mudhorn/node_exporter
1
0
Fork 0
mirror of https://github.com/prometheus/node_exporter.git synced 2024-12-24 21:24:30 -08:00
Code Issues Actions Packages Projects Releases Wiki Activity

Migrate dashboards to new grafonnet library (#3147)
Some checks failed
golangci-lint / lint (push) Has been cancelled
Details

Browse source 
Migrated away from deprecated Grafonnet library. This replaces panels using Angular JS which are disabled by default in Grafana 11 and will be unsupported in Grafana 12.

Fixes #3046

---------

Signed-off-by: Tom <12222103+critchtionary@users.noreply.github.com>
This commit is contained in:
Tom 2024-12-19 15:49:22 +00:00 committed by GitHub
parent ff97e35a71
commit d0c1d00d18
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
3 changed files with 871 additions and 894 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

825
docs/node-mixin/dashboards/use.libsonnet
View file

@ -1,469 +1,480 @@
local grafana = import 'github.com/grafana/grafonnet-lib/grafonnet/grafana.libsonnet';
local grafana = import 'github.com/grafana/grafonnet/gen/grafonnet-latest/main.libsonnet';
local dashboard = grafana.dashboard;
local row = grafana.row;
local prometheus = grafana.prometheus;
local template = grafana.template;
local graphPanel = grafana.graphPanel;
local variable = dashboard.variable;
local row = grafana.panel.row;
local prometheus = grafana.query.prometheus;
local timeSeriesPanel = grafana.panel.timeSeries;
local tsOptions = timeSeriesPanel.options;
local tsStandardOptions = timeSeriesPanel.standardOptions;
local tsQueryOptions = timeSeriesPanel.queryOptions;
local tsCustom = timeSeriesPanel.fieldConfig.defaults.custom;
local tsLegend = tsOptions.legend;
local c = import '../config.libsonnet';
local datasourceTemplate = {
current: {
text: 'default',
value: 'default',
},
hide: 0,
label: 'Data Source',
name: 'datasource',
options: [],
query: 'prometheus',
refresh: 1,
regex: '',
type: 'datasource',
};
local datasource = variable.datasource.new(
'datasource', 'prometheus'
);
local tsCommonPanelOptions =
variable.query.withDatasourceFromVariable(datasource)
+ tsCustom.stacking.withMode('normal')
+ tsCustom.withFillOpacity(100)
+ tsCustom.withShowPoints('never')
+ tsLegend.withShowLegend(false)
+ tsOptions.tooltip.withMode('multi')
+ tsOptions.tooltip.withSort('desc');
local CPUUtilisation =
graphPanel.new(
timeSeriesPanel.new(
'CPU Utilisation',
datasource='$datasource',
span=6,
format='percentunit',
stack=true,
fill=10,
legend_show=false,
) { tooltip+: { sort: 2 } };
)
+ tsCommonPanelOptions
+ tsStandardOptions.withUnit('percentunit');
local CPUSaturation =
// TODO: Is this a useful panel? At least there should be some explanation how load
// average relates to the "CPU saturation" in the title.
graphPanel.new(
timeSeriesPanel.new(
'CPU Saturation (Load1 per CPU)',
datasource='$datasource',
span=6,
format='percentunit',
stack=true,
fill=10,
legend_show=false,
) { tooltip+: { sort: 2 } };
)
+ tsCommonPanelOptions
+ tsStandardOptions.withUnit('percentunit');
local memoryUtilisation =
graphPanel.new(
timeSeriesPanel.new(
'Memory Utilisation',
datasource='$datasource',
span=6,
format='percentunit',
stack=true,
fill=10,
legend_show=false,
) { tooltip+: { sort: 2 } };
)
+ tsCommonPanelOptions
+ tsStandardOptions.withUnit('percentunit');
local memorySaturation =
graphPanel.new(
timeSeriesPanel.new(
'Memory Saturation (Major Page Faults)',
datasource='$datasource',
span=6,
format='rds',
stack=true,
fill=10,
legend_show=false,
) { tooltip+: { sort: 2 } };
)
+ tsCommonPanelOptions
+ tsStandardOptions.withUnit('rds');
local networkOverrides = tsStandardOptions.withOverrides(
[
tsStandardOptions.override.byRegexp.new('/Transmit/')
+ tsStandardOptions.override.byRegexp.withPropertiesFromOptions(
tsCustom.withTransform('negative-Y')
),
]
);
local networkUtilisation =
graphPanel.new(
timeSeriesPanel.new(
'Network Utilisation (Bytes Receive/Transmit)',
datasource='$datasource',
span=6,
format='Bps',
stack=true,
