    }
+   }
+  ]
+ },
+ "labels": []
+}
--- a/roles/cobald/files/influxdb.repo
+++ b/roles/cobald/files/influxdb.repo
@@ -0,0 +1,6 @@
+[influxdb]
+name = InfluxDB Repository - RHEL \$releasever
+baseurl = https://repos.influxdata.com/rhel/\$releasever/\$basearch/stable
+enabled = 1
+gpgcheck = 1
+gpgkey = https://repos.influxdata.com/influxdb.key
--- a/roles/cobald/files/init-cobaldmodules.sh
+++ b/roles/cobald/files/init-cobaldmodules.sh
@@ -0,0 +1,5 @@
+#!/bin/sh
+
+[ -f /usr/local/lib/cobaldmodules/setup.py -a \
+	-d /usr/local/lib/cobaldmodules/cobaldmodules ] && \
+		pip3 install --no-deps --editable /usr/local/lib/cobaldmodules
--- a/roles/cobald/files/telegraf.Dockerfile
+++ b/roles/cobald/files/telegraf.Dockerfile
@@ -0,0 +1,8 @@
+FROM centos:7
+
+COPY influxdb.repo /etc/yum.repos.d/influxdb.repo
+
+RUN yum -y install telegraf &&\
+    yum clean all && rm -rf /var/cache/yum
+
+CMD telegraf
--- a/roles/cobald/library/TODO.md
+++ b/roles/cobald/library/TODO.md
@@ -0,0 +1,11 @@
+# Tests
+    ANSIBLE_LIBRARY=. ansible -m influx_bucket -a "base='http://192.168.122.140:28086' org='my-org' auth_token='87-fEnSlQldFi1T_CLHsrHxH-T9VKey-qzUbVH6tmR2QzL4oZzbUPwzS1wzOoIkyfmyGbRv75yLjYfztxziivw==' name='bucky' description='test 123'" localhost -vvv
+    ANSIBLE_LIBRARY=. ansible -m influx_token -a "base='http://192.168.122.140:28086' org='my-org' auth_token='87-fEnSlQldFi1T_CLHsrHxH-T9VKey-qzUbVH6tmR2QzL4oZzbUPwzS1wzOoIkyfmyGbRv75yLjYfztxziivw==' key='foo' description='test 123' permissions=\"{{'[{\\\"action\\\": \\\"write\\\",\\\"resource\\\": {\\\"type\\\": \\\"buckets\\\"} }]'|from_json}}\"" localhost -vvv
+    ANSIBLE_LIBRARY=. ansible -m influx_dashboard -create -a "base='http://192.168.122.140:28086' org='my-org' token='2Mji-PvTzgn2oie5p36pJ-vxqWCnxczMWGrnYz2nUHj6Q6XvdIGiLPmK4DjX16KGhOjxQ5dWymDusE8qJrhFFg==' data='{{lookup(\"file\", \"../files/influxdb-dashboard-cobald.json\")}}'" localhost -vvv
+Missing: lot of stuff, e.g. missing tokens, invalid data like bucket instead of buckets ...
+
+# TODO
+* tests
+* state (present/absent)
+* `module_utils/urls.py` (https://github.com/ansible/ansible/blob/devel/lib/ansible/module_utils/urls.py)
+* see module notes
--- a/roles/cobald/library/influx_bucket.py
+++ b/roles/cobald/library/influx_bucket.py
@@ -0,0 +1,161 @@
+#!/usr/bin/env python
+import requests
+from ansible.module_utils.basic import AnsibleModule
+
+DOCUMENTATION = r'''
+---
+module: influx2_bucket
+short_description: create bucket in influxdb2
+description: create bucket in influxdb2
+notes:
+    - just works with influxdb version 2
+    - does not remove buckets
+    - no way to configure data retention
+
+options:
+    base:
+        description: URL for path, e.g. `https://localhost:8086`
+        type: str
+        required: True
+    org:
+        description: influxdb2 organisation
+        type: str
+        required: True
+    auth_token:
+        description: influxdb2 authentication token
+        type: str
+        required: True
+    name:
+        description: name of the bucket
+        type: str
+        required: True
+    force:
+        description: force creation even if bucket already exists
+            (adds a new one)
+        type: bool
+        required: False
+        default: False
+author:
+    - Thorsten M. (@thoto)
+'''
+
+EXAMPLES = r'''
+- name: "fetch auth token"
+  raw: influx auth list --user my-user --hide-headers | cut -f 3
+  register: influx_token_fetch
+  delegate_to: ed-influxdb-2
+
+- name: "create bucket"
+  influx_bucket:
+    base: "http://localhost:8086"
+    org: "my-org"
+    token: "{{influx_token_fetch.stdout_lines[0]}}"
+    name: "bucky"
+'''
+
+
+def get_org_id(base, org_name, h):
+    ro = requests.get("{base}/api/v2/orgs".format(base=base), headers=h,
+                      json={"org": org_name})
+    ro.raise_for_status()
+    org_id = [o["id"] for o in ro.json()["orgs"] if o["name"] == org_name]
+    return org_id[0]
+
+
+class Bucket:
+    def __init__(self, base, h, org, description, name):
+        self.base = base
+        self.h = h
+        self.org_id = get_org_id(base, org, h)
+        self.description = description
+        self.name = name
+
+        self.result = None
+        self.f = None
+
+    def check(self):
+        ra = requests.get("{base}/api/v2/buckets".format(
+                            base=self.base),
+                          params={"orgID": self.org_id, "name": self.name},
+                          headers=self.h)
+        if ra.status_code == 404:
+            # orgID + name -> 404 on empty set. Just name -> 200 but buckets=[]
+            x = []
+        else:
+            ra.raise_for_status()
+            x = [i for i in ra.json()["buckets"]
+                 if self.name == i["name"] and i["orgID"] == self.org_id]
+        assert(len(x) == 1 or len(x) == 0)
+
+        update = None
+        if len(x) == 0:  # create
+            self.result = None
+            self.f = lambda: self._create({
+                "orgID": self.org_id,
+                "description": self.description if self.description else None,
+                "name": self.name,
+                "retentionRules": []
+            })
+        else:
+            self.result = x[0]
+            if self.description == x[0].get("description", ""):
+                return False  # everything matches -> no change needed
+            else:
+                self.result = x[0]
+                update = {"id": x[0]["id"],
+                          "description": self.description}
+            self.f = lambda: self._update(**update)
+        return True
+
+    def run(self):
+        if not self.f:
+            self.check()
+        self.f()
+
+    def _update(self, id, description):
+        ra = requests.patch(
+               "{base}/api/v2/buckets/{id}".format(
+                   base=self.base, id=id),
+               headers=self.h, json={"description": description})
+        ra.raise_for_status()
+        return ra
+
+    def _create(self, data):
+        ra = requests.post("{base}/api/v2/buckets".format(base=self.base),
+                           headers=self.h, json=data)
+        ra.raise_for_status()
+        self.result = ra.json()
+        return ra
+
+
+if __name__ == "__main__":
+    result = dict(changed=False, message="")
+    module = AnsibleModule(
+        argument_spec=dict(
+            base=dict(type="str", required=True),
+            org=dict(type="str", required=True),
+            auth_token=dict(type="str", required=True),
+            name=dict(type="str", required=True),
+            description=dict(type="str", default=""),
+            force=dict(type="bool", default=False),
+        ),
+        supports_check_mode=True
+    )
+    h = {"Authorization": "Token {token}".format(
+        token=module.params["auth_token"])}
+    b = Bucket(module.params["base"], h, org=module.params["org"],
+               description=module.params["description"],
+               name=module.params["name"])
+
+    changed = b.check()
+    if b.result:
+        result['bucket_id'] = b.result["id"]
+    result['changed'] = changed
+
+    if module.check_mode:
+        module.exit_json(**result)
+
+    if changed or module.params["force"]:
+        b.run()
+        result['bucket_id'] = b.result["id"]
+    module.exit_json(**result)
--- a/roles/cobald/library/influx_dashboard.py
+++ b/roles/cobald/library/influx_dashboard.py
@@ -0,0 +1,147 @@
+#!/usr/bin/env python
+import json
+import requests
+from ansible.module_utils.basic import AnsibleModule
+
+DOCUMENTATION = r'''
+---
+module: influx2_dashboard
+short_description: create dashboard in influxdb2
+description: create dashboard in influxdb2
+notes:
+    - just works with influxdb version 2
+    - does not create dashboard description
+    - does not update dashboards
+    - just creates a dashboard if it does not exist.
+
+options:
+    base:
+        description: URL for path, e.g. `https://localhost:8086`
+        type: str
+        required: True
+    org:
+        description: influxdb2 organisation
+        type: str
+        required: True
+    auth_token:
+        description: influxdb2 authentication token
+        type: str
+        required: True
+    data:
+        description: exported dashboard json file
+        type: json
+        required: True
+    force:
+        description: force creation even if dashboard already exists
+            (adds a new one)
+        type: bool
+        required: False
+        default: False
+author:
+    - Thorsten M. (@thoto)
+'''
+
+EXAMPLES = r'''
+- name: "fetch auth token"
+  raw: influx auth list --user my-user --hide-headers | cut -f 3
+  register: influx_token_fetch
+  delegate_to: ed-influxdb-2
+
+- name: "create dashboard"
+  influx_dashboard:
+    base: "http://localhost:8086"
+    org: "my-org"
+    auth_token: "{{influx_token_fetch.stdout_lines[0]}}"
+    data: "{{lookup('file', 'influxdb-dashboard-cobald.json')}}"
+'''
+
+
+def parse_dashboard_template(data):
+    j = json.loads(data)
+
+    return {
+        'name': j["content"]["data"]["attributes"]["name"],
+        'cells': {i["relationships"]["view"]["data"]["id"]: i["attributes"]
