Prometheus dashboards are tied to "Local" datasource

[dillaman]

The ansible installer configures a "prometheus" datasource, so lots of dashboards fail to load due to the missing data. These should either just use the "default" datasource or be switched to "prometheus".

[mgr-prometheus]$ git grep Local
alert-status.json:      "name": "Local",
ceph-at-a-glance.json:      "name": "Local",
ceph-backend-storage.json:      "name": "Local",
ceph-cluster.json:      "name": "Local",
ceph-cluster.json:          "datasource": "Local",
ceph-health.json:      "name": "Local",
ceph-osd-information.json:          "datasource": "Local",
ceph-osd-information.json:          "datasource": "Local",
ceph-osd-information.json:          "datasource": "Local",
ceph-osd-information.json:          "datasource": "Local",
ceph-osd-information.json:          "datasource": "Local",
ceph-osd-information.json:          "datasource": "Local",
ceph-osd-information.json:          "datasource": "Local",
ceph-osd-information.json:          "datasource": "Local",
ceph-osd-information.json:          "datasource": "Local",
ceph-osd-information.json:          "datasource": "Local",
ceph-osd-information.json:          "datasource": "Local",
ceph-osd-information.json:          "datasource": "Local",
ceph-pools.json:      "name": "Local",
ceph-rgw-workload.json:      "name": "Local",
iops-by-server.json:      "name": "Local",
iscsi-client-details.json:      "name": "Local",
iscsi-overview.json:      "name": "Local",
iscsi-overview.json:          "datasource": "Local",
iscsi-overview.json:          "datasource": "Local",
iscsi-overview.json:          "datasource": "Local",
iscsi-overview.json:          "datasource": "Local",
iscsi-overview.json:          "datasource": "Local",
latency-by-server.json:      "name": "Local",
mds-performance.json:      "name": "Local",
network-usage-by-node.json:      "name": "Local",
network-usage-by-node.json:          "datasource": "Local",
network-usage-by-node.json:          "datasource": "Local",
osd-node-detail.json:      "name": "Local",

[zmc]

@dillaman, we use "Local" as our datasource name: https://github.com/ceph/cephmetrics/blob/master/ansible/roles/ceph-grafana/defaults/main.yml#L16

Did you by chance override it?

[dillaman]

Negative ghost rider — half the charts are at default and the other are Local. Consistency is important. The ansible scripts set up Prometheus as “prometheus”.

[zmc]

@dillaman can I see the inventory and any group_vars files you're using?

[dillaman]

@zmc it’s been a month since I wrote this bug. I haven’t had his environment for weeks.