mgr/dashboard: Fix Grafana Ceph Cluster health status widget

The health status widget doesn't show any status because it requires its
query to return a single result. But in case a mgr instance had failed,
it would return more, provided the incident has happened in the
requested time frame.

This is simply an issue of the `instant` switch being disabled for that
widget. As only one mgr instance can ever be providing data at a time,
enabling `instant` completely solves that issue.

Fixes: https://tracker.ceph.com/issues/51212
Signed-off-by: Patrick Seidensal <pseidensal@suse.com>

diff --git a/monitoring/grafana/dashboards/ceph-cluster.json b/monitoring/grafana/dashboards/ceph-cluster.json
index a08dc26f34f0d031ee4958f7d7888ea8123523cd..5603b064af680d2f08ddb4f36dd1786a7896cb99 100644 (file)
--- a/monitoring/grafana/dashboards/ceph-cluster.json
+++ b/monitoring/grafana/dashboards/ceph-cluster.json
@@ -109,6 +109,7 @@
         {
           "expr": "ceph_health_status{instance=~'$instance'}",
           "format": "time_series",
+          "instant": true,
           "interval": "$interval",
           "intervalFactor": 1,
           "refId": "A",