fill=10,
legend_show=false,
)
.addSeriesOverride({ alias: '/Receive/', stack: 'A' })
.addSeriesOverride({ alias: '/Transmit/', stack: 'B', transform: 'negative-Y' })
{ tooltip+: { sort: 2 } };
+ tsCommonPanelOptions
+ tsStandardOptions.withUnit('Bps')
+ networkOverrides;
local networkSaturation =
graphPanel.new(
timeSeriesPanel.new(
'Network Saturation (Drops Receive/Transmit)',
datasource='$datasource',
span=6,
format='Bps',
stack=true,
fill=10,
legend_show=false,
)
.addSeriesOverride({ alias: '/ Receive/', stack: 'A' })
.addSeriesOverride({ alias: '/ Transmit/', stack: 'B', transform: 'negative-Y' })
{ tooltip+: { sort: 2 } };
+ tsCommonPanelOptions
+ tsStandardOptions.withUnit('Bps')
+ networkOverrides;
local diskIOUtilisation =
graphPanel.new(
timeSeriesPanel.new(
'Disk IO Utilisation',
datasource='$datasource',
span=6,
format='percentunit',
stack=true,
fill=10,
legend_show=false,
) { tooltip+: { sort: 2 } };
)
+ tsCommonPanelOptions
+ tsStandardOptions.withUnit('percentunit');
local diskIOSaturation =
graphPanel.new(
timeSeriesPanel.new(
'Disk IO Saturation',
datasource='$datasource',
span=6,
format='percentunit',
stack=true,
fill=10,
legend_show=false,
) { tooltip+: { sort: 2 } };
)
+ tsCommonPanelOptions
+ tsStandardOptions.withUnit('percentunit');
local diskSpaceUtilisation =
graphPanel.new(
timeSeriesPanel.new(
'Disk Space Utilisation',
datasource='$datasource',
span=12,
format='percentunit',
stack=true,
fill=10,
legend_show=false,
) { tooltip+: { sort: 2 } };
)
+ tsCommonPanelOptions
+ tsStandardOptions.withUnit('percentunit');
{
_clusterTemplate:: template.new(
name='cluster',
datasource='$datasource',
query='label_values(node_time_seconds, %s)' % $._config.clusterLabel,
current='',
hide=if $._config.showMultiCluster then '' else '2',
refresh=2,
includeAll=false,
sort=1
),
_clusterVariable::
variable.query.new('cluster')
+ variable.query.withDatasourceFromVariable(datasource)
+ variable.query.queryTypes.withLabelValues(
$._config.clusterLabel,
'node_time_seconds',
)
+ (if $._config.showMultiCluster then variable.query.generalOptions.showOnDashboard.withLabelAndValue() else variable.query.generalOptions.showOnDashboard.withNothing())
+ variable.query.refresh.onTime()
+ variable.query.selectionOptions.withIncludeAll(false)
+ variable.query.withSort(asc=true),
grafanaDashboards+:: {
'node-rsrc-use.json':
dashboard.new(
'%sUSE Method / Node' % $._config.dashboardNamePrefix,
time_from='now-1h',
tags=($._config.dashboardTags),
timezone='utc',
refresh='30s',
graphTooltip='shared_crosshair',
uid=std.md5('node-rsrc-use.json')
)
.addTemplate(datasourceTemplate)
.addTemplate($._clusterTemplate)
.addTemplate(
template.new(
+ dashboard.time.withFrom('now-1h')
+ dashboard.withTags($._config.dashboardTags)
+ dashboard.withTimezone('utc')
+ dashboard.withRefresh('30s')
+ dashboard.graphTooltip.withSharedCrosshair()
+ dashboard.withUid(std.md5('node-rsrc-use.json'))
+ dashboard.withVariables([
datasource,
$._clusterVariable,
variable.query.new('instance')
+ variable.query.withDatasourceFromVariable(datasource)
+ variable.query.queryTypes.withLabelValues(
'instance',
'$datasource',
'label_values(node_exporter_build_info{%(nodeExporterSelector)s, %(clusterLabel)s="$cluster"}, instance)' % $._config,
refresh='time',
sort=1
)
)
.addRow(
row.new('CPU')
.addPanel(CPUUtilisation.addTarget(prometheus.target('instance:node_cpu_utilisation:rate%(rateInterval)s{%(nodeExporterSelector)s, instance="$instance", %(clusterLabel)s="$cluster"} != 0' % $._config, legendFormat='Utilisation')))
.addPanel(CPUSaturation.addTarget(prometheus.target('instance:node_load1_per_cpu:ratio{%(nodeExporterSelector)s, instance="$instance", %(clusterLabel)s="$cluster"} != 0' % $._config, legendFormat='Saturation')))
)
.addRow(
row.new('Memory')
.addPanel(memoryUtilisation.addTarget(prometheus.target('instance:node_memory_utilisation:ratio{%(nodeExporterSelector)s, instance="$instance", %(clusterLabel)s="$cluster"} != 0' % $._config, legendFormat='Utilisation')))
