I've followed the instructions to add Kibana to my BlueData 4.0 controller:
docker exec -it monitoring-<controller_ip> bash
, where is the IP address of the Controller.bdconfig --getallenv | grep bdshared_elasticsearch_admin
The system returns the username and password. For example:
export bdshared_elasticsearch_adminpass='adef3a31-fbec-474d-8b74-a50b3355399f';
export bdshared_elasticsearch_admin='elastic';
<A.B.C.D>:5610
to access the Kibana interface, where <A.B.C.D>
is either the Controller IP address, cluster IP address (if platform HA is enabled), or the IP address of a Gateway host.However, when I try to access the service on the controller A.B.C.D:5610
I get the following error (I have tried for about 15 minutes):
Kibana server is not ready yet
How can I resolve this issue?
From the primary controller (see note below), I tried to connect to elasticsearch:
[centos@ip-10-1-0-242 ~]$ curl localhost:9210
curl: (7) Failed connect to localhost:9210; Connection refused
It returned an error, so I found the epic-monitoring-** service ...
[centos@ip-10-1-0-242 ~]$ docker ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
6e61f6a9eb5f bluedata/sklearngpu:1.0 "/sbin/init" 2 weeks ago Up 32 hours bluedata-15
35e029a991ea epic/webhdfs:1.0 "/usr/bin/supervisord" 2 weeks ago Up 32 hours epic-webhdfs-10.1.0.242
55f607fa361d epic/monitoring:1.12 "/etc/start-all.sh..." 2 weeks ago Up 32 hours epic-monitoring-10.1.0.242
15b5090844ed epic/nagios:1.9 "/bin/sh -c '/usr/..." 2 weeks ago Up 32 hours epic-nagios-10.1.0.242
b380915b6c3b epic/apache-hdfs-centos:2.7.2 "/usr/local/bin/st..." 2 weeks ago Up 32 hours epic-apache-hdfs-centos
and restarted it:
[centos@ip-10-1-0-242 ~]$ docker restart 55f607fa361d
55f607fa361d
I then followed the instructions again to enable kibana, and after waiting a few minutes, I was able to login to kibana with the credentials returned by bdconfig --getallenv | grep bdshared_elasticsearch_admin
NOTE:
To find the current primary controller run:
bdconfig --hainfo
, or Role: Primary Controller