+                  for i in j["content"]["included"] if i["type"] == "cell"},
+        'views': {i["id"]: i["attributes"]
+                  for i in j["content"]["included"] if i["type"] == "view"},
+    }
+
+
+def get_auth(base, org_name, token):
+    h = {"Authorization": "Token {token}".format(token=token)}
+
+    rm = requests.get("{base}/api/v2/me".format(base=base), headers=h)
+    rm.raise_for_status()
+    # me_name = rm.json()["name"]
+    me = rm.json()["id"]
+
+    ro = requests.get("{base}/api/v2/orgs".format(base=base), headers=h,
+                      json={"userID": me})
+    ro.raise_for_status()
+    org_id = [o["id"] for o in ro.json()["orgs"] if o["name"] == org_name]
+
+    return {"org_id": org_id[0], "org_name": org_name, "uid": me, "h": h}
+
+
+def check(base, auth, dashboard):
+    h = auth["h"]
+    rd = requests.get("{base}/api/v2/dashboards".format(base=base), headers=h)
+    rd.raise_for_status()
+    x = [i for i in rd.json()["dashboards"]
+         if i["name"] == dashboard["name"] and i["orgID"] == auth["org_id"]]
+    return len(x) == 0, x
+
+
+def create(base, auth, dashboard):
+    h = auth["h"]
+
+    # create dashboard
+    rd = requests.post("{base}/api/v2/dashboards".format(base=base), headers=h,
+                       json={"orgID": auth["org_id"],
+                             "name": dashboard["name"]})
+    rd.raise_for_status()
+    dash_id = rd.json()["id"]
+
+    for k, v in dashboard["cells"].items():
+        # create cells in dashboard
+        rc = requests.post(
+            "{base}/api/v2/dashboards/{dash_id}/cells".format(
+                base=base, dash_id=dash_id),
+            headers=h, json=v)
+        rc.raise_for_status()
+        cell_id = rc.json()["id"]
+
+        # create view of cell in dashboard
+        rv = requests.patch(
+             "{base}/api/v2/dashboards/{dash_id}/cells/{cell_id}/view".format(
+                base=base, dash_id=dash_id, cell_id=cell_id),
+             headers=h, json=dashboard["views"][k])
+        rv.raise_for_status()
+
+
+if __name__ == "__main__":
+    result = dict(changed=False, message="")
+    module = AnsibleModule(
+        argument_spec=dict(
+            base=dict(type="str", required=True),
+            org=dict(type="str", required=True),
+            auth_token=dict(type="str", required=True),
+            data=dict(type="json", required=True),
+            force=dict(type="bool", default=False),
+        ),
+        supports_check_mode=True
+    )
+    dashboard = parse_dashboard_template(module.params["data"])
+    auth = get_auth(module.params["base"], module.params["org"],
+                    module.params["auth_token"])
+    changed, x = check(module.params["base"], auth, dashboard)
+
+    result['changed'] = changed
+
+    if module.check_mode:
+        module.exit_json(**result)
+
+    if changed or module.params["force"]:
+        create(module.params["base"], auth, dashboard)
+    module.exit_json(**result)
--- a/roles/cobald/library/influx_token.py
+++ b/roles/cobald/library/influx_token.py
@@ -0,0 +1,232 @@
+#!/usr/bin/env python
+import requests
+from ansible.module_utils.basic import AnsibleModule
+
+DOCUMENTATION = r'''
+---
+module: influx2_token
+short_description: generate token via influxdb2 api
+description: generate token via influxdb2 api
+notes:
+    - just works with influxdb version 2
+    - needs token to authenticate against API (use
+      `influx auth list --user my-user --hide-headers | cut -f 3`
+    - tokens may not be removed
+    - permissions can not be updated. a new token is created and the old
+      one is not removed.
+options:
+    base:
+        description: URL for path, e.g. `https://localhost:8086`
+        type: str
+        required: True
+    org:
+        description: influxdb2 organisation
+        type: str
+        required: True
+    auth_token:
+        description: influxdb2 authentication token
+        type: str
+        required: True
+    key:
+        description: some key used to identify token. This is put into
+                     the tokens description
+        type: str
+        required: True
+    description:
+        description: textual description for token. key gets appended
+        type: str
+        required: False
+    permissions:
+        description: list of permissions, each dict(action, resource)
+        type: list
+        required: True
+    force:
+        description: force creation even if dashboard already exists
+            (adds a new one)
+        type: bool
+        required: False
+        default: False
+author:
+    - Thorsten M. (@thoto)
+'''
+
+EXAMPLES = r'''
+- name: "fetch auth token"
+  raw: influx auth list --user my-user --hide-headers | cut -f 3
+  register: influx_token_fetch
+  delegate_to: ed-influxdb-2
+
+- name: "create dashboard"
+  influx_token:
+    base: "http://localhost:8086"
+    org: "my-org"
+    auth_token: "{{influx_token_fetch.stdout_lines[0]}}"
+    key: "foo123"
+    description: "token for foo key"
+    permissions:
+    - action: "write"
+      resource:
+        type: "buckets"
+  register: ed-influx-token
+
+- debug: msg="Token: {{ed-influx-token.token}}"
+'''
+
+
+def get_org_id(base, org_name, h):
+    ro = requests.get("{base}/api/v2/orgs".format(base=base), headers=h,
+                      json={"org": org_name})
+    ro.raise_for_status()
+    org_id = [o["id"] for o in ro.json()["orgs"] if o["name"] == org_name]
+    return org_id[0]
+
+
+def marker(key):
+    return "__ANSIBLE_TOKEN_{key}__".format(key=key)
+
+
+def _filter_perms(perms):
+    for r in perms:
+        r["resource"] = {k: v for k, v in r["resource"].items() if v}
+    return perms
+
+
+class Token:
+    def __init__(self, base, h, data):
+        self.base = base
+        self.h = h
+        self.marker = marker(data["key"])
+
+        self.org_id = data["org_id"]
+        self.perms = _filter_perms(data["perms"])
+        self.description = data["description"]+" "+self.marker
+
+        self.result_token = None
+        self.f = None
+
+    def check(self):
+        ra = requests.get("{base}/api/v2/authorizations".format(
+                            base=self.base),
+                          params={"orgID": self.org_id},
+                          headers=self.h)
+        ra.raise_for_status()
+
+        update = None
+        for i in ra.json()["authorizations"]:
+            if self.marker not in i["description"] \
+                    or i["orgID"] != self.org_id:
+                continue
+            if self._match_perms(self.perms, i["permissions"]):
+                self.result_token = i
+                if self.description == i["description"]:
+                    return False  # everything matches -> no change needed
+                else:
+                    update = {"auth_id": i["id"],
+                              "description": self.description}
+            # TODO: may remove token because permissions do not match?
+        if update:
+            self.f = lambda: self._update(**update)
+        else:
+            self.result_token = None
+            self.f = lambda: self._create({
+                "orgID": self.org_id,
+                "description": self.description,
+                "permissions": self.perms
+            })
+        return True
+
+    def run(self):
+        if not self.f:
+            self.check()
+        self.f()
+
+    def _match_perms(self, pa, pb):
+        def g(match, lst):
+            for idx, i in enumerate(lst):
+                if i['action'] != match['action']:
+                    continue
+                for k, v in match['resource'].items():
+                    if k not in i['resource'] or i['resource'][k] != v:
+                        continue
+                else:  # first best match
+                    return idx
+            else:
+                raise ValueError
+
+        b = [b.copy() for b in pb]
+        for i in pa:
+            try:
+                b.pop(g(i, b))
+            except ValueError:
+                return False  # permission i not present in b
+
+        if b:  # not empty
+            return False  # some permissions in b not in a
+        return True
+
+    def _update(self, auth_id, description):
+        ra = requests.patch(
+               "{base}/api/v2/authorizations/{auth_id}".format(
+                   base=self.base, auth_id=auth_id),
+               headers=self.h, json={"description": description})
+        ra.raise_for_status()
+        return ra
+
+    def _create(self, data):
+        ra = requests.post("{base}/api/v2/authorizations".format(
+                          base=self.base),
+                          headers=self.h, json=data)
+        ra.raise_for_status()
+        self.result_token = ra.json()
+        return ra
+
+
+if __name__ == "__main__":
+    result = dict(changed=False, message="")
+    module = AnsibleModule(
+        argument_spec=dict(
+            base=dict(type="str", required=True),
+            org=dict(type="str", required=True),
+            auth_token=dict(type="str", required=True),
+            key=dict(type="str", required=True),
