Grafana: broken pipe

[ivanelgran23]

Hello! I installed the docker-composer hom7-prom-all image, but unfortunately grafana is constantly spamming with errors like these:
grafana | t=2021-12-15T18:27:24+0500 lvl=eror msg="Request error" logger=context userId=0 orgId=1 uname= error="Context.HTML:write tcp [::1]:3000->[::1]:43370: write: broken pipe"
grafana | t=2021-12-15T18:27:25+0500 lvl=eror msg="Request error" logger=context userId=0 orgId=1 uname= error="Context.HTML:write tcp [::1]:3000->[::1]:43376: write: broken pipe"
Please tell me how it can be fixed?

[talkyoo]

Same Error here - with hom7-influx-grafana.

[lukeescude]

I can confirm this is happening to me as well.

[haup]

Hi, I am using the "hom7-prom-all" and changed a the definition of grafana quite a bit, there are no errors anymore. It seems that the healthcheck triggered the request error

grafana:
#    image: grafana/grafana:master
   image: grafana/grafana:latest
   container_name: grafana
   volumes:
     - /data/grafana_data:/var/lib/grafana
     - ./grafana/provisioning/:/etc/grafana/provisioning/
   environment:
     - GF_SECURITY_ALLOW_EMBEDDING=true
     - GF_AUTH_ANONYMOUS_ENABLED=true
     - GF_AUTH_ANONYMOUS_ORG_ROLE=Viewer
#      - GF_AUTH_OAUTH_AUTO_LOGIN=true
     - GF_SECURITY_ADMIN_USER=${ADMIN_USER:-admin}
     - GF_SECURITY_ADMIN_PASSWORD=${ADMIN_PASSWORD:-admin}
     - GF_USERS_ALLOW_SIGN_UP=false
     - GF_EXPLORE_ENABLED=true
   restart: unless-stopped
   ports:
     - "9030:3000"
   healthcheck:
     test: ["CMD-SHELL", "wget --quiet --tries=1 --spider http://localhost:3000/api/health || exit 1"]
     interval: 1s
     timeout: 3s
     retries: 30
   labels:
     org.label-schema.group: "monitoring"
   logging:
     options:
       max-size: "50m"