istio version 1.6.1 deployed with add-on kiali (1.18.2) enabled, grafana and prometheus disabled
configured Urls for External: grafana, prometheus and jaeger.
kiali works as expected.
But istio Component status masthead UI shows:
prometheus Not found
grafana Not found
istio-tracing Not found
is this related to https://github.com/kiali/kiali/pull/2598. ?
Hi @bethmage, could you share with us more details about your installation ?
It seems related to Istio check feature we introduced recently.
Is there an issue we can track ?
Thanks
addonComponents:
prometheus:
enabled: false
kiali:
enabled: true
grafana:
enabled: false
tracing:
enabled: false
Jaeger v1.16.0, Grafana v7.0.1 and Prometheus v2.18.1 are installed with stable helm Charts.
apiVersion: v1
data:
config.yaml: |-
istio_component_namespaces:
grafana: monitoring-dev
tracing: monitoring-dev
istiod: istio-system
prometheus: monitoring-dev
istio_namespace: istio-system
auth:
strategy: login
deployment:
accessible_namespaces: [’**’]
login_token:
signing_key: xxxxxxxxxxxxxxxxxx
server:
port: 20001
web_root: /kiali
external_services:
istio:
url_service_version: http://istiod.istio-system:15014/version
tracing:
enabled: true
auth:
type: none
use_kiali_token: false
in_cluster_url: http://jaeger-trace.monitoring-dev
url: https://jaeger.dev.test.com
whitelist_istio_system: [jaeger-query, istio-ingressgateway]
grafana:
enabled: true
auth:
type: basic
username: admin
password: xxxxxxxxxxxxxxxx
insecure_skip_verify: true
dashboards:
- name: ‘Istio Service Dashboard’
variables:
namespace: var-namespace
service: var-service
- name: ‘Istio Workload Dashboard’
variables:
namespace: var-namespace
workload: var-workload
in_cluster_url: http://grafana.monitoring-dev
url: https://grafana.dev.test.com
prometheus:
url: http://prometheus-server.monitoring-dev
custom_metrics_url: http://prometheus-server.monitoring-dev
…
apiVersion: v1
data:
mesh: |-
accessLogEncoding: TEXT
accessLogFile: /dev/stdout
accessLogFormat: “”
defaultConfig:
concurrency: 2
configPath: ./etc/istio/proxy
connectTimeout: 10s
controlPlaneAuthPolicy: NONE
discoveryAddress: istiod.istio-system.svc:15012
drainDuration: 45s
parentShutdownDuration: 1m0s
proxyAdminPort: 15000
proxyMetadata:
DNS_AGENT: “”
serviceCluster: istio-proxy
tracing:
zipkin:
address: jaeger-collector.monitoring-dev:9411
disableMixerHttpReports: true
disablePolicyChecks: true
enableAutoMtls: true
enablePrometheusMerge: false
enableTracing: true
ingressClass: istio
ingressControllerMode: STRICT
ingressService: istio-ingressgateway
protocolDetectionTimeout: 100ms
reportBatchMaxEntries: 100
reportBatchMaxTime: 1s
sdsUdsPath: unix:/etc/istio/proxy/SDS
trustDomain: cluster.local
trustDomainAliases: null
meshNetworks: ‘networks: {}’
values.yaml: |-
appNamespaces:
autoscaleEnabled: true
autoscaleMax: 5
autoscaleMin: 1
configMap: true
configNamespace: istio-config
configSource:
subscribedResources:
cpu:
targetAverageUtilization: 80
deploymentLabels: {}
enableProtocolSniffingForInbound: false
enableProtocolSniffingForOutbound: true
enabled: true
env: {}
hub: “”
image: pilot
ingress:
ingressClass: istio
ingressControllerMode: STRICT
ingressService: istio-ingressgateway
jwksResolverExtraRootCA: “”
keepaliveMaxServerConnectionAge: 30m
meshNetworks:
networks: {}
namespace: istio-system
nodeSelector: {}
plugins:
podAnnotations: {}
podAntiAffinityLabelSelector:
podAntiAffinityTermLabelSelector:
policy:
enabled: false
replicaCount: 1
resources:
requests:
cpu: 500m
memory: 2048Mi
rollingMaxSurge: 100%
rollingMaxUnavailable: 25%
tag: “”
tolerations:
traceSampling: 1
…
Hi @lucas.ponce, what else do you need about my installation?
also the kiali logs does not show any errors.
Thanks
Hi @bethmage,
Apoligizes, in the first previous response, I didn’t get the notification, in the last one as you put my nick I could see it.
I asked @xeviknal to take a looks as he is familiar with the feature.
HI @bethmage,
Can you confirm that you are running both prometheus, jaeger-tracing and grafana in another namespace than istio-system?
I guess that the deployments for tracing, prometheus and grafana are living in the monitornig-dev namespace, right?
I think that current implementation of that istio health icon in the masthead is only checking components living in the istio root namespace.
Thanks,
Hi @xeviknal, Yes. prometheus, jaeger-tracing and grafana are in the monitornig-dev namespace.
istio in the istio-system namespace.
Great! I’ve created a new issue for tracking this scenario. Thanks for reporting that out.
If you want, you can disable the istio health component in the meanwhile. You should change the CM/CR to disable the following flag:
external_services:
istio:
istio_status_enabled: false