+            description=dict(type="str", default=""),
+            permissions=dict(type="list", elements='dict', options=dict(
+                action=dict(type='str', choices=['read', 'write'],
+                            required=True),
+                resource=dict(type='dict', options=dict(
+                    id=dict(type='str'),
+                    name=dict(type='str'),
+                    org=dict(type='str'),
+                    orgID=dict(type='str'),
+                    type=dict(type='str', required=True, choices=[
+                        "authorizations", "buckets", "dashboards", "orgs",
+                        "sources", "tasks", "telegrafs", "users", "variables",
+                        "scrapers", "secrets", "labels", "views", "documents",
+                        "notificationRules", "notificationEndpoints", "checks",
+                        "dbrp", "flows", "annotations", "functions"]),
+                ), required=True),
+            ), required=True),
+            force=dict(type="bool", default=False),
+        ),
+        supports_check_mode=True
+    )
+    h = {"Authorization": "Token {token}".format(
+        token=module.params["auth_token"])}
+    t = Token(module.params["base"], h, {
+        "org_id": get_org_id(module.params["base"], module.params["org"], h),
+        "key": module.params["key"],
+        "perms": module.params["permissions"],
+        "description": module.params["description"]})
+
+    changed = t.check()
+    if t.result_token:
+        result['token'] = t.result_token["token"]
+    result['changed'] = changed
+
+    if module.check_mode:
+        module.exit_json(**result)
+
+    if changed or module.params["force"]:
+        t.run()
+        result['token'] = t.result_token["token"]
+    module.exit_json(**result)
--- a/roles/cobald/tasks/dockerimage-generic.yml
+++ b/roles/cobald/tasks/dockerimage-generic.yml
@@ -0,0 +1,34 @@
+- file:
+    path: "/container/docker-images/cobald.{{cobald_image_tag|default('latest')}}/"
+    state: directory
+    owner: "{{unpriv_user}}"
+    group: docker
+
+- template:
+    src: cobald.Dockerfile
+    dest: "/container/docker-images/cobald.{{cobald_image_tag|default('latest')}}/Dockerfile"
+    owner: "{{unpriv_user}}"
+    group: docker
+  register: cobald_cp_dockerfile
+
+- copy:
+    src: "{{item}}"
+    dest: "/container/docker-images/cobald.{{cobald_image_tag|default('latest')}}/{{item}}"
+    owner: "{{unpriv_user}}"
+    group: docker
+    mode: 0755
+  with_items:
+  - cobald-entrypoint.sh
+  - init-cobaldmodules.sh
+  register: cobald_cp_files
+
+- docker_image:
+    name: "cobald"
+    tag: "{{cobald_image_tag|default('latest')}}"
+#    pull: False
+    build:
+      pull: False
+      path: "/container/docker-images/cobald.{{cobald_image_tag|default('latest')}}/"
+    source: build
+    force_source: "{{cobald_cp_dockerfile.changed or cobald_cp_files.changed}}"
+
--- a/roles/cobald/tasks/facts.yml
+++ b/roles/cobald/tasks/facts.yml
@@ -0,0 +1,11 @@
+- block:
+  - docker_container_info:
+      name: "{{ container_name | mandatory }}"
+    register: cobald_container_info
+
+  - set_fact:
+      cobald_container_hostname: |-
+        {{cobald_container_info.container.Config.Hostname | default('cobald-'+
+          lookup('password', '/dev/null chars=ascii_lowercase length=6')) }}
+  when: cobald_container_hostname is not defined
+
--- a/roles/cobald/tasks/grafana.yml
+++ b/roles/cobald/tasks/grafana.yml
@@ -0,0 +1,54 @@
+- name: create influx token for grafana
+  influx_token:
+    base: "http://localhost:{{influx_pubport}}"
+    org: "my-org"
+    auth_token: "{{influx_admin_token}}"
+    description: grafana read access
+    key: grafana
+    permissions:
+    - action: read
+      resource:
+        type: buckets
+  register: influx_grafana_token
+
+- name: run grafana
+  docker_container:
+    name: ed-grafana
+    image: grafana/grafana:7.5.7
+    hostname: ed-grafana
+    domainname: cobald.local
+    networks:
+    - name: "{{docker_network}}"
+    published_ports:
+    - "3000:3000"
+    state: started
+    detach: True
+    cleanup: True
+
+- wait_for:
+    host: localhost
+    port: 3000
+
+- community.grafana.grafana_datasource:
+    grafana_url: http://localhost:3000
+    grafana_user: admin
+    grafana_password: admin
+    name: InfluxDB
+    ds_type: influxdb
+    ds_url: "{{influx_url}}"
+    additional_json_data:
+      defaultBucket: "{{influx_bucket}}"
+      organization: "{{influx_org}}"
+      version: Flux
+    additional_secure_json_data:
+      token: "{{influx_grafana_token.token}}"
+  register: das
+
+- community.grafana.grafana_dashboard:
+    grafana_url: http://localhost:3000
+    grafana_user: admin
+    grafana_password: admin
+    state: present
+    commit_message: updated by ansible
+    overwrite: yes
+    json_data: "{{lookup('file', 'grafana-dashboard.json')|from_json}}"
--- a/roles/cobald/tasks/influxdb.yml
+++ b/roles/cobald/tasks/influxdb.yml
@@ -0,0 +1,76 @@
+- name: run influxdb in docker container
+  docker_container:
+    name: ed-influxdb
+    image: influxdb
+    hostname: "{{influx_hostname}}"
+    domainname: "{{influx_domainname}}"
+    networks:
+    - name: "{{ docker_network }}"
+    published_ports:
+    - "{{influx_pubport}}:8086"
+    volumes:
+    - "ed-influxdb-data:/var/lib/influxdb2"
+    - "ed-influxdb-config:/etc/influxdb2"
+    - "/container/volumes/influxdb-backup/:/backup"
+    env:
+      DOCKER_INFLUXDB_INIT_MODE: setup
+      DOCKER_INFLUXDB_INIT_USERNAME: "{{influx_admin_user}}"
+      DOCKER_INFLUXDB_INIT_PASSWORD: "{{influx_admin_pw}}"
+      DOCKER_INFLUXDB_INIT_ORG: "{{influx_org}}"
+      DOCKER_INFLUXDB_INIT_BUCKET: my-bucket
+    state: started
+    detach: True
+    cleanup: True
+    networks_cli_compatible: True
+
+- name: add ansible connection to influxdb container
+  add_host:
+    name: ed-influxdb
+    ansible_connection: docker
+    ansible_docker_extra_args: "-H=ssh://{{ansible_host}}"
+  changed_when: False
+
+- name: wait for influx to run
+  raw: until curl http://localhost:8086 ; do sleep 1 ; done
+  changed_when: False
+  delegate_to: ed-influxdb
+
+- name: fetch influxdb auth token
+  raw: influx auth list --user my-user --hide-headers --json
+  register: influx_token_fetch
+  changed_when: False
+  delegate_to: ed-influxdb
+
+- name: set influxdb admin token
+  set_fact:
+    influx_admin_token:
+      "{{(influx_token_fetch.stdout | from_json | first).token}}"
+
+- name: create influxdb bucket for cobald
+  influx_bucket:
+    base: "http://localhost:{{influx_pubport}}"
+    org: "my-org"
+    auth_token: "{{influx_admin_token}}"
+    name: "{{influx_bucket}}"
+
+- name: create influxdb dashboard
+  influx_dashboard:
+    base: "http://localhost:{{influx_pubport}}"
+    org: "my-org"
+    auth_token: "{{influx_admin_token}}"
+    data: "{{lookup('file', 'influxdb-dashboard-cobald.json')}}"
+  when: influxdb_dashboard | default(True)
+
+- name: create influxdb write access token for telegraf
+  influx_token:
+    base: "http://localhost:{{influx_pubport}}"
+    org: "my-org"
+    auth_token: "{{influx_admin_token}}"
+    description: cobald tardis telegraf monitoring plugin
+    key: telegraf_cobaldtardis
+    permissions:
+    - action: write
+      resource:
+        type: buckets
+        name: "{{influx_bucket}}"
+  register: influx_telegraf_token
--- a/roles/cobald/tasks/main.yml
+++ b/roles/cobald/tasks/main.yml
@@ -0,0 +1,136 @@
+- include_vars: cobald-slurm.yml
+  when: cobald_slurm | default(False)
+
+- name: build cobald:slurm docker image
+  include_role:
+    name: slurm
+    tasks_from: dockerimage
+  vars:
+    slurm_image_prefix: cobald
+    image_name: "{{cobald_image_tag}}"
+    dockerfile: "{{ lookup('template', 'cobald.Dockerfile') }}"
+    files:
+    - dest: cobald-entrypoint.sh
+      content: "{{ lookup('file', 'cobald-entrypoint.sh') }}"
+    - dest: init-cobaldmodules.sh
+      content: "{{ lookup('file', 'init-cobaldmodules.sh') }}"
+  when: cobald_slurm | default(False)
+
+- name: build generic cobald docker image
+  include_tasks: dockerimage-generic.yml
+  when: not (cobald_slurm | default(False))
+
+- name: make cobald data volume
+  file:
+    path: "/container/volumes/cobald/"
+    state: directory
+    owner: "{{unpriv_user}}"
+    group: docker
+
+- name: copy cobald config (does nothing yet)
+  copy:
+    src: cobald-config/
+    dest: /container/volumes/cobald
+    owner: "{{unpriv_user}}"
+    group: docker
+  when: False
+
+- name: ensure network for cobald container exists
+  docker_network:
+    name: "{{cobald_docker_network}}"
+    state: present
+
+# docker run -v $(pwd)/cobald-config-host:/etc/cobald -v $(pwd)/cobald:/cobald --rm -it cobald  bash
+
+- name: install git
+  yum:
+    name: git
+    state: present
+
+- name: make directories for cobald configuration and modules
+  file:
+    path: "{{item}}"
+    owner: "{{unpriv_user}}"
+    group: "{{unpriv_user}}"
+    mode: "a=rx,u=rwx"
+    state: directory
+  with_items:
+  - "~{{unpriv_user}}/cobald/modules"
+  - "~{{unpriv_user}}/cobald"
+
+- name: clone cobald code from git
+  git:
