I would like to get a table in my grafana dashboard that lists all annotations shown on the dashboard (gathered via dashboard configuration and (if possible) also the ones contained in panels.
Any idea? I cannot find a detailed documentation on that...
Related
I have a test Grafana setup pointing to my AWS CloudWatch service and CloudWatch logs. I can setup a graph pointing to a specific log group or groups, but what I want is to be able to dynamically set the logs groups, e.g., aws/containerinsights/$cluster/application. But I can't see how to do that or whether its even possible. The log group selection appears to be only from a dropdown and can't be edited. Is this possible? If it's not, I don't see how you can create a dashboard that can be used across clusters.
Usually, what looks like a dropdown (for example also dashboard variable) in the Grafana also takes user input - you can write/paste own text there. So write desired string with dashboard variable manually and then use offered Create: ... option:
We are using grafana to visualize our metrics and very glad to use it.
For one dashboard I need to create custom variable to show host name. For example, we have the first variable: host with query collectd.localhost.ping.ping-* and regexp /ping-(.*)/. It returns a list of hosts that we ping. Now I want to add showing title depends on selected host. I think need to add another custom variable title like map host1:Title1;host2:Title2 and in the panel's title write something like Proxy ${title:json}.${host} in result get Proxy TitleN.
How to do it? We use 7.2.1 grafana version.
Thank you!
I'm not sure if this is a great answer, but given it's been nearly two years without any answer, I'll proffer it. Having just commented how I was surprised no one had in fact answered, I kept poking around. Eventually I found that if added a variable interactively, and then I clicked dashboard settings followed by JSON Model I got to a description that included new data in the templating field. I pasted that into my original JSON description of the dashboard, and restarted Grafana --- the change persisted.
I'm trying to configure a Grafana dashboard for multiple Prometheus datasources. These two datasources are present in two different servers.
Prometheus scrape the metrics from three jobs, prometheus itself, node-exporter for host metrics, advisor for docker containers metrics.
This is what prometheus.yaml looks like:
global:
scrape_interval: 5s
scrape_configs:
- job_name: 'prometheus'
static_configs:
- targets: ['localhost:9090']
- job_name: 'node-exporter'
static_configs:
- targets: ['node-exporter:9100']
- job_name: 'cadvisor'
static_configs:
- targets: ['cadvisor:8080']
The same yaml configuration is being used in both the servers.
I created two datasources in Grafana UI and their respective names are Prometheus, and Prometheus-Certify. The first datasource is the default one. I'm using Grafana v7.1.0.
This is what my dashboard looks like for the datasource Prometheus:
There are other panels as well, but showing these panels would be enough to explain the issue.
If I click on the Host dropdown option, then I can see the hosts cadvisor:8080, localhost:9090, and node-exporter:9100.
Requirement:
What I'm trying to achieve is to configure this dashboard for both the data sources. I can see all the options in the datasource dropdown menu default, Prometheus, Prometheus-Certify. I want the dashboard to show me the data from a particular source when I select that particular datasource. And this is where I'm getting stuck.
I did my research on this and found that we can show data from two different sources at the same time in a panel. I saw this option in the Grafana documentation and also this thread which actually gets me closer to what I'm trying to achieve.
So I followed this approach and configured --Mixed-- datasources for a panel.
It worked flawlessly and performing the same operation on different panels resulted in this:
Though it worked but, this is not what I was trying to do. The problems here are:
I'll have to do this for every panel in order to see data from both the sources.
It shows data from both sources at the same time. I want to see data from the selected datasource only, not both.
If I add another datasource in the future, I'll have to perform all the operations all over again.
I saw [templates and variables][6] being suggested in many places. I gave it a shot but it didn't do anything.
If changing variables is the way to go then please provide suggestions specific to this issue only as I've already gone through that section and couldn't find anything specific to this.
I hope I was able to explain my problem to you people. I just want to see the data from the selected datasource without creating a new dashboard. Your help would be very much appreciated. Thank you.
I realized that the dashboard I imported creates a variable DS_PROMETHEUS, and its type is Datasource.
This variable holds the value of the selected datasource. I was unable to see the data from the selected datasource because the datasource was kinda hardcoded in the panels' queries.
As you can see in the image above the preselected datasource is Prometheus. Changing this value to $DS_PROMETHEUS did the job.
So it all boiled down to how do I make this variable as the datasource for every query in every panel? The question now was how to run a query/command to modify settings of multiple panels in one go.
After doing some research I found this thread. It has been categorized as a feature request. Person/People have already developed a tool to handle these kinds of scenarios. As of now, I haven't used this tool but I'll update my answer once I use it.
The simple and the old-school way was to replace prometheus with $DS_PROMETHEUS everywhere in the dashboard's JSON. So I did that and Voila! it worked.
For me disabling a default Prometheus datasource did the trick. Without a default data source you can select the Prometheus datasource in the query.
Please see picture,
I didn't add Grafana, Grafana metrics, Graphite Carbon metrics... I don't know where these dashboards are defined but I need to remove them
You can go to the Manage Dashboards page, select the ones you don't want and delete them.
I cannot seem to figure out how to get certain metrics from GCP into Stackdriver (google monitoring) in a usable way. They can be viewed using Stackdriver's "Metrics Explorer" tool, but not saved into a graph or alerting policy. As a specific example, only a handful of the metrics outlined in this table are available:
https://cloud.google.com/monitoring/api/metrics_gcp#gcp-container
Again, I can use the "Metrics Explorer" tool to immediately visualize any one of them in an ad-hoc graph, but I cannot create an alerting policy or any sort of persistent monitoring for anything except for CPU Usage, Disk Usage, Page Faults, and Used Memory. Does anyone know how to get one of these metrics (such as container/cpu/usage_time) into an alerting policy?
Metrics Explorer includes access to more metrics than are currently available for alerting. We're working on addressing this both for UIs and APIs. Please stay tuned to release notes.
Thanks for using Stackdriver!
You can create a dashboard and add those metrics that you want to view. To do so please follow the instructions below. Further information on dashboards and on creating charts and alerts can be found in this documentation link.
At the Google Cloud Platform click on Products & Services which is the icon with the four bars at the top left hand corner.
On the menu go to the Stackdriver section and click on Monitoring.
A new ‘Stackdriver Monitoring’ window for your project will open up.
At the left-side navigation list click on Dashboards and select Create Dashboard. During this step you also have the option to add a chart to an existing dashboard by clicking on its name.
Give a name to the dashboard you’re creating.
At the right-hand side you can see different options to customize. Click on ‘Add Chart’ and fill in the requested information:
Give a name to the chart you’re creating.
Resource Type: Choose your resource type, e.g ‘GCE VM Instance’.
Metric: Choose the metrics that you want to add, e.g ‘compute.googleapis.com/instance/uptime ‘. Metric information for each product can be found in this documentation link.
Fill in all other information as needed: Filter, ‘Group By’, etc.
You can add more metrics by clicking the ‘Add Metric’ button.
Once you have finished click on the ‘Save’ button to save and create this new chart.
At the left-hand menu list click Dashboard and select the dashboard you have created to view it.