.addPanel(memorySaturation.addTarget(prometheus.target('instance:node_vmstat_pgmajfault:rate%(rateInterval)s{%(nodeExporterSelector)s, instance="$instance", %(clusterLabel)s="$cluster"} != 0' % $._config, legendFormat='Major page Faults')))
)
.addRow(
row.new('Network')
.addPanel(
networkUtilisation
.addTarget(prometheus.target('instance:node_network_receive_bytes_excluding_lo:rate%(rateInterval)s{%(nodeExporterSelector)s, instance="$instance", %(clusterLabel)s="$cluster"} != 0' % $._config, legendFormat='Receive'))
.addTarget(prometheus.target('instance:node_network_transmit_bytes_excluding_lo:rate%(rateInterval)s{%(nodeExporterSelector)s, instance="$instance", %(clusterLabel)s="$cluster"} != 0' % $._config, legendFormat='Transmit'))
)
.addPanel(
networkSaturation
.addTarget(prometheus.target('instance:node_network_receive_drop_excluding_lo:rate%(rateInterval)s{%(nodeExporterSelector)s, instance="$instance", %(clusterLabel)s="$cluster"} != 0' % $._config, legendFormat='Receive'))
.addTarget(prometheus.target('instance:node_network_transmit_drop_excluding_lo:rate%(rateInterval)s{%(nodeExporterSelector)s, instance="$instance", %(clusterLabel)s="$cluster"} != 0' % $._config, legendFormat='Transmit'))
)
)
.addRow(
row.new('Disk IO')
.addPanel(diskIOUtilisation.addTarget(prometheus.target('instance_device:node_disk_io_time_seconds:rate%(rateInterval)s{%(nodeExporterSelector)s, instance="$instance", %(clusterLabel)s="$cluster"} != 0' % $._config, legendFormat='{{device}}')))
.addPanel(diskIOSaturation.addTarget(prometheus.target('instance_device:node_disk_io_time_weighted_seconds:rate%(rateInterval)s{%(nodeExporterSelector)s, instance="$instance", %(clusterLabel)s="$cluster"} != 0' % $._config, legendFormat='{{device}}')))
)
.addRow(
row.new('Disk Space')
.addPanel(
diskSpaceUtilisation.addTarget(prometheus.target(
|||
sort_desc(1 -
(
max without (mountpoint, fstype) (node_filesystem_avail_bytes{%(nodeExporterSelector)s, fstype!="", instance="$instance", %(clusterLabel)s="$cluster"})
/
max without (mountpoint, fstype) (node_filesystem_size_bytes{%(nodeExporterSelector)s, fstype!="", instance="$instance", %(clusterLabel)s="$cluster"})
) != 0
)
||| % $._config, legendFormat='{{device}}'
))
'node_exporter_build_info{%(nodeExporterSelector)s, %(clusterLabel)s="$cluster"}' % $._config,
)
+ variable.query.refresh.onTime()
+ variable.query.withSort(asc=true),
])
+ dashboard.withPanels(
grafana.util.grid.makeGrid([
row.new('CPU')
+ row.withPanels([
CPUUtilisation + tsQueryOptions.withTargets([prometheus.new('$datasource', 'instance:node_cpu_utilisation:rate%(rateInterval)s{%(nodeExporterSelector)s, instance="$instance", %(clusterLabel)s="$cluster"} != 0' % $._config) + prometheus.withLegendFormat('Utilisation')]),
CPUSaturation + tsQueryOptions.withTargets([prometheus.new('$datasource', 'instance:node_load1_per_cpu:ratio{%(nodeExporterSelector)s, instance="$instance", %(clusterLabel)s="$cluster"} != 0' % $._config) + prometheus.withLegendFormat('Saturation')]),
]),
row.new('Memory')
+ row.withPanels([
memoryUtilisation + tsQueryOptions.withTargets([prometheus.new('$datasource', 'instance:node_memory_utilisation:ratio{%(nodeExporterSelector)s, instance="$instance", %(clusterLabel)s="$cluster"} != 0' % $._config) + prometheus.withLegendFormat('Utilisation')]),
memorySaturation + tsQueryOptions.withTargets([prometheus.new('$datasource', 'instance:node_vmstat_pgmajfault:rate%(rateInterval)s{%(nodeExporterSelector)s, instance="$instance", %(clusterLabel)s="$cluster"} != 0' % $._config) + prometheus.withLegendFormat('Major page Faults')]),
]),
row.new('Network')
+ row.withPanels([
networkUtilisation + tsQueryOptions.withTargets([
prometheus.new('$datasource', 'instance:node_network_receive_bytes_excluding_lo:rate%(rateInterval)s{%(nodeExporterSelector)s, instance="$instance", %(clusterLabel)s="$cluster"} != 0' % $._config) + prometheus.withLegendFormat('Receive'),