+    repo: https://github.com/thoto/cobald
+    dest: "~{{unpriv_user}}/cobald-src"
+    version: bugfix/mixed_construction_methods
+    update: no
+  become: yes
+  become_user: "{{unpriv_user}}"
+  register: cobald_git_pull
+
+- name: clone tardis code from git
+  git:
+    repo: https://github.com/MatterMiners/tardis
+    dest: "~{{unpriv_user}}/tardis-src"
+    version: master
+  become: yes
+  become_user: "{{unpriv_user}}"
+  register: tardis_git_pull
+
+- name: "get unpriv_user {{unpriv_user}} uid and gid"
+  getent:
+    database: passwd
+    key: "{{unpriv_user}}"
+
+- name: run pip install on cobald and tardis
+  docker_container:
+    image: "cobald:{{cobald_image_tag|default('latest')}}"
+    name: "cobald-src-{{item.name}}-install"
+    volumes:
+    - "~{{unpriv_user}}/{{item.name}}-src:/usr/local/src/{{item.name}}:rw"
+    state: started
+    detach: False
+    cleanup: True
+    user: "{{getent_passwd[unpriv_user][1]}}:{{getent_passwd[unpriv_user][2]}}"
+    entrypoint: ""
+    command: |
+      bash -c 'HOME=/tmp pip3 install --editable /usr/local/src/{{item.name}}'
+  with_items:
+  - name: cobald
+    run: "{{cobald_git_pull.changed}}"
+  - name: tardis
+    run: "{{tardis_git_pull.changed}}"
+  when: item.run
+
+- import_tasks: telegraf.yml
+
+- name: get cobald hostname
+  include_tasks: facts.yml
+  when: cobald_container_hostname is not defined
+
+- name: run cobald container
+  docker_container:
+    name: "{{ container_name | default('cobald') }}"
+    image: "cobald:{{cobald_image_tag|default('latest')}}"
+    hostname: "{{cobald_container_hostname}}"
+    domainname: "{{ cobald_domainname | default('cobald.local')}}"
+    volumes: "{{default_mounts + cobald_mounts }}"
+    networks:
+    - name: "{{cobald_docker_network}}"
+    networks_cli_compatible: True
+#    env:
+#      slurmuser: "{{slurm_user}}"
+#    privileged: "{{ container_privileged | bool }}"
+    state: started
+    detach: True
+    cleanup: True
+    interactive: True
+    # command: python3 -m cobald.daemon /etc/cobald/config.yaml
+  vars:
+    default_mounts: "{{cobald_slurm_mounts | default([])}}"
+    cobald_mounts:
+    - "~{{unpriv_user}}/cobald:/etc/cobald"
+    # - /container/volumes/cobald:/etc/cobald:ro
+    - "~{{unpriv_user}}/cobald/modules:/usr/local/src/cobaldmodules"
+    - "~{{unpriv_user}}/cobald-src:/usr/local/src/cobald:ro"
+    - "~{{unpriv_user}}/tardis-src:/usr/local/src/tardis:ro"
--- a/roles/cobald/tasks/telegraf.yml
+++ b/roles/cobald/tasks/telegraf.yml
@@ -0,0 +1,68 @@
+- name: setup directories for telegraf
+  file:
+    path: "/container/{{item}}/telegraf/"
+    state: directory
+    owner: "{{unpriv_user}}"
+    group: docker
+  loop:
+  - docker-images
+  - volumes
+
+- name: copy telegraf Dockerfile
+  copy:
+    src: telegraf.Dockerfile
+    dest: /container/docker-images/telegraf/Dockerfile
+    owner: "{{unpriv_user}}"
+    group: docker
+  register: cobald_cp_telegraf_dockerfile
+
+- name: copy telegraf repo file
+  copy:  # telegraf is found in influxdb repo
+    src: influxdb.repo
+    dest: /container/docker-images/telegraf/influxdb.repo
+    owner: "{{unpriv_user}}"
+    group: docker
+
+- name: docker image for telegraf
+  docker_image:
+    name: "ed-telegraf"
+    build:
+      pull: False
+      path: "/container/docker-images/telegraf/"
+    source: build
+    force_source: "{{cobald_cp_telegraf_dockerfile.changed}}"
+
+- import_tasks: influxdb.yml
+  tags: influxdb
+
+- name: generate telegraf config
+  template:
+    src: telegraf.conf.j2
+    dest: /container/volumes/telegraf/telegraf.conf
+    owner: "{{unpriv_user}}"
+    group: docker
+  vars:
+    influx_token: "{{influx_telegraf_token.token}}"
+    influx_url: "http://{{influx_hostname}}:8086"
+  register: telegraf_config_gen
+
+- name: run telegraf container
+  docker_container:
+    name: ed-telegraf
+    image: ed-telegraf
+    hostname: telegraf
+    domainname: cobald.local
+    networks:
+    - name: "{{docker_network | default('bridge') }}"
+    volumes:
+    - "/container/volumes/telegraf/telegraf.conf:/etc/telegraf/telegraf.conf:ro"
+    state: started
+    recreate: "{{ telegraf_config_gen.changed | default(False) | bool }}"
+    detach: True
+#   cleanup: True
+    networks_cli_compatible: True
+
+- import_tasks: grafana.yml
+  vars:
+    influx_url: "http://{{influx_hostname}}:8086"
+  tags: influxdb
--- a/roles/cobald/templates/cobald.Dockerfile
+++ b/roles/cobald/templates/cobald.Dockerfile
@@ -0,0 +1,69 @@
+FROM {{ cobald_docker_base_image | default("docker.io/library/centos:7") }}
+
+RUN yum update -y && \
+	yum install -y python3 git && pip3 install --upgrade pip && \
+    yum clean all && rm -rf /var/cache/yum
+
+ARG REPOCOBALD=https://github.com/MatterMiners/cobald
+ARG REPOTARDIS=https://github.com/MatterMiners/tardis
+
+RUN git clone $REPOCOBALD /usr/local/src/cobald && \
+	git clone $REPOTARDIS /usr/local/src/tardis
+
+RUN mkdir /etc/cobald /var/log/cobald && \
+	( getent passwd cobald > /dev/null || \
+		useradd -m -d /var/lib/cobald --no-log-init --system cobald ) && \
+	chown cobald:cobald /var/log/cobald
+
+#RUN mkdir /cobald && python3 -m venv /cobald && source /cobald/bin/activate &&\
+#	pip3 install --upgrade pip && pip3 install cobald
+
+RUN mkdir /usr/local/src/cobaldmodules /usr/local/lib/cobaldmodules && \
+	 ln -s /usr/local/src/cobaldmodules/setup.py \
+		/usr/local/lib/cobaldmodules/setup.py && \
+	 ln -s /usr/local/src/cobaldmodules/cobaldmodules \
+		/usr/local/lib/cobaldmodules/cobaldmodules && \
+	 chown cobald:cobald /usr/local/lib/cobaldmodules
+
+RUN pip3 install --editable /usr/local/src/cobald  && \
+	pip3 install --editable /usr/local/src/cobald[contrib]
+
+RUN pip3 install --editable /usr/local/src/tardis&& \
+	pip3 install --editable /usr/local/src/tardis[contrib]
+
+ENV PYTHONPATH=/usr/local/src/cobaldmodules
+
+# pip3 install --editable .
+# pip3 install --editable .[contrib]
+# pip3 install --upgrade --editable /etc/cobald/modules/
+# su cobald -c "python3 -m cobald.daemon /etc/cobald/config.yaml"
+
+VOLUME /usr/local/src/cobaldmodules
+
+VOLUME /etc/cobald
+
+RUN mkdir -p /usr/local/lib/entrypoints.d/
+
+COPY init-cobaldmodules.sh /usr/local/lib/entrypoints.d/50-init-cobaldmodules.sh
+
+RUN chmod 755 /usr/local/lib/entrypoints.d/50-init-cobaldmodules.sh
+
+RUN echo -e "#!/bin/sh\npython3 -m cobald.daemon /etc/cobald/config.yaml" >> /etc/docker-init.d/70-cobald && chmod 755 /etc/docker-init.d/70-cobald
+
+{% if cobald_docker_default_command | default(True) -%}
+COPY cobald-entrypoint.sh /usr/local/sbin/cobald-entrypoint.sh
+
+RUN chmod 755 /usr/local/sbin/cobald-entrypoint.sh
+
+ENTRYPOINT [ "/usr/local/sbin/cobald-entrypoint.sh" ]
+
+RUN yum -y install iproute &&\
+    yum clean all && rm -rf /var/cache/yum
+
+USER cobald
+
+STOPSIGNAL SIGINT
+
+# CMD "python3 -m cobald.daemon /etc/cobald/config.yaml"
+CMD /etc/docker-init.d/60-cobald
+{%- endif %}
--- a/roles/cobald/templates/telegraf.conf.j2
+++ b/roles/cobald/templates/telegraf.conf.j2
--- a/roles/cobald/vars/cobald-slurm.yml
+++ b/roles/cobald/vars/cobald-slurm.yml
@@ -0,0 +1,12 @@
+cobald_image_tag: slurm
+cobald_docker_base_image: "slurm:slurmd"
+cobald_docker_default_command: False
+cobald_docker_network: "{{slurm.network}}"
+cobald_domainname: "{{slurm.domain}}"
+cobald_slurm_mounts: "{{slurm.mounts}}"
+#- /container/volumes/slurm/:/etc/slurm/:rw
+##- "{{slurm_cfg_path | mandatory}}:/etc/slurm/:rw"
+#- /container/volumes/munge/munge.key:/etc/munge/munge.key:rw
+## - "{{slurm_munge_path | mandatory}}:/etc/munge/munge.key:rw"
+#- slurm-shared:/shared/:rw
+## - "{{slurm_shared_path | mandatory}}:{{slurm_shared_target | default('/shared')}}:rw"
--- a/roles/cobald/vars/main.yml
+++ b/roles/cobald/vars/main.yml
@@ -0,0 +1,3 @@
+cobald_docker_network: "{{docker_network}}"
+influx_domainname: "{{ cobald_domainname }}"
+influx_hostname: "ed-influxdb"
--- a/roles/slurm/defaults/main.yml
+++ b/roles/slurm/defaults/main.yml
@@ -1,5 +1,9 @@
-container_privileged: False
 slurm_user: slurm
 slurm_log_path_ctld: /var/log/slurm/slurmctld.log
 slurm_log_path_d: /var/log/slurm/slurmd.log
 slurm_log_path_sched: /var/log/slurm/slurmsched.log
+slurm_prefix: slurm
+slurm_domain: slurm.local
+container_privileged: False
+docker_network: slurm
+slurm_image_prefix: slurm
--- a/roles/slurm/files/docker-init
+++ b/roles/slurm/files/docker-init
@@ -0,0 +1,18 @@
+#!/bin/bash
+function trp_term(){
+        echo pkill -P $pids
+        for j in $pids ; do
+                pkill -P $j
+                kill -SIGTERM $j
+        done
+}
+trap trp_term SIGINT SIGTERM
+pids=""
+for i in /etc/docker-init.d/* ; do
+        [ ! -f $i ] && break
+        $i &
+        pids="$pids $!"
+done
+wait $pids
+
+# TODO: call start scripts like "foo.sh start" and "foo.sh stop" to avoid pkill
--- a/roles/slurm/files/entry-munge.sh
+++ b/roles/slurm/files/entry-munge.sh
@@ -0,0 +1,7 @@
+#!/usr/bin/env bash
+set -e
+
+if [ -f "/etc/munge/munge.key" ] ; then
+	chown munge:munge /etc/munge/munge.key
+	chmod 600 /etc/munge/munge.key
+fi
--- a/roles/slurm/files/entrypoint.sh
+++ b/roles/slurm/files/entrypoint.sh
@@ -1,9 +1,8 @@
 #!/usr/bin/env bash
 set -e

