Impact 2024: The Industrial Data and AI Conference for and by Users | Nominate Speakers Now for a Ch...
@Everton Colling What an impressively detailed walkthrough. This will definitely help me set up a workaround while we wait for native support of machine states in CDF.Thank you very much!
[...] Custom Scripting within Grafana: If you're comfortable with scripting, you might explore using a custom panel that enables transforming the source data using javascript. Then you can map these integer values to their descriptive counterparts. One example of such panel is the Apache ECharts Panel for Grafana. [...] Thanks for the feedback @Everton Colling and @Glen Sykes . The machine states feature sounds like it might be what we need to address this.I wonder, if I create a custom panel, I would still need to use another data source than the Cognite data source for Grafana to retrieve the metadata from time series, no? I don’t suppose it would be good practice, or even possible, to use the CDF Javascript SDK within an Apache ECharts Panel to connect directly from the panel to CDF. Or is perhaps data models the way to go to retrieve the time series metadata in Grafana using the Cognite datasource plugin? We haven’t started using data models in CDF yet, but plan to in the near fut
@sharat @Anders Nygaard . This indeed seemed to be the problem.We have been using Cognite’s bootstrap CLI tool to set up datasets and access. It must have been a misconfiguration in the early phase of our CDF setup that caused this. I found the labels by manually setting up an access group with access not scoped to datasets.
Sure @Everton Colling. Take this time series, which is a 2bit time series, indicating the position of a switch. Where the values are most often 0 or 1 (on or off), and occasionally 2 or 3 (indicating in-between-states). The time series has metadata fields indicating the meaning of each value. In Grafana it would be useful to display these as a table, or event-list (not Cognite-style events with a duration, but instantaneous events).Something like this:Or perhaps better, something like this, with the name of the time series and its status text: These examples are realised by creating new string time series - which duplicates data from the underlying time series, and is unable to handle multiple datapoints with the same timestamp.My question is how I can extract the “statusText” metadata fields in Grafana, which would let me create such event lists without first creating a string time series.I hope this was clarifying :)
Hi, and thanks for the reply. Sure, displaying the datapoints in a jupyter notebook would be one way to access data in a tabular format. This we can easily set up ourselves, or perhaps even better a Grafana or Power BI dashboard for our domain experts.But for our self-service data consumers, I think it would be even easier if by a single click of a button in Charts or Fusion UI they would get the datapoints displayed as a scrollabale table with timestamp and value.Some of our SMEs prefer accessing data (both raw datapoints and aggregates) in a tabular/list format rather than a line plot - particluarly for data not related to industrial equipment, but to settlement, finance, etc.
Thanks @Vincent Gay .I sure hope that some funtionality for handling geodata in CDF will preserved and further developed/maintained, even though Geospatial’s future is uncertain!Assigning assets to features work fine, but I struggle to find a working syntax for retrieving the coordinates of one specific asset. Take as example the piece below, which works to lists the three features that match the constraint given by “stContains".client.geospatial.list_features( feature_type_external_id="land_test_ft", filter={ "stContains": { "property": "outline", "value": {"wkt": "Point(9 60)"} }, #"containsAny": { # "property": "assetIds", # "value": [my_asset_id] #} })But when I uncomment the four lines with the “containsAny” constraint as suggested by the documentation https://developer.cognite.com/dev/concepts/resource_types/geospatial/#searching-for-features,I get a not very helpful error in return:Additional error in
Thanks @Pasindu Perera, I can confirm that this syntax works!
Thanks @Nicholas Gerstle. For these use cases the need is only read access.Dividing into separate datasets might not be feasible when there is a partial overlap between the access need of individual groups. But I was not aware of the possibility to grant access to individual time series. I think this may be the solution I’m looking for. The only issue then would be that we would need to handle this manually - it seems that the Bootstrap CLI tool cannot provide access to individual time series. If I may also provide a feature request, I would have loved if there was a functionality in CDF Data Explorer where our data owners (owners of individual datasets) could select a group of time series (through filtering, search, asset id, ..) and then by the click of a button add these time series to the scope of the timeseries:read capability of selected groups. (The same would be relevant for assets, sequences, and events).Something like this:
Azure Managed grafana does not include the Cognite Data Fusion data source. It only has a limited set of data sournce plugins which has been whitelisted by Microsoft. We have requested to get the CDF data source whitelisted, but it has not yet been done by Microsoft. If you want to try out grafana you could also try the free tier at grafana.com which many of our customers are using. Hi @Murad Sæter. Just wanted to inquire whether you have had any progress in the dialogue with Microsoft to have the CDF data source whitelisted for Azure Managed Grafana. Azure Managed Grafana is our prefered Grafana solution, and it would be really useful to us to have the CDF datasource available there.
The issue is related to support ticket #8268 that I posted previously. The temporary workaround was to never delete any labels, but it would be nice if @Cognite can revisit this, as we do sometimes delete labels and need to recreate them later.
It works now.I was missing capabilities for sessions:list, sessions:create, sessions:deleteThanks!
Thanks. Groups setup, as in capabilities, or something else?I’ve got the dataPipelinesApiExperiment, monitoringTaskApiExperiment, alertsApiExperiment capabilities set for my user.
Thanks! I was also just made aware that Charts already by default displays data in local time, and even handles daylight savings time. It is probably how most of our users would want it, but being able to choose UTC or another timezone might also be useful.Regarding the alignment parameter, it helps, but isn’t quite what I would be after, as it would require an aggregation of data, and I suppose daylight savings time would be an issue.
Hi. We discovered a possible bug today.A chart was set up by one user The chart contained sensitive timeseries that only a small group had access to. The chart also contained a simple calculation: subtraction between two sensitive timeseries.This chart was made into a public chart. Another user, without access to the sensitive timeseries opened the chart. The sensitive time series did not appear (which is as it should be), but the calculation did - that is until he zoomed in the chart so that the calculation was done again, then the calculation disappeared.There seems to be a caching of calcluation results in the chart, and these cached results are not governed by the access privilegues tothe underlying timeseries.
Great to see the pointwise calculations! This fixes a problem for us. Our source systems have hysteresis thresholds, so that the values do not update while staying the same. This resulted in highly inaccurate visualisations when forward-filling was done on aggregated vaules. With pointwise calculations I can do a ffill interpolation on the timeseries and it displays properly.Would it be an idea to have the line display pointwise step-interpolated by default when “step interpolation” is chosen on the left side menu?
Already have an account? Login
Enter your username or e-mail address. We'll send you an e-mail with instructions to reset your password.
Sorry, we're still checking this file's contents to make sure it's safe to download. Please try again in a few minutes.
Sorry, our virus scanner detected that this file isn't safe to download.