I should have checked my branch before I sent the mail! Thanks John/Ricardo. This is looking good - nice work, I especially like the integrated API docs! However, back to the grafana piece, I'm not sure where I should see the grafana integration - should a link appear? I imported the ceph-cluster.json dashboard from monitoring/grafana/dashboards and checked that it displayed natively in grafana before making the root_url change to grafana.ini. I'm using grafana (v5.2) and prom as containers (host networking) on the same host, and here are my settings; $ grep root_url /opt/docker/grafana-prom/etc/grafana.ini root_url = http://192.168.122.114:3000/api/grafana/proxy $ bin/ceph dashboard get-grafana-api-url http://192.168.122.114:3000 $ bin/ceph dashboard get-grafana-api-auth-method password $ bin/ceph dashboard get-grafana-api-password admin $ bin/ceph dashboard get-grafana-api-username admin If I try to access the grafana instance (localhost:3000) from a browser, I get a json error message - {"message":"Not found"}, and the url is rewritten to http://192.168.122.114:3000/api/grafana/proxy/login I know the dashboard is there - curl -u admin:admin -XGET http://localhost:3000/api/search?query=Ceph%20-%20Cluster [{"id":1,"uid":"yetcefHmk","title":"Ceph - Cluster","uri":"db/ceph-cluster","url":"/api/grafana/proxy/d/yetcefHmk/ceph-cluster","type":"dash-db","tags":["ceph","cluster"],"isStarred":false}] $ curl -u admin:admin -XGET http://localhost:3000/api/dashboards/uid/yetcefHmk | json_pp | head -n 20 % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 100 18142 0 18142 0 0 343k 0 --:--:-- --:--:-- --:--:-- 347k { "dashboard" : { "annotations" : { "list" : [] }, "graphTooltip" : 0, "links" : [], "refresh" : "30s", "schemaVersion" : 16, "__requires" : [ { "name" : "Grafana", "version" : "5.0.0", "id" : "grafana", "type" : "grafana" }, { "type" : "panel", "version" : "5.0.0", "id" : "graph", <snip> So once again, what dumb thing am I doing? Also, when you enable/disable mgr modules you get thrown out of the UI as the mgr reloads. I'm assuming this is a known behavior (?) On Thu, Jun 21, 2018 at 10:48 PM Ricardo Marques <rimarques@xxxxxxxx> wrote: > > Hi Paul, > > On 06/21/2018 01:21 AM, Paul Cuzner wrote: > > Hi, > > > > I tried today to follow the grafana installation steps[1] (based on > > master, from yesterday), but it failed with command not found. The > > docs show grafana* variables that you need to set with ceph dashboard > > set-* but when I try this I just get invalid command. > > Which command are you executing exactly? > > I've tried this on latest master and it worked: > > # bin/ceph dashboard set-grafana-api-url http://localhost:3001 > *** DEVELOPER MODE: setting PATH, PYTHONPATH and LD_LIBRARY_PATH *** > 2018-06-21 10:43:09.483 7ff913cbf700 -1 WARNING: all dangerous and > experimental features are enabled. > 2018-06-21 10:43:09.503 7ff913cbf700 -1 WARNING: all dangerous and > experimental features are enabled. > Option GRAFANA_API_URL updated > > > > > When I look at the dashboard/module.py or settings.py, I don't see the > > grafana variables being defined. > > > > Any pointers? > > > > PC > > > > [1] http://docs.ceph.com/docs/master/mgr/dashboard/ > > -- > > To unsubscribe from this list: send the line "unsubscribe ceph-devel" in > > the body of a message to majordomo@xxxxxxxxxxxxxxx > > More majordomo info at http://vger.kernel.org/majordomo-info.html > > > > -- > Best Regards, > Ricardo Marques > SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton, > HRB 21284 (AG Nürnberg) -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html