Currently, the plugin only supports remote CSV files that are available over HTTP. There's no support for loading CSV files from disk.

The grafana version: v7.2.1

the csv file: http://10.67.230.111:443/Capacity.csv

Storage,Type,Total,Used,Free
CN-test01,Total,107.047,86.149,20.897
CN-test01,FC,51.188,36.714,14.474
CN-test01,SSD,55.859,49.436,6.424
CN-test02,Total,212.646,157.699,54.947
CN-test02,FC,212.646,157.699,54.947
CN-test03,Total,96.135,88.646,7.489
CN-test03,NL,96.135,88.646,7.489

The grafana detail error

An unexpected error happened
Details
Error: Minified React error #130; visit https://reactjs.org/docs/error-decoder.html?invariant=130&args[]=undefined&args[]= for the full message or use the non-minified dev environment for full errors and additional helpful warnings.

    in Unknown
    in t
    in t
    in div
    in div
    in div
    in Unknown
    in x
    in ho
    in yo
    in QueryOperationRow
    in div
    in t
    in div
    in Unknown
    in x
    in x
    in l
    in Tn
    in t
    in Cn
    in t
    in div
    in div
    in div
    in div
    in div
    in t
    in t
    in t
    in div
    in Mo
    in div
    in dn
    in div
    in div
    in t
    in div
    in t
    in div
    in div
    in t
    in x
    in div
    in t
    in x
    in r
    in Unknown
    in t
    in Unknown
    in h
    in Unknown
    in Unknown
    in t
    in t
    in l

Could you please take a look at it ?

Looking at the addition of selecting field names using regex (the example of /Bid/ giving all Bid* columns). This was covered in a little more detail on the discussion here

In our old data visualization system (a bunch of JS and HighCharts) we split the time series into multiple ones by their year, so we could plot a year-over-year chart

Like this

I'm thinking how I could replicate a similar thing in Grafana and the built-in timeshift feature was implemented by the datasource. I would probably try stacking multiple -1y, -2y, -3y etc queries. It would be manual, but it would work.

Outputting multiple time series directly by a specified time interval would be best but I don't know if that would be outside scope of this plugin.

(The new Grafana time series transformations could have something like this but as far I can see, they don't in their current featureset. In fact, they seem to be made with the opposite idea: merging series. )

correct link : https://grafana.com/grafana/plugins/marcusolsson-csv-datasource/installation
link found in the about section of the plugin in your github page : https://grafana.com/plugins/grafana-csv-datasource

hope this helps

Hello,

I've had to restart our Grafana docker today and for some reason I get:

"local mode has been disabled by your administrator"

If I go to my datasource and test it says it's working.

Using Grafana 8.0.1 and Sign CSV 0.6.1

Any help would be great!

Hello,

Apologies as it’s not an issue, but wasn’t sure how to contact you.

What a great plugin, I’ve been using it for a while on my Linux Grafana boxer work.

How do I check what version I’m using?
How can I upgrade to the latest?

thanks

hi,
got this issue while retrieving my csv
Get "http://192.168.10.239:12345/test.csv": net/http: invalid header field name ""
my csv looks like :
Time,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,chaud Z1,chaud Z2,chaud Z3,froid Z1,froid Z2,froid Z3,regul Z1,regul Z2,regul Z3 2020-12-03 10:30:30,15.32,15.42,14.31,15.12,15.62,15.62,14.51,14.71,16.02,14.84,15.22,14.71,14.51,15.42,15.01,16.02,14.71,15.42,15.42,14.54,14.94,15.42,15.52,15.09,15.04,250.37,249.62,250.65,249.89,250.68,250.98,250.10,250.01,250.06 2020-12-03 10:30:35,15.32,15.42,14.31,15.12,15.62,15.62,14.61,14.71,16.02,14.84,15.22,14.71,14.51,15.42,15.01,16.02,14.71,15.42,15.42,14.54,14.84,15.42,15.52,15.09,14.94,250.37,249.62,250.66,249.89,250.67,250.97,250.10,250.02,250.05 2020-12-03 10:30:40,15.32,15.42,14.31,15.12,15.62,15.62,14.61,14.61,16.02,14.84,15.22,14.71,14.51,15.42,15.01,16.02,14.71,15.42,15.42,14.54,14.84,15.42,15.52,15.09,14.94,250.37,249.62,250.66,249.88,250.67,250.97,250.07,250.02,250.07 2020-12-03 10:30:45,15.32,15.42,14.31,15.12,15.62,15.62,14.61,14.61,16.02,14.84,15.32,14.71,14.51,15.42,15.01,16.02,14.71,15.42,15.42,14.54,14.84,15.42,15.52,15.09,14.94,250.37,249.62,250.66,249.88,250.67,250.98,250.04,250.02,250.07
meaning the dates are not in colums but in lines (as most of the CSV logging somethign does.
is your plugin able to withstand this kind of presentation ?

