def generate_sfn_execution_duration_graph(name: str,
cloudwatch_data_source: str, *args,
**kwargs):
"""
Generate step function graph
"""
targets = [
CloudwatchMetricsTarget(
alias=DURATION_MINIMUM_ALIAS,
namespace=NAMESPACE,
metricName="ExecutionTime",
statistics=["Minimum"],
dimensions={"StateMachineArn": name},
),
CloudwatchMetricsTarget(
alias=DURATION_AVERAGE_ALIAS,
namespace=NAMESPACE,
metricName="ExecutionTime",
statistics=["Average"],
dimensions={"StateMachineArn": name},
),
CloudwatchMetricsTarget(
alias=DURATION_MAXIMUM_ALIAS,
namespace=NAMESPACE,
metricName="ExecutionTime",
statistics=["Maximum"],
dimensions={"StateMachineArn": name},
),
]
yAxes = YAxes(
YAxis(format=MILLISECONDS_FORMAT, decimals=2),
YAxis(format=MILLISECONDS_FORMAT, decimals=2),
)
seriesOverrides = [
{
"alias": DURATION_MINIMUM_ALIAS,
"color": "#C8F2C2",
"lines": False
},
{
"alias": DURATION_AVERAGE_ALIAS,
"color": "#FADE2A",
"fill": 0
},
{
"alias": DURATION_MAXIMUM_ALIAS,
"color": "rgb(77, 159, 179)",
"fillBelowTo": DURATION_MINIMUM_ALIAS,
"lines": False,
},
]
alert = None
return Graph(
title="Step function execution duration",
dataSource=cloudwatch_data_source,
targets=targets,
seriesOverrides=seriesOverrides,
yAxes=yAxes,
transparent=TRANSPARENT,
editable=EDITABLE,
alert=alert,
alertThreshold=ALERT_THRESHOLD,
).auto_ref_ids()
dashboard = Dashboard(
title="{}s Dashboard".format(GRR_COMPONENT).title(),
rows=[
Row(panels=[panel(GRR_COMPONENT) for panel in row])
for row in GENERAL_PANELS
] + [
Row(panels=[
Graph(
title="QPS",
targets=[
Target(
expr='sum(rate(frontend_request_count_total[1m]))',
legendFormat="Requests",
),
],
yAxes=YAxes(left=YAxis(format="reqps")),
),
Graph(
title="Request Latency",
targets=[
Target(
expr=
'sum(rate(frontend_request_latency_sum[10m])) / sum(rate(frontend_request_latency_count[10m]))',
legendFormat="Latency",
),
],
yAxes=YAxes(left=YAxis(format=SECONDS_FORMAT)),
),
Graph(
title="Well Known Flows Requests Rate by Flow",
targets=[
),
Target(
expr=
'aws_ec_hash_based_cmds_average{environment="${environment}", dimension_cache_cluster_id=~"${cluster}.*"}',
legendFormat="hash-{{dimension_cache_cluster_id}}",
refId='G',
),
Target(
expr=
'aws_ec_list_based_cmds_average{environment="${environment}", dimension_cache_cluster_id=~"${cluster}.*"}',
legendFormat="list-{{dimension_cache_cluster_id}}",
refId='H',
),
],
yAxes=YAxes(
YAxis(format="µs"),
YAxis(format=SHORT_FORMAT),
),
),
]),
Row(panels=[
Graph(
title="Redis Commands Latency (avg)",
dataSource=DATASOURCE,
targets=[
Target(
expr=
'aws_ec_set_type_cmds_latency_average{environment="${environment}", dimension_cache_cluster_id=~"${cluster}.*"}',
legendFormat="set_type-{{dimension_cache_cluster_id}}",
refId='A',
),
Target(
),
Graph(
title="Threadpool Latency vs. Queuing Time",
targets=[
Target(
expr=
'sum(rate(threadpool_working_time_sum{job="grr_worker"}[10m])) / sum(rate(threadpool_working_time_count{job="grr_worker"}[10m]))',
legendFormat="Latency",
),
Target(
expr=
'sum(rate(threadpool_queueing_time_sum{job="grr_worker"}[10m])) / sum(rate(threadpool_queueing_time_count{job="grr_worker"}[10m]))',
legendFormat="Queueing Time",
),
],
yAxes=YAxes(left=YAxis(format=SECONDS_FORMAT)),
),
Graph(
title=
"Rate of Flow States a GRR Worker has moved through by GRR Worker Instance",
targets=[
Target(
expr=
'sum by (instance) (rate(grr_worker_states_run_total{job="grr_worker"}[10m]))',
legendFormat="{{instance}}",
),
],
),
]),
],
).auto_panel_ids()
from grr_grafanalib_dashboards.config import CLIENT_LOAD_STATS_DATA_SOURCE
client_id_variable = Template(name="ClientID", query="", type="textbox")
dashboard = Dashboard(
title="Client Load Stats Dashboard",
templating=Templating([client_id_variable]),
rows=[
Row(panels=[
Graph(
title="User CPU Usage",
targets=[