-if [ -f "/etc/munge/munge.key" ] ; then
-	chown munge:munge /etc/munge/munge.key
-	chmod 600 /etc/munge/munge.key
-fi
+for i in /usr/local/lib/entrypoints.d/* ; do
+    [ -f $i ] && /bin/sh $i || break
+done

-exec "$@"
+exec "${@:-/bin/bash}"
--- a/roles/slurm/files/slurm-base.Dockerfile
+++ b/roles/slurm/files/slurm-base.Dockerfile
@@ -0,0 +1,35 @@
+FROM docker.io/library/centos:7
+
+RUN yum install -y epel-release && \
+	yum install -y slurm && \
+	yum clean all && rm -rf /var/cache/yum
+
+RUN	yum install -y less iproute bind-utils nmap-ncat net-tools && \
+	yum clean all && rm -rf /var/cache/yum
+
+RUN mkdir -p /usr/local/lib/entrypoints.d/
+
+COPY --chown=root:root entry-munge.sh /usr/local/lib/entrypoints.d/10-munge.sh
+COPY --chown=root:root entrypoint.sh /usr/local/sbin/entrypoint.sh
+
+RUN chmod 755 /usr/local/lib/entrypoints.d/10-munge.sh && \
+	chmod 755 /usr/local/sbin/entrypoint.sh
+
+ENTRYPOINT [ "/usr/local/sbin/entrypoint.sh" ]
+
+ARG slurmuser=slurm
+ENV slurmuser=${slurmuser}
+
+RUN useradd -d /var/lib/slurm -m --no-log-init --system $slurmuser &&\
+	slurm-setuser -u $slurmuser -g $slurmuser -y
+
+COPY docker-init /usr/local/sbin/docker-init
+RUN mkdir /etc/docker-init.d  && chmod 755 /usr/local/sbin/docker-init
+COPY start-scripts/10-munge /etc/docker-init.d/10-munge
+RUN chmod 755 /etc/docker-init.d/10-munge
+
+ARG moreusers
+RUN function mu { [ -z "$1" ] || useradd -d $2 -m --no-log-init --system $1 ;};\
+	echo "${moreusers}" | tr ',' '\n' | while read i ; do mu $i ; done
+
+CMD /usr/local/sbin/docker-init
--- a/roles/slurm/files/slurmctld.Dockerfile
+++ b/roles/slurm/files/slurmctld.Dockerfile
@@ -1,32 +1,11 @@
-FROM docker.io/library/centos:7 as base
-
-RUN yum install -y epel-release && \
-	yum install -y slurm && \
-	yum clean all && rm -rf /var/cache/yum
-
-RUN	yum install -y less iproute bind-utils nmap-ncat net-tools && \
-	yum clean all && rm -rf /var/cache/yum
-
-COPY entrypoint.sh /usr/local/sbin/entrypoint.sh
-
-RUN chown root:root /usr/local/sbin/entrypoint.sh && \
-	chmod 755 /usr/local/sbin/entrypoint.sh
-
-ENTRYPOINT [ "/usr/local/sbin/entrypoint.sh" ]
-
-ARG slurmuser=slurm
-ENV slurmuser=${slurmuser}
-
-RUN useradd -d /var/lib/slurm -m --no-log-init --system $slurmuser &&\
-	slurm-setuser -u $slurmuser -g $slurmuser -y
+FROM slurm:base