prometheus.new('$datasource', 'instance:node_network_transmit_bytes_excluding_lo:rate%(rateInterval)s{%(nodeExporterSelector)s, instance="$instance", %(clusterLabel)s="$cluster"} != 0' % $._config) + prometheus.withLegendFormat('Transmit'),
]),
networkSaturation + tsQueryOptions.withTargets([
prometheus.new('$datasource', 'instance:node_network_receive_drop_excluding_lo:rate%(rateInterval)s{%(nodeExporterSelector)s, instance="$instance", %(clusterLabel)s="$cluster"} != 0' % $._config) + prometheus.withLegendFormat('Receive'),
prometheus.new('$datasource', 'instance:node_network_transmit_drop_excluding_lo:rate%(rateInterval)s{%(nodeExporterSelector)s, instance="$instance", %(clusterLabel)s="$cluster"} != 0' % $._config) + prometheus.withLegendFormat('Transmit'),
]),
]),
row.new('Disk IO')
+ row.withPanels([
diskIOUtilisation + tsQueryOptions.withTargets([prometheus.new('$datasource', 'instance_device:node_disk_io_time_seconds:rate%(rateInterval)s{%(nodeExporterSelector)s, instance="$instance", %(clusterLabel)s="$cluster"} != 0' % $._config) + prometheus.withLegendFormat('{{device}}')]),
diskIOSaturation + tsQueryOptions.withTargets([prometheus.new('$datasource', 'instance_device:node_disk_io_time_weighted_seconds:rate%(rateInterval)s{%(nodeExporterSelector)s, instance="$instance", %(clusterLabel)s="$cluster"} != 0' % $._config) + prometheus.withLegendFormat('{{device}}')]),
]),
], panelWidth=12, panelHeight=7)
+ grafana.util.grid.makeGrid([
row.new('Disk Space')
+ row.withPanels([
diskSpaceUtilisation + tsQueryOptions.withTargets([
prometheus.new(
'$datasource',
|||
sort_desc(1 -
(
max without (mountpoint, fstype) (node_filesystem_avail_bytes{%(nodeExporterSelector)s, fstype!="", instance="$instance", %(clusterLabel)s="$cluster"})
/
max without (mountpoint, fstype) (node_filesystem_size_bytes{%(nodeExporterSelector)s, fstype!="", instance="$instance", %(clusterLabel)s="$cluster"})
) != 0
)
||| % $._config
) + prometheus.withLegendFormat('{{device}}'),
]),
]),
], panelWidth=24, panelHeight=7, startY=34),
),
'node-cluster-rsrc-use.json':
dashboard.new(
'%sUSE Method / Cluster' % $._config.dashboardNamePrefix,
time_from='now-1h',
tags=($._config.dashboardTags),
timezone='utc',
refresh='30s',
graphTooltip='shared_crosshair',
uid=std.md5('node-cluster-rsrc-use.json')
)
.addTemplate(datasourceTemplate)
.addTemplate($._clusterTemplate)
.addRow(
row.new('CPU')
.addPanel(
CPUUtilisation
.addTarget(prometheus.target(
|||
((
instance:node_cpu_utilisation:rate%(rateInterval)s{%(nodeExporterSelector)s, %(clusterLabel)s="$cluster"}
*
instance:node_num_cpu:sum{%(nodeExporterSelector)s, %(clusterLabel)s="$cluster"}
) != 0 )
/ scalar(sum(instance:node_num_cpu:sum{%(nodeExporterSelector)s, %(clusterLabel)s="$cluster"}))
||| % $._config, legendFormat='{{ instance }}'
))
)
.addPanel(
CPUSaturation
.addTarget(prometheus.target(
|||
(
instance:node_load1_per_cpu:ratio{%(nodeExporterSelector)s, %(clusterLabel)s="$cluster"}
/ scalar(count(instance:node_load1_per_cpu:ratio{%(nodeExporterSelector)s, %(clusterLabel)s="$cluster"}))
) != 0
||| % $._config, legendFormat='{{instance}}'
))
)
)
.addRow(
row.new('Memory')
.addPanel(
memoryUtilisation
.addTarget(prometheus.target(
|||
(
instance:node_memory_utilisation:ratio{%(nodeExporterSelector)s, %(clusterLabel)s="$cluster"}
/ scalar(count(instance:node_memory_utilisation:ratio{%(nodeExporterSelector)s, %(clusterLabel)s="$cluster"}))
) != 0
||| % $._config, legendFormat='{{instance}}',
))
)
.addPanel(memorySaturation.addTarget(prometheus.target('instance:node_vmstat_pgmajfault:rate%(rateInterval)s{%(nodeExporterSelector)s, %(clusterLabel)s="$cluster"}' % $._config, legendFormat='{{instance}}')))
)
.addRow(
row.new('Network')
.addPanel(
networkUtilisation
.addTarget(prometheus.target('instance:node_network_receive_bytes_excluding_lo:rate%(rateInterval)s{%(nodeExporterSelector)s, %(clusterLabel)s="$cluster"} != 0' % $._config, legendFormat='{{instance}} Receive'))