I've been having lots of issues to make my own release with the new changes, if you could click the button on your end it would be greatly appreciated :)

I updated the marcusolsson-csv-datasource plugin to try to use the new regex support in the field names.
After the update, the plugin does not start anymore. The error message is the following.

May 28 08:10:19 linux-7x33.suse grafana-server[32040]: t=2021-05-28T08:10:19+0200 lvl=eror msg="Failed to start plugin" logger=plugins.backend pluginId=grafana-csv-plugin error="Unrecognized remote plugin message: \n\nThis usually means that the plugin is either invalid or simply\nneeds to be recompiled to support the latest protocol."

What did I wrong, or what can I do to resolve this?
I tried to remove and re-add the plugin, but I got no change. :-(
Thank you very much for supporting.

I successfully imported my datasource, headers are recognized.

h1,h2
v1,v3
v1,v3
v2,v3

I would like to make a piechart out of one column. It should count all different values in this column.

For the data above, if I select column h1, it should return a piechart with v1:66% and v2:33%.

I am stuck at the point where it successfully count all values but does not make a difference between v1 and v2.

Can you help me achieve that ?

I’ve spent weeks trying to get my Citrix Powershell outputs into Grafana. Some I have via the Influx Powershell module, but others I haven’t. Now I simply export to csv to my IIS web server every 5 mins and import using this plugin and all is good. Thanks!!

One think it didn’t wonder about it thresholds/Colour/Mappings. I have a some cells in my CSV that say “failed” which is correct but I’d like a way to show that cell as Red and others with different colours.

Anyway fantastic work!

Some users have reported errors when loading local files:

• #4 (comment)
• #36 (comment)

From the log lines in this comment, it seems like Grafana is trying to parse the local file path it as a URL.

The plugin does use the URL field in the data source config to store the file path since that's only way to display the path in the data source list. You can see this by switching between HTTP and Local.

My initial thought was that recent versions of Grafana had introduced additional validation for the URL field. Unfortunately, I seem to be unable to reproduce this on my machine. I've tried both Brave and Safari, for the following versions:

• 7.3.7
• 7.4.0
• 7.4.1

I'd like to figure out under what circumstances this happens. If not, I'll have to stop reusing the URL field for the file path. This will make it hard to find the right data source from the list though.

Hello,

As requested, there seems to be an issue with this plugin if it's on a version greater than 0.1.0 if the csv is hosted in IIS.

Please read #31

If I'm on 0.1.0 all works fine, if I upgrade to 0.3.3 they stop working. If I run locally, in Gish, on a Python Website all works, it's just IIS.

Thanks

Hello-
While attempting to graph sample data from https://raw.githubusercontent.com/CSSEGISandData/COVID-19/master/csse_covid_19_data/csse_covid_19_time_series/time_series_covid19_deaths_global.csv I observe my Grafana dashboard message: Unable to graph data

When I switch to table view, yes the data is visible, but isn't the whole point to graph data with Grafana's amazing UI?

Surely this is available now, is it not?

See screen capture...

When using "path" the plugin keeps warning that the handler is invalid. The plugin returns data even with this warning.

The data source returns Success.

I always get the error "Plugin Health Check Failed" with my own CSV file as well as with your example file from the other case

Is it true that it is unable to graph data from the grafana-csv-datasource? I got:

And there is no way I found to switch it to graph view, only table view.

I was having gaps in my charts and quickly found the culprit using the table view

Viewing as string

Viewing as time

The original format is arguably pretty bad, however, Grafana itself via the override date format understands it no problem, but the plugin Time type does not.

Hi,

I'm in the situation that I need to store JMeter test results as CSV files instead of into DB.