Target(target='Mean User CPU Rate', ),
Target(target='Max User CPU Rate', ),
],
yAxes=YAxes(left=YAxis(max=105, format="percent")),
),
Graph(
title="System CPU Usage",
targets=[
Target(target='Mean System CPU Rate', ),
Target(target='Max System CPU Rate', ),
],
yAxes=YAxes(left=YAxis(max=105, format="percent")),
),
]),
Row(panels=[
Graph(
title="Resident Memory",
targets=[
Target(target='Mean Resident Memory MB', ),
Target(
expr=
'sum(process_resident_memory_bytes{job="fleetspeak"})',
legendFormat="Resident Memory",
),
]),
Graph(
title="CPU Usage",
targets=[
Target(
expr=
'avg(rate(process_cpu_seconds_total{job="fleetspeak"}[30s])) * 100',
legendFormat="Average Process CPU Usage",
),
],
yAxes=YAxes(left=YAxis(max=105, format="percent")),
),
]),
Row(panels=[
Graph(
title="Datastore Latency per Operation",
targets=[
Target(
expr=
'sum by (operation) (rate(fleetspeak_server_datastore_operations_completed_latency_sum[10m]) / rate(fleetspeak_server_datastore_operations_completed_latency_count[10m]))',
legendFormat="{{operation}}",
),
],
yAxes=YAxes(left=YAxis(format=SECONDS_FORMAT)),
),
Heatmap(
OPS_FORMAT,
RowPanel,
SHORT_FORMAT,
SqlTarget,
YAxes,
YAxis,
)
dashboard = Dashboard(
title="Random stats from SQL DB",
panels=[
RowPanel(title="New row", gridPos=GridPos(h=1, w=24, x=0, y=8)),
Graph(
title="Some SQL Queries",
dataSource="Your SQL Source",
targets=[
SqlTarget(
rawSql=
'SELECT date as "time", metric FROM example WHERE $__timeFilter("time")',
refId="A",
),
],
yAxes=YAxes(
YAxis(format=OPS_FORMAT),
YAxis(format=SHORT_FORMAT),
),
gridPos=GridPos(h=8, w=24, x=0, y=9),
),
],
).auto_panel_ids()
def workflow_stats(collapse: bool) -> Row:
return Row(
title="Workflow Stats",
collapse=collapse,
panels=[
Graph(
title="Accepted Workflow",
dataSource=DATASOURCE,
targets=[
Target(
expr=
'sum(rate(flyte:propeller:all:workflow:accepted{project=~"$project", domain=~"$domain", wf=~"$workflow"}[5m]))',
refId='A',
),
],
yAxes=YAxes(
YAxis(format=OPS_FORMAT),
YAxis(format=SHORT_FORMAT),
),
),
Graph(
title="Successful Workflow",
dataSource=DATASOURCE,
targets=[
Target(
expr=
'sum(rate(flyte:propeller:all:workflow:success_duration_ms_count{project=~"$project", domain=~"$domain", wf=~"$workflow"}[5m]))',
refId='A',
),
],
yAxes=YAxes(
YAxis(format=OPS_FORMAT),
YAxis(format=SHORT_FORMAT),
),
),
Graph(
title="Failed Workflow",
dataSource=DATASOURCE,
targets=[
Target(
expr=
'sum(rate(flyte:propeller:all:workflow:failure_duration_ms_count{project=~"$project", domain=~"$domain", wf=~"$workflow"}[5m]))',
refId='A',
),
],
yAxes=YAxes(
YAxis(format=OPS_FORMAT),
YAxis(format=SHORT_FORMAT),
),
),
Graph(
title="Aborted Workflow",
dataSource=DATASOURCE,
targets=[
Target(
expr=
'sum(rate(flyte:propeller:all:workflow:workflow_aborted{project=~"$project", domain=~"$domain", wf=~"$workflow"}[5m]))',
refId='A',
),
],
yAxes=YAxes(
YAxis(format=OPS_FORMAT),
YAxis(format=SHORT_FORMAT),
),
),
Graph(
title="Successful workflow execution time by Quantile",
dataSource=DATASOURCE,
targets=[
Target(
expr=
'sum(flyte:propeller:all:workflow:success_duration_ms{project=~"$project", domain=~"$domain", wf=~"$workflow"}) by (quantile)',
refId='A',
),
],
yAxes=single_y_axis(format=MILLISECONDS_FORMAT),
),
Graph(
title="Failed workflow execution time by Quantile",
dataSource=DATASOURCE,
targets=[
Target(
expr=
'sum(flyte:propeller:all:workflow:failed_duration_ms{project=~"$project", domain=~"$domain", wf=~"$workflow"}) by (quantile)',
refId='A',
),
],
yAxes=single_y_axis(format=MILLISECONDS_FORMAT),
),
Graph(
title="Node queuing latency by Quantile",
dataSource=DATASOURCE,
targets=[
Target(
expr=
'sum(flyte:propeller:all:node:queueing_latency_ms{project=~"$project", domain=~"$domain", wf=~"$workflow"}) by (quantile)',
refId='A',
),
],
yAxes=single_y_axis(format=MILLISECONDS_FORMAT),
),
])