.addTarget(prometheus.target('instance:node_network_transmit_bytes_excluding_lo:rate%(rateInterval)s{%(nodeExporterSelector)s, %(clusterLabel)s="$cluster"} != 0' % $._config, legendFormat='{{instance}} Transmit'))
)
.addPanel(
networkSaturation
.addTarget(prometheus.target('instance:node_network_receive_drop_excluding_lo:rate%(rateInterval)s{%(nodeExporterSelector)s, %(clusterLabel)s="$cluster"} != 0' % $._config, legendFormat='{{instance}} Receive'))
.addTarget(prometheus.target('instance:node_network_transmit_drop_excluding_lo:rate%(rateInterval)s{%(nodeExporterSelector)s, %(clusterLabel)s="$cluster"} != 0' % $._config, legendFormat='{{instance}} Transmit'))
)
)
.addRow(
row.new('Disk IO')
.addPanel(
diskIOUtilisation
.addTarget(prometheus.target(
|||
(
instance_device:node_disk_io_time_seconds:rate%(rateInterval)s{%(nodeExporterSelector)s, %(clusterLabel)s="$cluster"}
/ scalar(count(instance_device:node_disk_io_time_seconds:rate%(rateInterval)s{%(nodeExporterSelector)s, %(clusterLabel)s="$cluster"}))
) != 0
||| % $._config, legendFormat='{{instance}} {{device}}'
))
)
.addPanel(
diskIOSaturation
.addTarget(prometheus.target(
|||
(
+ dashboard.time.withFrom('now-1h')
+ dashboard.withTags($._config.dashboardTags)
+ dashboard.withTimezone('utc')
+ dashboard.withRefresh('30s')
+ dashboard.graphTooltip.withSharedCrosshair()
+ dashboard.withUid(std.md5('node-cluster-rsrc-use.json'))
+ dashboard.withVariables([
datasource,
$._clusterVariable,
variable.query.withDatasourceFromVariable(datasource)
+ variable.query.refresh.onTime()
+ variable.query.withSort(asc=true),
])
+ dashboard.withPanels(
grafana.util.grid.makeGrid([
row.new('CPU')
+ row.withPanels([
CPUUtilisation + tsQueryOptions.withTargets([
prometheus.new(
'$datasource',
|||
((
instance:node_cpu_utilisation:rate%(rateInterval)s{%(nodeExporterSelector)s, %(clusterLabel)s="$cluster"}
*
instance:node_num_cpu:sum{%(nodeExporterSelector)s, %(clusterLabel)s="$cluster"}
) != 0 )
/ scalar(sum(instance:node_num_cpu:sum{%(nodeExporterSelector)s, %(clusterLabel)s="$cluster"}))
||| % $._config
) + prometheus.withLegendFormat('{{ instance }}'),
]),
CPUSaturation + tsQueryOptions.withTargets([
prometheus.new(
'$datasource',
|||
(
instance:node_load1_per_cpu:ratio{%(nodeExporterSelector)s, %(clusterLabel)s="$cluster"}
/ scalar(count(instance:node_load1_per_cpu:ratio{%(nodeExporterSelector)s, %(clusterLabel)s="$cluster"}))
) != 0
||| % $._config
) + prometheus.withLegendFormat('{{ instance }}'),
]),
]),
row.new('Memory')
+ row.withPanels([
memoryUtilisation + tsQueryOptions.withTargets([
prometheus.new(
'$datasource',
|||
(
instance:node_memory_utilisation:ratio{%(nodeExporterSelector)s, %(clusterLabel)s="$cluster"}
/ scalar(count(instance:node_memory_utilisation:ratio{%(nodeExporterSelector)s, %(clusterLabel)s="$cluster"}))
) != 0
||| % $._config
) + prometheus.withLegendFormat('{{ instance }}'),
]),
memorySaturation + tsQueryOptions.withTargets([
prometheus.new(
'$datasource',
'instance:node_vmstat_pgmajfault:rate%(rateInterval)s{%(nodeExporterSelector)s, %(clusterLabel)s="$cluster"}' % $._config
) + prometheus.withLegendFormat('{{ instance }}'),
]),
]),
row.new('Network')
+ row.withPanels([
networkUtilisation + tsQueryOptions.withTargets([
prometheus.new(
'$datasource',
'instance:node_network_receive_bytes_excluding_lo:rate%(rateInterval)s{%(nodeExporterSelector)s, %(clusterLabel)s="$cluster"} != 0' % $._config
) + prometheus.withLegendFormat('{{ instance }} Receive'),
prometheus.new(
'$datasource',
'instance:node_network_transmit_bytes_excluding_lo:rate%(rateInterval)s{%(nodeExporterSelector)s, %(clusterLabel)s="$cluster"} != 0' % $._config
) + prometheus.withLegendFormat('{{ instance }} Transmit'),
]),
networkSaturation + tsQueryOptions.withTargets([
prometheus.new(
'$datasource',
'instance:node_network_receive_drop_excluding_lo:rate%(rateInterval)s{%(nodeExporterSelector)s, %(clusterLabel)s="$cluster"} != 0' % $._config
) + prometheus.withLegendFormat('{{ instance }} Receive'),
prometheus.new(
'$datasource',