 RUN	yum install -y slurm-slurmctld && \
 	yum clean all && rm -rf /var/cache/yum

+COPY start-scripts/20-slurmctld /etc/docker-init.d/20-slurmctld
+RUN chmod 755 /etc/docker-init.d/20-slurmctld
+
 ENV SLURMCTLD_LOG_PATH="/var/log/slurm/slurmctld.log"
 ENV SLURMD_LOG_PATH="/var/log/slurm/slurmd.log"
 ENV SLURM_SCHED_LOG_PATH="/var/log/slurm/slurmsched.log"
-
-CMD bash -c 'cat <({ su -s /bin/sh -c "munged -F" munge & \
-	su -s /bin/sh -c "slurmctld -D" ${slurmuser} 2>/dev/null 1>/dev/null & \
-	tail --retry --pid $! -f ${SLURMCTLD_LOG_PATH} ${SLURM_SCHED_LOG_PATH} & })'
--- a/roles/slurm/files/slurmd.Dockerfile
+++ b/roles/slurm/files/slurmd.Dockerfile
@@ -1,32 +1,11 @@
-FROM docker.io/library/centos:7
-
-RUN yum install -y epel-release && \
-	yum install -y slurm && \
-	yum clean all && rm -rf /var/cache/yum
-
-RUN	yum install -y less iproute bind-utils nmap-ncat net-tools && \
-	yum clean all && rm -rf /var/cache/yum
-
-COPY entrypoint.sh /usr/local/sbin/entrypoint.sh
-
-RUN chown root:root /usr/local/sbin/entrypoint.sh && \
-	chmod 755 /usr/local/sbin/entrypoint.sh
-
-ENTRYPOINT [ "/usr/local/sbin/entrypoint.sh" ]
-
-ARG slurmuser=slurm
-ENV slurmuser=${slurmuser}
-
-RUN useradd -d /var/lib/slurm -m --no-log-init --system $slurmuser &&\
-	slurm-setuser -u $slurmuser -g $slurmuser -y
+FROM slurm:base