Would I be able to somehow compare 2 or more load tests stored in CSV files with this grafana-csv-datasource?
I mean, to compare in the Graph panel, which requires a Time field and a Number field. The problem to me is that different tests have different timestamps. So,

• How can I chart 2 or more load tests in the same Graph panel?
• How to make sure they line up nicely instead each in their own time period?

Thx!

This has some interesting HTTP client config based on standard settings -- it would be great if the standard setting were easy to use from the SDK.

perhaps something like:
https://github.com/grafana/grafana-plugin-sdk-go/blob/master/experimental/rest_client.go#L33

Windows server 2019 Datacenter Version 1809, OS build 17763.1817.
grafana-csv-datasource-0.41
Downloaded plugin as zip file, installed in plugins, restarted server.

Error: Minified React error #130; visit https://reactjs.org/docs/error-decoder.html?invariant=130&args[]=undefined&args[]= for the full message or use the non-minified dev environment for full errors and additional helpful warnings.

in Unknown
in div
in t
in form
in div
in div
in t
in div
in div
in div
in t
in t
in div
in t
in t
in x
in ConnectWithCleanUp(Connect(t))
in r
in Unknown
in t
in Unknown
in h
in Unknown
in Unknown
in t
in t
in l

First Grate plugin - really a missing part of the Grafana toolbox, and now it is here ;)

Not sure if it is a bug or feature request or Plotly panel issue, sorry for that.

I'm currently trying to visualize data from the CSV plugin via the Plotly Plugin panel (AE3E) but it i'm not able to make it work.
The CSV plugin returns data in the Table plugin and also other plugins as seen below

The data is normally accessed with "data.series[0].fields[1..x].values.buffer" in the plotly panel plugin, but It does not return any data(so it seams).

Is there a other way to call the data via the Script?

It would be nice if one could select the decimal delimiter in the data source settings.
In Germany some COVID-Tables are filled with numbers like 45,67.
Grafana cannot interpret these values as numbers, only as strings.
Thus its impossible to color the values. =(

Mentioned on the discussion page and seems this would be a good request, will outline below.

The possibility of selecting csvs from a folder and setting this as the data source (the data source would be linked to a csv folder for example). Then when adding a panel and selecting this data source, it would give a drop down of the csvs within the csv folder from the data source and from here you can select which csv you want to use?

The discussion link is here

I have the grafana cvs plugin working nicely. Just wondering whether it is possible to show the data as cumulative graph.

Would it be possible by using Grafana's transform functionality?

Or with the cvs plugin - expression functionality?
I had a look at this, but I don't understand how it works.
Are there examples of how to use this?

E.g. When I have a table with the following columns:
Timestamp,Total,Team1,Team2

Would it be possible to calculate the value for Team3 as: Team3 = Total - Team1 - Team2

Is it possible to negate a value, for example:
Team4 = Team2 * -1

Hello there,

I'm quite uncomfortable posting an issue for something that might not be... Data is not updating for me. Eventhough my source CSV changes, nothing changes in the dashboard. Is the plugin downloading the target file on each refresh ? Or does he download the file just once and then only display the cached file ?

Thank you.

Regards,

Baptiste

I have a local .csv file of the type
time,value
2021-07-02 11:00:00, 101
2021-07-02 11:03:06, 113
2021-07-02 11:05:53, 120
2021-07-02 11:07:32, 132

When I import it in Grafana, the table view shows the data as follows:

But when I try to view it as time series, it shows No Graphable Fields. the current system time is 11:40, therefore time is not an issue.

I'm looking to run this locally and so am trying to mount my csv file to the Docker container such that it is available to the Grafana server. I have tried various paths in place of WHERE? but I haven't succeeded in finding the right path. Is it home, data, plugins?

docker run --rm --name grafana -p 3000:3000 -e GF_INSTALL_PLUGINS=marcusolsson-csv-datasource -v $PWD/:/WHERE?/ grafana/grafana:7.5.2

installing marcusolsson-csv-datasource @ 0.4.1
from: https://grafana.com/api/plugins/marcusolsson-csv-datasource/versions/0.4.1/download
into: /var/lib/grafana/plugins