'instance:node_network_transmit_drop_excluding_lo:rate%(rateInterval)s{%(nodeExporterSelector)s, %(clusterLabel)s="$cluster"} != 0' % $._config
) + prometheus.withLegendFormat('{{ instance }} Transmit'),
]),
]),
row.new('Disk IO')
+ row.withPanels([
diskIOUtilisation + tsQueryOptions.withTargets([
prometheus.new(
'$datasource',
|||
instance_device:node_disk_io_time_seconds:rate%(rateInterval)s{%(nodeExporterSelector)s, %(clusterLabel)s="$cluster"}
/ scalar(count(instance_device:node_disk_io_time_seconds:rate%(rateInterval)s{%(nodeExporterSelector)s, %(clusterLabel)s="$cluster"}))
||| % $._config
) + prometheus.withLegendFormat('{{ instance }} {{device}}'),
]),
diskIOSaturation + tsQueryOptions.withTargets([prometheus.new(
'$datasource',
|||
instance_device:node_disk_io_time_weighted_seconds:rate%(rateInterval)s{%(nodeExporterSelector)s, %(clusterLabel)s="$cluster"}
/ scalar(count(instance_device:node_disk_io_time_weighted_seconds:rate%(rateInterval)s{%(nodeExporterSelector)s, %(clusterLabel)s="$cluster"}))
) != 0
||| % $._config, legendFormat='{{instance}} {{device}}'
))
)
)
.addRow(
row.new('Disk Space')
.addPanel(
diskSpaceUtilisation
.addTarget(prometheus.target(
|||
sum without (device) (
max without (fstype, mountpoint) ((
node_filesystem_size_bytes{%(nodeExporterSelector)s, %(fsSelector)s, %(fsMountpointSelector)s, %(clusterLabel)s="$cluster"}
-
node_filesystem_avail_bytes{%(nodeExporterSelector)s, %(fsSelector)s, %(fsMountpointSelector)s, %(clusterLabel)s="$cluster"}
) != 0)
)
/ scalar(sum(max without (fstype, mountpoint) (node_filesystem_size_bytes{%(nodeExporterSelector)s, %(fsSelector)s, %(fsMountpointSelector)s, %(clusterLabel)s="$cluster"})))
||| % $._config, legendFormat='{{instance}}'
))
)
||| % $._config
) + prometheus.withLegendFormat('{{ instance }} {{device}}')]),
]),
], panelWidth=12, panelHeight=7)
+ grafana.util.grid.makeGrid([
row.new('Disk Space')
+ row.withPanels([
diskSpaceUtilisation + tsQueryOptions.withTargets([
prometheus.new(
'$datasource',
|||
sum without (device) (
max without (fstype, mountpoint) ((
node_filesystem_size_bytes{%(nodeExporterSelector)s, %(fsSelector)s, %(fsMountpointSelector)s, %(clusterLabel)s="$cluster"}
-
node_filesystem_avail_bytes{%(nodeExporterSelector)s, %(fsSelector)s, %(fsMountpointSelector)s, %(clusterLabel)s="$cluster"}
) != 0)
)
/ scalar(sum(max without (fstype, mountpoint) (node_filesystem_size_bytes{%(nodeExporterSelector)s, %(fsSelector)s, %(fsMountpointSelector)s, %(clusterLabel)s="$cluster"})))
||| % $._config
) + prometheus.withLegendFormat('{{ instance }}'),
]),
]),
], panelWidth=24, panelHeight=7, startY=34),
),
} +
if $._config.showMultiCluster then {
'node-multicluster-rsrc-use.json':
dashboard.new(
'%sUSE Method / Multi-cluster' % $._config.dashboardNamePrefix,
time_from='now-1h',
tags=($._config.dashboardTags),
timezone='utc',
refresh='30s',
graphTooltip='shared_crosshair',
uid=std.md5('node-multicluster-rsrc-use.json')
)
.addTemplate(datasourceTemplate)
.addRow(
row.new('CPU')
.addPanel(
CPUUtilisation
.addTarget(prometheus.target(
|||
sum(
((
instance:node_cpu_utilisation:rate%(rateInterval)s{%(nodeExporterSelector)s}
*
instance:node_num_cpu:sum{%(nodeExporterSelector)s}
) != 0)
/ scalar(sum(instance:node_num_cpu:sum{%(nodeExporterSelector)s}))
) by (%(clusterLabel)s)
||| % $._config, legendFormat='{{%(clusterLabel)s}}' % $._config
))
)
.addPanel(
CPUSaturation
.addTarget(prometheus.target(
|||
sum((
instance:node_load1_per_cpu:ratio{%(nodeExporterSelector)s}
/ scalar(count(instance:node_load1_per_cpu:ratio{%(nodeExporterSelector)s}))
) != 0) by (%(clusterLabel)s)
||| % $._config, legendFormat='{{%(clusterLabel)s}}' % $._config
))
)
)
.addRow(
row.new('Memory')
.addPanel(
memoryUtilisation
.addTarget(prometheus.target(
|||
sum((
instance:node_memory_utilisation:ratio{%(nodeExporterSelector)s}
/ scalar(count(instance:node_memory_utilisation:ratio{%(nodeExporterSelector)s}))
) != 0) by (%(clusterLabel)s)
||| % $._config, legendFormat='{{%(clusterLabel)s}}' % $._config
))
)
.addPanel(
memorySaturation
.addTarget(prometheus.target(