 RUN yum install -y slurm-slurmd && \
 	yum clean all && rm -rf /var/cache/yum

+COPY start-scripts/30-slurmd /etc/docker-init.d/30-slurmd
+RUN chmod 755 /etc/docker-init.d/30-slurmd
+
 ENV SLURMCTLD_LOG_PATH="/var/log/slurm/slurmctld.log"
 ENV SLURMD_LOG_PATH="/var/log/slurm/slurmd.log"
 ENV SLURM_SCHED_LOG_PATH="/var/log/slurm/slurmsched.log"
-
-CMD bash -c 'cat <({ su -s /bin/sh -c "munged -F" munge & \
-	slurmd -D 2>/dev/null 1>/dev/null & \
-	tail --retry --pid $! -f ${SLURMD_LOG_PATH} ${SLURM_SCHED_LOG_PATH} & })'
--- a/roles/slurm/files/start-scripts/10-munge
+++ b/roles/slurm/files/start-scripts/10-munge
@@ -0,0 +1,2 @@
+#!/bin/sh
+exec su -s /bin/sh -c "munged -F" munge
--- a/roles/slurm/files/start-scripts/20-slurmctld
+++ b/roles/slurm/files/start-scripts/20-slurmctld
@@ -0,0 +1,4 @@
+#!/bin/sh
+su -s /bin/sh -c "slurmctld -D" ${slurmuser} 2>/dev/null 1>/dev/null &
+tail --retry --pid $! -f ${SLURMCTLD_LOG_PATH} ${SLURM_SCHED_LOG_PATH}
+
--- a/roles/slurm/files/start-scripts/30-slurmd
+++ b/roles/slurm/files/start-scripts/30-slurmd
@@ -0,0 +1,4 @@
+#!/bin/sh
+slurmd -D 2>/dev/null 1>/dev/null &
+tail --retry --pid $! -f ${SLURMD_LOG_PATH} ${SLURM_SCHED_LOG_PATH}
+
--- a/roles/slurm/handlers/main.yml
+++ b/roles/slurm/handlers/main.yml
@@ -1,3 +1,3 @@
 - name: reconfigure slurm 
-  command:
-    cmd: docker container exec -it slurm-ctl scontrol reconfigure
+  shell:
+    cmd: "docker container exec -it {{slurm_prefix}}-ctl scontrol reconfigure || docker container restart {{slurm_prefix}}-ctl && docker container exec -it {{slurm_prefix}}-ctl scontrol reconfigure"
--- a/roles/slurm/tasks/dockerimage.yml
+++ b/roles/slurm/tasks/dockerimage.yml
@@ -1,31 +1,29 @@
- file:
-    path: "/container/docker-images/{{item}}"
-    state: directory
-    owner: "{{unpriv_user}}"
-    group: docker
+- name: build slurm base docker image
+  include_tasks: dockerimage_build.yml
+  vars:
+    slurm_image_prefix: "{{slurm_base_image_prefix | default('slurm') }}"
+    image_name: base
+    dockerfile: "{{lookup('file', 'slurm-base.Dockerfile')}}"
+    files:
+    - dest: entrypoint.sh
+      content: "{{ lookup('file', 'entrypoint.sh') }}"
+    - dest: entry-munge.sh
+      content: "{{ lookup('file', 'entry-munge.sh') }}"
+    - dest: docker-init
+      content: "{{ lookup('file', 'docker-init') }}"
+    - dest: start-scripts/10-munge
+      content: "{{ lookup('file', 'start-scripts/10-munge') }}"
+    image_args:
+      moreusers: >-
+        {% for a in slurm_user_accounts | default([]) -%}
+        {{a['name']}} {{a['dir']}}{{loop.last | ternary('',',')}}
+        {%- endfor %}
+  when: not slurm_baseimg_build_chg | default(False)

- copy:
-    src: "{{item}}.Dockerfile"
-    dest: "/container/docker-images/{{item}}/Dockerfile"
-    owner: "{{unpriv_user}}"
-    group: docker
-  register: slurm_cp_dockerfile
-
- copy:
-    src: "entrypoint.sh"
-    dest: "/container/docker-images/{{item}}/entrypoint.sh"
-    owner: root
-    group: root
-    mode: u=rwx,g=rx,o=rx
-  register: slurm_cp_entrypt
-
- docker_image:
-    name: "slurm-{{item}}"
-#    pull: False
-    build:
-      pull: False
-      path: "/container/docker-images/{{item}}"
-#      target: "{{item}}" # unsupported on old docker-py versions as in el7 
-    source: build
-    force_source: "{{slurm_cp_dockerfile.changed or slurm_cp_entrypt.changed}}"
+- set_fact:
+    slurm_baseimg_build_chg:
+      "{{(slurm_baseimg_build_chg | default(False)) or
+         slurm_img_build.changed}}"

+- name: "build slurm base docker image {{image_name}}"
+  include_tasks: dockerimage_build.yml
--- a/roles/slurm/tasks/dockerimage_build.yml
+++ b/roles/slurm/tasks/dockerimage_build.yml
@@ -0,0 +1,43 @@
+- name: create directories for docker image build
+  file:
+    path: "/container/docker-images/{{slurm_image_prefix}}-{{image_name}}/{{item}}"
+    state: directory
+    owner: "{{unpriv_user}}"
+    group: docker
+  loop: "{{ [''] + (files | map(attribute='dest') | map('dirname') |
+                    unique | select | list) }}"
+
+- name: "copy Dockerfile {{slurm_image_prefix}}:{{image_name}}"
+  copy:
+    content: "{{dockerfile}}"
+    dest: "/container/docker-images/{{slurm_image_prefix}}-{{image_name}}/Dockerfile"
+    owner: "{{unpriv_user}}"
+    group: docker
+  register: slurm_cp_dockerfile
+
+- name: copy requisite files
+  copy:
+    content: "{{ item.content }}"
+    dest: "/container/docker-images/{{slurm_image_prefix}}-{{image_name}}/{{item.dest}}"
+    owner: root
+    group: root
+    mode: u=rwx,g=rx,o=rx
+  loop: "{{ files | default([]) }}"
+  loop_control:
+    label: "{{ item.dest }}"
+  register: slurm_cp_files
+
+- name: "build docker image {{slurm_image_prefix}}:{{image_name}}"
+  docker_image:
+    name: "{{slurm_image_prefix}}"
+    tag: "{{image_name}}"
+#    pull: False
+    build:
+      args: "{{image_args | default(omit)}}"
+      pull: False
+      path: "/container/docker-images/{{slurm_image_prefix}}-{{image_name}}/"
+    source: build
+    force_source: "{{slurm_cp_dockerfile.changed or
+                     slurm_cp_files.changed or
+                     slurm_baseimg_build_chg | default(False) }}"
+  register: slurm_img_build
--- a/roles/slurm/tasks/main.yml
+++ b/roles/slurm/tasks/main.yml
@@ -3,10 +3,27 @@
    name: [ slurm, slurm-doc ]
    state: present

- include_tasks: dockerimage.yml
+- name: build docker images for slurm
+  include_tasks:
+    file: dockerimage.yml
  loop:
-    - slurmctld
-    - slurmd
+  - name: slurmctld
+    dockerfile: "{{ lookup('file', 'slurmctld.Dockerfile') }}"
+    files:
+    - dest: start-scripts/20-slurmctld
+      content: "{{ lookup('file', 'start-scripts/20-slurmctld') }}"
+  - name: slurmd
+    dockerfile: "{{ lookup('file', 'slurmd.Dockerfile') }}"
+    files:
+    - dest: start-scripts/30-slurmd
+      content: "{{ lookup('file', 'start-scripts/30-slurmd') }}"
+  vars:
+    image_name: "{{image.name | default(omit) }}"
+    dockerfile: "{{image.dockerfile | default(omit) }}"
+    files: "{{image.files | default(omit) }}"
+  loop_control:
+    loop_var: image
+    label: "{{ image.name }}"

 - name: generate munge key
  shell:
@@ -20,7 +37,8 @@
    group: munge
    mode: u=rw,g=,o=

- file:
+- name: create munge key directory for containers
+  file:
    path: /container/volumes/munge
    state: directory
    owner: munge
@@ -35,7 +53,8 @@
    src: /etc/munge/munge.key
    dest: /container/volumes/munge/munge.key

- file:
+- name: make slurm directory
+  file:
    path: /container/volumes/slurm/
    state: directory

@@ -47,12 +66,14 @@
  loop:
    - slurm.conf
    - cgroup.conf
+  vars:
+    alloc_nodes: "{{ [ slurm_prefix+'-submit1' ] + extra_nodes | default([])}}"
  notify: reconfigure slurm
  tags: [ slurm-config ]

 - name: "create docker network to make service discovery work"
  docker_network:
-    name: slurm
+    name: "{{ docker_network }}"
    state: present

 - name: "create docker volume for shared access between nodes"
@@ -60,28 +81,19 @@
    name: slurm-shared
    state: present

- set_fact:
-    slurm_nodes: # default nodes: controller and submit machine
-    - machine: ctl
-      image: slurm-slurmctld
-    - machine: submit1
-      image: slurm-slurmd
-      extra_mounts:
-      - "/home/{{unpriv_user}}/job3/:/mnt/:rw"
-  tags: [ slurm-config ]
-    
 # TODO: reserve some address using docker_network_info and assign as aux
 # address to enable slurmctld to get a static address in order to be
 # reachable from slurm running on docker host to enable submitting jobs.

