Monitoring Helm values reference

Helm values reference table for the monitoring application.

Key

Type

Default

Description

chronograf.env

object

stock settings for SQuaRE Phalanx deployment

Environment for chronograf

chronograf.envFromSecret

string

"monitoring"

Chronograf will read environment variables from this secret

chronograf.image

object

quay.io/influxdb/chronograf 1.10.3

chronograf image settings

chronograf.ingress

object

disabled; must be enabled and configured for each site

Chronograf ingress

chronograf.oauth.enabled

bool

false

Enable Chronograf oauth Never enable it: it breaks the deployment because it’s expecting a static TOKEN_SECRET. Instead, to get oauth working, leave this setting as-is and just configure all the correct environment variables (see below).

chronograf.resources

object

1Gi/1CPU request, 30Gi/4CPU limit

Chronograf resource requests/limits

chronograf.service

object

1 replica, ClusterIP

Chronograf service

chronograf.updateStrategy

object

Recreate

Chronograf update strategy

config.influxdbHostname

string

"monitoring.lsst.cloud"

Hostname for the singleton InfluxDBv2 collection point

config.influxdbOrg

string

"square"

InfluxDBv2 organization

cronjob.debug

bool

false

set to true to enable debug logging

cronjob.image.pullPolicy

string

"IfNotPresent"

imagePullPolicy for cronjobs

cronjob.image.repository

string

"ghcr.io/lsst-sqre/rubin-influx-tools"

repository for rubin-influx-tools, which supplies tools and dashboards

cronjob.image.tag

string

the appVersion of the chart

tag for rubin-influx-tools

cronjob.schedule.bucketmaker

string

"*/15 * * * *"

bucketmaker schedule

cronjob.schedule.bucketmapper

string

"3-59/15 * * * *"

bucketmapper schedule

cronjob.schedule.taskmaker

string

"6-59/15 * * * *"

taskmaker schedule

global.enabledServices

string

Set by Argo CD

services enabled in this RSP instance

global.host

string

Set by Argo CD

Host name for instance identification

global.vaultSecretsPath

string

Set by Argo CD

Base path for Vault secrets

influxdb2.adminUser.bucket

string

"monitoring_raw_"

Bucket to dump raw monitoring data into

influxdb2.adminUser.existingSecret

string

"monitoring"

Where we store secrets to run the server

influxdb2.adminUser.organization

string

"square"

InfluxDB internal organization

influxdb2.adminUser.retention_policy

string

"30d"

How long to keep data

influxdb2.adminUser.user

string

"admin"

User name

influxdb2.ingress

object

disabled, must be enabled and configured at each site

InfluxDB2 ingress configuration.

influxdb2.livenessProbe.failureThreshold

int

10

Number of checks to conclude whether InfluxDB has died

influxdb2.livenessProbe.periodSeconds

int

10

Period between checks for whether InfluxDB is still alive

influxdb2.resources

object

See values.yaml

Resource limits and requests for the InfluxDB server instance

influxdb2.startupProbe.enabled

bool

true

Whether to enable a startup probe

influxdb2.startupProbe.failureThreshold

int

60

Number of checks to conclude whether InfluxDB won’t start. High to allow up to 10 minutes for startup, because checking many shards can be slow.

influxdb2.startupProbe.initialDelaySeconds

int

30

How long to wait before checking the first time

influxdb2.startupProbe.periodSeconds

int

10

Period between checking whether InfluxDB has started