|||
sum((
instance:node_vmstat_pgmajfault:rate%(rateInterval)s{%(nodeExporterSelector)s}
) != 0) by (%(clusterLabel)s)
||| % $._config, legendFormat='{{%(clusterLabel)s}}' % $._config
))
)
)
.addRow(
row.new('Network')
.addPanel(
networkUtilisation
.addTarget(prometheus.target(
|||
sum((
instance:node_network_receive_bytes_excluding_lo:rate%(rateInterval)s{%(nodeExporterSelector)s}
) != 0) by (%(clusterLabel)s)
||| % $._config, legendFormat='{{%(clusterLabel)s}} Receive' % $._config
))
.addTarget(prometheus.target(
|||
sum((
instance:node_network_transmit_bytes_excluding_lo:rate%(rateInterval)s{%(nodeExporterSelector)s}
) != 0) by (%(clusterLabel)s)
||| % $._config, legendFormat='{{%(clusterLabel)s}} Transmit' % $._config
))
)
.addPanel(
networkSaturation
.addTarget(prometheus.target(
|||
sum((
instance:node_network_receive_drop_excluding_lo:rate%(rateInterval)s{%(nodeExporterSelector)s}
) != 0) by (%(clusterLabel)s)
||| % $._config, legendFormat='{{%(clusterLabel)s}} Receive' % $._config
))
.addTarget(prometheus.target(
|||
sum((
instance:node_network_transmit_drop_excluding_lo:rate%(rateInterval)s{%(nodeExporterSelector)s}
) != 0) by (%(clusterLabel)s)
||| % $._config, legendFormat='{{%(clusterLabel)s}} Transmit' % $._config
))
)
)
.addRow(
row.new('Disk IO')
.addPanel(
diskIOUtilisation
.addTarget(prometheus.target(
|||
sum((
instance_device:node_disk_io_time_seconds:rate%(rateInterval)s{%(nodeExporterSelector)s}
/ scalar(count(instance_device:node_disk_io_time_seconds:rate%(rateInterval)s{%(nodeExporterSelector)s}))
) != 0) by (%(clusterLabel)s, device)
||| % $._config, legendFormat='{{%(clusterLabel)s}} {{device}}' % $._config
))
)
.addPanel(
diskIOSaturation
.addTarget(prometheus.target(
|||
sum((
instance_device:node_disk_io_time_weighted_seconds:rate%(rateInterval)s{%(nodeExporterSelector)s}
/ scalar(count(instance_device:node_disk_io_time_weighted_seconds:rate%(rateInterval)s{%(nodeExporterSelector)s}))
) != 0) by (%(clusterLabel)s, device)
||| % $._config, legendFormat='{{%(clusterLabel)s}} {{device}}' % $._config
))
)
)
.addRow(
row.new('Disk Space')
.addPanel(
diskSpaceUtilisation
.addTarget(prometheus.target(
|||
sum (
sum without (device) (
max without (fstype, mountpoint, instance, pod) ((
node_filesystem_size_bytes{%(nodeExporterSelector)s, %(fsSelector)s, %(fsMountpointSelector)s} - node_filesystem_avail_bytes{%(nodeExporterSelector)s, %(fsSelector)s, %(fsMountpointSelector)s}
) != 0)
)
/ scalar(sum(max without (fstype, mountpoint) (node_filesystem_size_bytes{%(nodeExporterSelector)s, %(fsSelector)s, %(fsMountpointSelector)s})))
) by (%(clusterLabel)s)
||| % $._config, legendFormat='{{%(clusterLabel)s}}' % $._config
))
)
+ dashboard.time.withFrom('now-1h')
+ dashboard.withTags($._config.dashboardTags)
+ dashboard.withTimezone('utc')
+ dashboard.withRefresh('30s')
+ dashboard.graphTooltip.withSharedCrosshair()
+ dashboard.withUid(std.md5('node-multicluster-rsrc-use.json'))
+ dashboard.withVariables([
datasource,
variable.query.withDatasourceFromVariable(datasource)
+ variable.query.refresh.onTime()
+ variable.query.withSort(asc=true),
])
+ dashboard.withPanels(
grafana.util.grid.makeGrid([
row.new('CPU')
+ row.withPanels([
CPUUtilisation + tsQueryOptions.withTargets([
prometheus.new(
'$datasource',
|||
sum(
((
instance:node_cpu_utilisation:rate%(rateInterval)s{%(nodeExporterSelector)s}
*
instance:node_num_cpu:sum{%(nodeExporterSelector)s}
) != 0)
/ scalar(sum(instance:node_num_cpu:sum{%(nodeExporterSelector)s}))
) by (%(clusterLabel)s)
||| % $._config
) + prometheus.withLegendFormat('{{%(clusterLabel)s}}'),
]),
CPUSaturation + tsQueryOptions.withTargets([
prometheus.new(
'$datasource',
|||
sum((
instance:node_load1_per_cpu:ratio{%(nodeExporterSelector)s}
/ scalar(count(instance:node_load1_per_cpu:ratio{%(nodeExporterSelector)s}))
) != 0) by (%(clusterLabel)s)
||| % $._config
) + prometheus.withLegendFormat('{{%(clusterLabel)s}}'),
]),
]),
row.new('Memory')
+ row.withPanels([