 - name: run slurm docker containers
  docker_container:
-    name: "slurm-{{item.machine}}"
-    hostname: "slurm-{{item.machine}}"
-    domainname: "slurm.local"
-    volumes: "{{default_mounts + ( item.extra_mounts | default([]) ) }}"
+    name: "{{ slurm_prefix }}-{{ item.machine }}"
+    hostname: "{{ slurm_prefix }}-{{ item.machine }}"
+    domainname: "{{ slurm_domain }}"
+    volumes: "{{ slurm_default_mounts + ( item.extra_mounts | default([]) ) }}"
+    ports: "{{ item.exposed_ports | default([]) }}"
    networks:
-    - name: "slurm"
+    - name: "{{ docker_network }}"
    env:
      slurmuser: "{{ slurm_user }}"
    image: "{{ item.image }}"
@@ -90,16 +102,20 @@
    cleanup: True
    privileged: "{{ container_privileged | bool }}"
    networks_cli_compatible: True
-  vars:
-    default_mounts:
-    - /container/volumes/slurm/:/etc/slurm/:rw
-    - /container/volumes/munge/munge.key:/etc/munge/munge.key:rw
-    - slurm-shared:/shared/:rw
-    slurm_nodes_all: | # add execute nodes
-      {% for i in range(1, 4) -%}
-        {% set _ = slurm_nodes.extend([
-          {'machine':'exec%s'|format(i), 'image': 'slurm-slurmd'}]) -%}
-      {%- endfor %}
-      {{ slurm_nodes }}
+    interactive: True
+  vars:  # see vars/main.yml
+    slurm_nodes_all: "{{ slurm_nodes_exec + slurm_nodes_std }}"
  loop: "{{ slurm_nodes_all }}"
+  loop_control:
+    label: "{{slurm_prefix}}-{{ item.machine }}"
  tags: [ slurm-config ]
+
+- name: export facts about slurm cluster to be used by other modules
+  set_fact:
+    slurm:
+      user: "{{slurm_user}}"
+      domain: "{{slurm_domain}}"
+      base_image: "slurm:base"
+      mounts: "{{slurm_default_mounts}}"
+      network: "{{docker_network}}"
+  tags: always
--- a/roles/slurm/templates/slurm.conf.j2
+++ b/roles/slurm/templates/slurm.conf.j2
@@ -164,5 +164,7 @@ SlurmSchedLogFile={{slurm_log_path_sched}}
 #
 # COMPUTE NODES
 NodeName=slurm-exec[1-{{num_nodes}}] CPUs=2 CoresPerSocket=2 State=UNKNOWN
-NodeName=slurm-submit1 CPUs=1 State=UNKNOWN
-PartitionName=debug Nodes=slurm-exec[1-{{num_nodes}}] AllocNodes=slurm-submit1 Default=YES MaxTime=INFINITE State=UP
+{% for i in alloc_nodes -%}
+NodeName={{i}} State=UNKNOWN
+{% endfor %}
+PartitionName=debug Nodes=slurm-exec[1-{{num_nodes}}] AllocNodes={{alloc_nodes | join(',')}} Default=YES MaxTime=INFINITE State=UP
--- a/roles/slurm/vars/main.yml
+++ b/roles/slurm/vars/main.yml
@@ -0,0 +1,20 @@
+slurm_nodes_std: # default nodes: controller and submit machine
+- machine: ctl
+  image: slurm:slurmctld
+  exposed_ports: [ "6817:6817/tcp" ]
+- machine: submit1
+  image: slurm:slurmd
+  extra_mounts:
+  - "/home/{{unpriv_user}}/job3/:/mnt/:rw"
+slurm_nodes_exec: | # extend range to execute nodes list
+  {% set slurm_nodes_exec = [] %}
+  {% for i in range(1, num_nodes+1) -%}
+    {% set _ = slurm_nodes_exec.extend([
+      {'machine':'exec%s'|format(i), 'image': 'slurm:slurmd'}]) -%}
+  {%- endfor %}
+  {{ slurm_nodes_exec }}
+slurm_default_mounts:
+- /container/volumes/slurm/:/etc/slurm/:rw
+- /container/volumes/munge/munge.key:/etc/munge/munge.key:rw
+- slurm-shared:/shared/:rw
+
Author	SHA1	Message	Date
Thorsten Müller	c7e931f29e	fix: building base image -> update child images	2021-06-23 14:29:32 +02:00
Thorsten Müller	a73f9ad6ad	additional user in slurm base docker image	2021-06-23 14:28:50 +02:00
Thorsten Müller	c35dc25c39	labels, some cleanup	2021-06-22 19:09:52 +02:00
Thorsten Müller	1f4dfe1821	build cobald image from slurm role, separated tags	2021-06-22 16:48:56 +02:00
Thorsten Müller	78850d4636	merged slurm_dockerimage back into slurm role	2021-06-22 00:26:00 +02:00
Thorsten Müller	f83801cb62	removed cobald_facts module	2021-06-21 21:34:24 +02:00
Thorsten Müller	e78e184375	WIP: cobald container containing and using slurm	2021-06-21 19:19:19 +02:00
Thorsten Müller	02e87d7c40	cleanup, requisite files instead of startupscripts	2021-06-18 12:03:14 +02:00
Thorsten Müller	4450c9bb65	WIP: separate slurm base and docker images	2021-06-17 22:50:40 +02:00
Thorsten Müller	6eb6984d6a	new startup for cobald containers	2021-06-17 14:55:34 +02:00
Thorsten Müller	cc43a39ea3	dashboard revision	2021-06-14 10:43:47 +02:00
Thorsten Müller	962d9b5ac9	grafana dashboard updated, wait_for	2021-06-10 10:51:01 +02:00
Thorsten Müller	e81fb5d445	cobald container termination signal	2021-06-09 16:26:14 +02:00
Thorsten Müller	73945b6cb9	shorter hostname for cobald container	2021-06-08 16:07:54 +02:00
Thorsten Müller	089ea914b6	updated dashboards	2021-06-08 12:31:46 +02:00
Thorsten Müller	dd1baa4aef	grafana	2021-06-08 12:31:13 +02:00
Thorsten Müller	ea3195a93c	minor fixes (entrypoint) and restructuring	2021-06-08 12:28:09 +02:00
Thorsten Müller	aef1499e65	fixed influxdb when container absent and wait_for	2021-06-02 17:04:32 +02:00
Thorsten Müller	c7203f58ff	fix: influxdb connection issue	2021-06-01 19:17:04 +02:00
Thorsten Müller	2e0d83cca1	host ed-c7-2, fixed htop install	2021-06-01 18:30:18 +02:00
Thorsten Müller	35882ca1a9	TODO for influx modules	2021-05-25 23:56:19 +02:00
Thorsten Müller	4e7f33338e	telegraf + influxdb	2021-05-25 23:47:03 +02:00
Thorsten Müller	ddc6c2bb4d	influx modules: fixes, permission match, py2, args	2021-05-25 19:13:57 +02:00
Thorsten Müller	f9e29a4e30	influx bucket module	2021-05-25 15:00:47 +02:00
Thorsten Müller	c26e962898	token module improved	2021-05-25 11:55:30 +02:00
Thorsten Müller	38c117d6fa	influxdb2 plugins	2021-05-21 20:28:50 +02:00
Thorsten Müller	ecb9724ee3	generic unpriv_user	2021-05-11 23:56:29 +02:00
Thorsten Müller	4373e0a4a2	cobald development environment	2021-05-11 23:48:49 +02:00
Thorsten Müller	19b71c9933	first cobald tardis	2021-05-10 12:20:27 +02:00