project-config/grafana/nodepool.template
Ian Wienand 99c8bd953b grafana: don't use bool for refresh
grafyaml actually knows this is deprecated, but it's actually more
than that and doesn't work to refresh the variable at all now.  "1"
means "on load" which is what we want.

Change-Id: I34ecdd30c2188cb7e6ec32e33c6a6e99b6240934
2020-06-24 10:54:04 +10:00

198 lines
6.1 KiB
Plaintext

#
# NOTE: EDIT THE TEMPLATE FILE AND RUN create-nodepool.sh
#
dashboard:
title: 'Nodepool: %PROVIDER%'
templating:
- name: region
includeAll: true
multi: true
query: stats.gauges.nodepool.provider.%STAT_LIST%
refresh: 1
type: query
rows:
- title: Description
height: 150px
panels:
- title: Description
content: |
%PROVIDER% Nodepool Status
==========================
This dashboard monitors the status of the nodepool environment for %PROVIDER%.
**This dashboard is managed by [Grafyaml](https://docs.openstack.org/infra/system-config/grafyaml.html).**
If you would like to make changes to this dashboard, please see the template in the `grafana` directory in
[project-config](https://opendev.org/openstack/project-config/src/branch/master/grafana/nodepool.template).
type: text
- title: Nodes
showTitle: true
height: 150px
panels:
- title: Building
span: 3
sparkline:
full: true
show: true
targets:
- target: sumSeries(stats.gauges.nodepool.provider.$region.nodes.building)
type: singlestat
valueName: current
- title: Ready
span: 3
sparkline:
full: true
show: true
targets:
- target: sumSeries(stats.gauges.nodepool.provider.$region.nodes.ready)
type: singlestat
valueName: current
- title: In Use
span: 3
sparkline:
full: true
show: true
targets:
- target: sumSeries(stats.gauges.nodepool.provider.$region.nodes.in-use)
type: singlestat
valueName: current
- title: Deleting
span: 3
sparkline:
full: true
show: true
targets:
- target: sumSeries(stats.gauges.nodepool.provider.$region.nodes.deleting)
type: singlestat
valueName: current
- title: Test Nodes
height: 400px
panels:
- title: Test Node History - $region
type: graph
span: 12
stack: true
repeat: region
minSpan: 4
tooltip:
value_type: individual
yaxes:
- label: "nodes"
- show: false
targets:
- target: alias(sumSeries(stats.gauges.nodepool.provider.$region.nodes.building), 'Building')
- target: alias(sumSeries(stats.gauges.nodepool.provider.$region.nodes.ready), 'Available')
- target: alias(sumSeries(stats.gauges.nodepool.provider.$region.nodes.in-use), 'In Use')
- target: alias(sumSeries(stats.gauges.nodepool.provider.$region.nodes.used), 'Used')
- target: alias(sumSeries(stats.gauges.nodepool.provider.$region.nodes.deleting), 'Deleting')
- target: alias(sumSeries(stats.gauges.nodepool.provider.$region.max_servers), 'Max')
seriesOverrides:
- alias: Max
stack: False
- title: Node Launches
showTitle: true
height: 250px
panels:
- title: Ready Node Launch Attempts
type: graph
span: 4
lines: false
bars: true
nullPointMode: null as zero
yaxes:
- label: "events / min"
- show: false
targets:
- target: aliasSub(summarize(stats_counts.nodepool.launch.provider.$region.ready, '1m'), '.*stats_counts.nodepool.launch.provider.(.*).ready.*', '\1')
- title: Time to Ready
type: graph
span: 4
lines: false
bars: true
nullPointMode: null as zero
yaxes:
- label: "time"
format: ms
- show: false
targets:
- target: aliasByNode(stats.timers.nodepool.launch.provider.$region.ready.mean, 5)
- title: Error Node Launch Attempts
type: graph
span: 4
lines: false
bars: true
nullPointMode: null as zero
yaxes:
- label: "events / min"
- show: false
targets:
- target: alias(smartSummarize(sumSeries(stats_counts.nodepool.launch.provider.$region.error.*), '1m'), "All Errors")
- title: API Operations
showTitle: true
height: 250px
panels:
- title: Create Server
type: graph
lines: true
nullPointMode: connected
span: 4
yaxes:
- format: ms
label: Time
- show: false
targets:
- target: aliasByNode(stats.timers.nodepool.task.$region.ComputePostServers.mean, 4)
- title: Get Server
type: graph
span: 4
lines: true
nullPointMode: connected
span: 4
yaxes:
- format: ms
label: Time
- show: false
targets:
- target: aliasByNode(stats.timers.nodepool.task.$region.ComputePostServers.mean, 4)
- title: Delete Server
type: graph
span: 4
lines: true
nullPointMode: connected
span: 4
yaxes:
- format: ms
label: Time
- show: false
targets:
- target: aliasByNode(stats.timers.nodepool.task.$region.ComputePostServers.mean, 4)
- title: List Servers
type: graph
lines: true
nullPointMode: connected
span: 4
yaxes:
- format: ms
label: Time
- show: false
targets:
- target: aliasByNode(stats.timers.nodepool.task.$region.ComputePostServers.mean, 4)
- title: Get Limits
type: graph
lines: true
nullPointMode: connected
span: 4
yaxes:
- format: ms
label: Time
- show: false
targets:
- target: aliasByNode(stats.timers.nodepool.task.$region.ComputePostServers.mean, 4)