Dashboard provisioning updateIntervalSeconds not working as expected

In a new Grafana setup I’m working with, we have a lot of organisations (500+) and I’d like to use dashboard provisioning.

I got everything to work as expected, the dashboard json files are being distributed to each organisation but the “updateIntervalSeconds” doesn’t seem to bite. I’d like to raise this from 3s to something way higher since the dashboards will not be changed that often and no need to hit the postgres that often with the query to the dashboard_provisioning table.

The setup I have is Grafana v5.1.3 (commit: 087143285) HA with PostgreSQL as backend.

Each organisation have it’s own yaml file in /etc/grafana/provisioning/dashboards with it’s name and specific orgId, looking something like this:

apiVersion: 1

- name: 'My_Orgs_Name'
  orgId: 2
  folder: 'Folder_Name'
  type: file
  disableDeletion: false
  updateIntervalSeconds: 120 #how often Grafana will scan for changed dashboards
    path: /var/lib/grafana/dashboards

Am I missing some configuration setting or have set it up wrong? Can I debug this somehow?

I just can see in the grafana log that it still looks to be queried every third second:

t=2018-06-13T09:41:59+0200 lvl=info msg="[SQL] SELECT \"id\", \"dashboard_id\", \"name\", \"external_id\", \"updated\" FROM \"dashboard_provisioning\" WHERE (name = $1) []interface {}{\"My_Orgs_Name\"} - took: 470.61µs" logger=sqlstore.xorm

t=2018-06-13T09:42:02+0200 lvl=info msg="[SQL] SELECT \"id\", \"dashboard_id\", \"name\", \"external_id\", \"updated\" FROM \"dashboard_provisioning\" WHERE (name = $1) []interface {}{\"My_Orgs_Name\"} - took: 605.883µs" logger=sqlstore.xorm


Please upgrade to at least Grafana v5.2.0.


Does updateIntervalSeconds having default value if not been set in yml file ? Will it look for change ?