-
Notifications
You must be signed in to change notification settings - Fork 26
Grafana bridge container does not return metrics for sensors restricted to a subset of nodes
If grafana-bridge is running in a container separately from pmcollector, it does not have direct access to the pmcollector sensor configuration data. This is why either the ZIMonSensors.cfg or the mmsdrfs file must be mounted in the grafana-bridge container at startup.
There is a slight difference between the two files. ZIMonSensors.cfg includes the sensor configuration information for a local node, whereas mmsdrfs provides a cluster-wide view. For example, on a non-nsdServer node, the GPFSNSDDisk sensor is disabled, so the period for that sensor is set to 0 in the local ZIMonSensors.cfg file. The mmsdrfs file on the same node will show a sensor period greater than 0 if the sensor is enabled on any node in the cluster.
To avoid any missing data, mount mmsdrfs instead of ZIMonSensors.cfg in the grafana-bridge container at startup.
# podman run -dt -p 4242:4242,9250:9250 -e "SERVER=9.1XX.XXX.X54" -e "APIKEYVALUE=c234c7af-c2d7-43c2-97b5-3bcbddbd7a1d" -e "PORT=4242" -e "PROMETHEUS=9250" -e "PROTOCOL=http" -e "BASICAUTH=True" -e "BASICAUTHUSER=scale_admin" -e "BASICAUTHPASSW=TXlWZXJ5U3Ryb25nUGFzc3cwcmQhCg==" -e "LOGLEVEL=5" -e "RAWCOUNTERS=False" -v /tmp:/var/log/ibm_bridge_for_grafana --mount type=bind,src=mmsdrfs,target=/var/mmfs/gen/mmsdrfs,ro=true --tz CET --pod new:my-bridge-basic-auth-test-pod --name bridge-basic-auth-test scale_bridge:test_9.0.1_dev
7d735c2df8e5026209d1c817063e79949d1f07ed88af94e02983be8c7e704ec2
You can verify that the expected metrics data has been returned correctly by querying the corresponding REST API metrics endpoint via a curl call.
# podman container inspect -f '{{.NetworkSettings.IPAddress}}' bridge-basic-auth-test
10.88.0.120
# curl http://10.88.0.120:9250/metrics_gpfs_nsddisk -u scale_admin:TXlWZXJ5U3Ryb25nUGFzc3cwcmQhCg==
# HELP gpfs_nsdds_bytes_read Desc not found
# TYPE gpfs_nsdds_bytes_read counter
gpfs_nsdds_bytes_read{node="scale-11",gpfs_disk_name="disk01"} 29320192.0 1764176090000
gpfs_nsdds_bytes_read{node="scale-11",gpfs_disk_name="disk02"} 3262899200.0 1764176090000
gpfs_nsdds_bytes_read{node="scale-11",gpfs_disk_name="disk03"} 3239584768.0 1764176090000
gpfs_nsdds_bytes_read{node="scale-11",gpfs_disk_name="disk04"} 8704.0 1764176090000
gpfs_nsdds_bytes_read{node="scale-11",gpfs_disk_name="disk05"} 8704.0 1764176090000
gpfs_nsdds_bytes_read{node="scale-11",gpfs_disk_name="disk06"} 8704.0 1764176090000
gpfs_nsdds_bytes_read{node="scale-11",gpfs_disk_name="disk07"} 8704.0 1764176090000
gpfs_nsdds_bytes_read{node="scale-11",gpfs_disk_name="disk08"} 8704.0 1764176090000
gpfs_nsdds_bytes_read{node="scale-11",gpfs_disk_name="disk09"} 8704.0 1764176090000
gpfs_nsdds_bytes_read{node="scale-11",gpfs_disk_name="disk10"} 8704.0 1764176090000
...Visit the IBM Storage Scale Knowledge Center for getting more info about the latest product updates
-
- Setup classic Grafana
- Make usage of Grafana Provisioning feature
-
- Installing RedHat community-powered Grafana operator from OperatorHub
- Creating Grafana instance using the RedHat community-powered Grafana-operator
- Creating Grafana Datasorce instance from Custom Resource managed by the RedHat community powered Grafana operator
- Importing the predefined dashboard from the example dashboards collection
- Exploring Grafana WEB interface for CNSA project in a k8s OCP environment
- How to setup Grafana instance to monitor multiple IBM Storage Scale clusters running in a cloud or mixed environment
- API key authentication
- Configurable bridge settings
- CherryPy builtin HTTP server settings
- How to setup HTTPS(SSL) connection
- Start and stop grafana-bridge with systemd
- Refresh IBM Storage Scale cluster configuration data cached by grafana bridge
- Accelerate the PrometheusExporter data retrieval time
- Optimize the performance of PrometheusExporter by using scrape_job params settings
- Grafana Dashboard Panel shows no metric values for a particular entity
- Missing Grafana-Operator on an OpenShift cluster
- Missing CherryPy packages
- What to do if your system is on Python < 3.8
- Grafana-bridge fails to start with Python3.8
- Grafana-bridge container time is different from a host time
- Grafana-bridge container does not return metrics for sensors restricted to a subset of nodes
- Verify that the grafana-bridge returns data as expected