Is your feature request related to a problem? Please describe.
I can use an existing pipeline to transfer time series data from an equinor source system to OmniaPlant Timeseries database, where it will be available for end users based on a different access protection scheme than in the source. Let's call it a privilege escalation bug.

Describe the solution you'd like
Time series tag metadata should include an access-group property
System to system pipelines, i.e. those creating tags, should include on access-group metadata to make it possible for Timeseries api to inherit/reproduce data access to it's end-users.

Describe alternatives you've considered
Don't allow transfer of data from other equinor source data systems.

Additional context

Is your feature request related to a problem? Please describe.
I want to make a timeseries in the timeseries API (e.g. the output of a machine learning algorithm), and I know some interesting things about the timeseries which I wants to store in the metadata of the generated timeseries object. Today it seems like I would have to encode all my interesting data into the "description" field, but this can get quite messy.

Describe the solution you'd like
I have 3 different alternatives in sinking order of difficulty and awesomeness:

1. Allow me to add arbitrary metadatafields which will all be searchable.
2. Allow me to add arbitrary metadatafields which will not be searchable, but I will see them when I get the timeseries object.
3. Add a few more hardcoded metadata fields which I can use for my own purpose. In this alternative maybe even give them lame names like "Info1", "Info2" etc to indicate that they have no semantic meaning.

For us point 1 is overkill, but clearly the coolest. We will be happy if we get any of them.

Describe alternatives you've considered
Encode the info into the description field. But this makes it hard to use programmatically.

Describe the bug
When querying data in batches on tags that (I think) I dont have access to I get a nice response that says which tags are 404 (not 403 but fair enough..) See screenshot 1.
This is nice since I still get data for tags that are valid and accessible

But when I send in a batch with invalid tag names, the whole batch fails, even for valid tags (see screenshot 2) This makes me lose the whole batch and

Trace ID
"traceId": "0HMT7I3PQEH7M:00000DB3"

To Reproduce
/query/data/latest?federationSource=IMS

[
    {
        "facility": "1219",
        "name":"1219-23PST0502_X",
          "endTime": "2023-09-01T05:00:00Z"
    },
    {
        "facility": "1219",
        "name":"1219-23PT0504_X",
        "endTime": "2023-09-01T05:00:00Z"
    },
     {
        "facility": "1219",
        "name":"GFA1110-23PT0504_X",
        "endTime": "2023-09-01T05:00:00Z"
    }

]

Expected behavior
I would like both requests to give the same format of error output, so that the whole batch does not get lost as in screenshot 2.

Screenshots
Screenshot 1

Screenshot 2

Criticality
Low/Medium

Is your feature request related to a problem? Please describe.
In 'CBM machine' project, one uses another agreggated function as the ones available in timeseries API, called EWMA.
Because of that, one uses extra components and storage for the sole purpose of calculating this value, which could in principle be useful also for monitoring or analytics purpose.

Describe the solution you'd like
The CBM project might need to come back with the exact requirement, but I would draft it here as : include EWMA to the available aggregate functions (together with MIN, MAX, AVG, COUNT...), as I believe it would be valuable for more than one consumers.
Using EWMA function requires an input parameter 'alpha' between 0 and 1 (as far I know).

Describe alternatives you've considered
Currently the CBM machine proof of concept uses SAP PAI for processing timeseries, and transform timeseries with a product called SAP Data intelligence (similar to Azure data factory, to transform data), and uses Python code to apply EWMA to the timeseries before applying CBM rules. This presents challenges in form of competence, introduction of new technology, costs...

Additional context
EWMA provides a way for smoothing the curve and removing 'noise' or false positives.