memoryUtilisation + tsQueryOptions.withTargets([
prometheus.new(
'$datasource',
|||
sum((
instance:node_memory_utilisation:ratio{%(nodeExporterSelector)s}
/ scalar(count(instance:node_memory_utilisation:ratio{%(nodeExporterSelector)s}))
) != 0) by (%(clusterLabel)s)
||| % $._config
) + prometheus.withLegendFormat('{{%(clusterLabel)s}}'),
]),
memorySaturation + tsQueryOptions.withTargets([
prometheus.new(
'$datasource',
|||
sum((
instance:node_vmstat_pgmajfault:rate%(rateInterval)s{%(nodeExporterSelector)s}
) != 0) by (%(clusterLabel)s)
|||
% $._config
) + prometheus.withLegendFormat('{{%(clusterLabel)s}}'),
]),
]),
row.new('Network')
+ row.withPanels([
networkUtilisation + tsQueryOptions.withTargets([
prometheus.new(
'$datasource',
|||
sum((
instance:node_network_receive_bytes_excluding_lo:rate%(rateInterval)s{%(nodeExporterSelector)s}
) != 0) by (%(clusterLabel)s)
||| % $._config
) + prometheus.withLegendFormat('{{%(clusterLabel)s}} Receive'),
prometheus.new(
'$datasource',
|||
sum((
instance:node_network_transmit_bytes_excluding_lo:rate%(rateInterval)s{%(nodeExporterSelector)s}
) != 0) by (%(clusterLabel)s)
||| % $._config
) + prometheus.withLegendFormat('{{%(clusterLabel)s}} Transmit'),
]),
networkSaturation + tsQueryOptions.withTargets([
prometheus.new(
'$datasource',
|||
sum((
instance:node_network_receive_drop_excluding_lo:rate%(rateInterval)s{%(nodeExporterSelector)s}
) != 0) by (%(clusterLabel)s)
||| % $._config
) + prometheus.withLegendFormat('{{%(clusterLabel)s}} Receive'),
prometheus.new(
'$datasource',
|||
sum((
instance:node_network_transmit_drop_excluding_lo:rate%(rateInterval)s{%(nodeExporterSelector)s}
) != 0) by (%(clusterLabel)s)
||| % $._config
) + prometheus.withLegendFormat('{{%(clusterLabel)s}} Transmit'),
]),
]),
row.new('Disk IO')
+ row.withPanels([
diskIOUtilisation + tsQueryOptions.withTargets([
prometheus.new(
'$datasource',
|||
sum((
instance_device:node_disk_io_time_seconds:rate%(rateInterval)s{%(nodeExporterSelector)s}
/ scalar(count(instance_device:node_disk_io_time_seconds:rate%(rateInterval)s{%(nodeExporterSelector)s}))
) != 0) by (%(clusterLabel)s, device)
||| % $._config
) + prometheus.withLegendFormat('{{%(clusterLabel)s}} {{device}}'),
]),
diskIOSaturation + tsQueryOptions.withTargets([prometheus.new(
'$datasource',
|||
sum((
instance_device:node_disk_io_time_weighted_seconds:rate%(rateInterval)s{%(nodeExporterSelector)s}
/ scalar(count(instance_device:node_disk_io_time_weighted_seconds:rate%(rateInterval)s{%(nodeExporterSelector)s}))
) != 0) by (%(clusterLabel)s, device)
||| % $._config
) + prometheus.withLegendFormat('{{%(clusterLabel)s}} {{device}}')]),
]),
], panelWidth=12, panelHeight=7)
+ grafana.util.grid.makeGrid([
row.new('Disk Space')
+ row.withPanels([
diskSpaceUtilisation + tsQueryOptions.withTargets([
prometheus.new(
'$datasource',
|||
sum (
sum without (device) (
max without (fstype, mountpoint, instance, pod) ((
node_filesystem_size_bytes{%(nodeExporterSelector)s, %(fsSelector)s, %(fsMountpointSelector)s} - node_filesystem_avail_bytes{%(nodeExporterSelector)s, %(fsSelector)s, %(fsMountpointSelector)s}
) != 0)
)
/ scalar(sum(max without (fstype, mountpoint) (node_filesystem_size_bytes{%(nodeExporterSelector)s, %(fsSelector)s, %(fsMountpointSelector)s})))
) by (%(clusterLabel)s)
||| % $._config
) + prometheus.withLegendFormat('{{%(clusterLabel)s}}'),
]),
]),
], panelWidth=24, panelHeight=7, startY=34),
),
} else {},
}

15
docs/node-mixin/jsonnetfile.json
View file

@ -4,20 +4,11 @@
{
"source": {
"git": {
"remote": "https://github.com/grafana/grafonnet-lib.git",
"subdir": "grafonnet"
"remote": "https://github.com/grafana/grafonnet.git",
"subdir": "gen/grafonnet-latest"
}
},
"version": "master"
},
{
"source": {
"git": {
"remote": "https://github.com/grafana/grafonnet-lib.git",
"subdir": "grafonnet-7.0"
}
},
"version": "master"
"version": "main"
}
],
"legacyImports": false

925
docs/node-mixin/lib/prom-mixin.libsonnet
View file

File diff suppressed because it is too large Load diff