✔ Installed marcusolsson-csv-datasource successfully

Restart grafana after installing plugins . <service grafana-server restart>
...
t=2021-04-15T20:48:08+0000 lvl=info msg="Path Home" logger=settings path=/usr/share/grafana
t=2021-04-15T20:48:08+0000 lvl=info msg="Path Data" logger=settings path=/var/lib/grafana
t=2021-04-15T20:48:08+0000 lvl=info msg="Path Logs" logger=settings path=/var/log/grafana
t=2021-04-15T20:48:08+0000 lvl=info msg="Path Plugins" logger=settings path=/var/lib/grafana/plugins
t=2021-04-15T20:48:08+0000 lvl=info msg="Path Provisioning" logger=settings path=/etc/grafana/provisioning
....

After applying a transformation, Max returns a value, Mean returns NaN.
Field 5 Max
39.23
Field 5 Count
10.00
Field 5 Mean
NaN

Reported in #6 (comment):

On Windows 10, importing files with backslashes in the path fails, e.g. c:\temp\file.csv, whereas importing files with forward slashes succeeds, e.g. c:/temp/file.csv

Can we add "Supported Time Formats" to the documentation somewhere? I struggled with this until I found the parser in detectTimeLayoutNaive.

If I'm not mistaken, the formats compatible with this plugin are as follows:

		"2006-01-02 15:04",
		"2006-01-02 15:04:05 MST",
		"2006-01-02 15:04:05.999999",
		"2006-01-02 15:04:05.999999 -07:00",
		"2006-01-02 15:04:05.999999Z",
		"2006-01-02T15:04",
		"2006-01-02T15:04:05 MST",
		"2006-01-02T15:04:05.999999",
		"2006-01-02T15:04:05.999999 -07:00",
		"2006-01-02T15:04:05.999999Z",
		"2006/1/2",
		"2006/01/02",
		"2006/01/02 15:04",
		"2006/01/02 15:04:05 MST",
		"2006/01/02 15:04:05.999999",
		"2006/01/02 15:04:05.999999 -07:00",
		"2006/01/02 15:04:05.999999Z",
		"2006/01/02T15:04",
		"2006/01/02T15:04:05 MST",
		"2006/01/02T15:04:05.999999",
		"2006/01/02T15:04:05.999999 -07:00",
		"2006/01/02T15:04:05.999999Z",
		"01/02/2006",
		"01/02/2006 15:04",
		"01/02/2006 15:04:05 MST",
		"01/02/2006 15:04:05.999999",
		"01/02/2006 15:04:05.999999 -07:00",
		"01/02/2006 15:04:05.999999Z",
		"01/02/2006T15:04",
		"01/02/2006T15:04:05 MST",
		"01/02/2006T15:04:05.999999",
		"01/02/2006T15:04:05.999999 -07:00",
		"01/02/2006T15:04:05.999999Z",```

Please consider adding support for csvq. This will allow for using an SQL-like query language for CSV data.

One benefit from would include the ability to add a delta column to the CSV table using SQL LAG() or LEAD() functions.

Example:

Original CSV, named test.csv:

Date, Processed
2021-04-15, 300
2021-04-14, 275

Augmented output with csvq could then be rendered in Grafana like this:

csvq 'select Date, Processed, LAG(Processed) over (order by Date) as Previous, (Processed-Previous) as Delta from test order by Date desc'

bash-5.0$ grafana-server -v
Version 6.7.2 (commit: 423a25fc32, branch: HEAD)

when creating a datasource based on the plugin, page http://172.22.0.6:3000/datasources/edit/2/ displays:

b.LegacyForms is undefined Evaluating http://172.22.0.6:3000/public/plugins/marcusolsson-csv-datasource/module.js Loading plugins/marcusolsson-csv-datasource/module
Data Source Error

actually i was using a time series data after using my query i am getting data into table but unable to fetch it on time series graph it keeps on showing as data doesnot have a time field.

Just installed Grafana 7.4.0. Installed plugin. Shows disabled because of "Modified signature":

We have encountered data source backend plugins that are unsigned. Grafana Labs cannot guarantee the integrity of unsigned plugins and recommends using signed plugins only.
The following plugins are disabled and not shown in the list below:

marcusolsson-csv-datasource

Modified signature - Valid signature but content has been modified

Note that unsigned front-end datasource plugins are still usable, but this is subject to change in the upcoming releases of Grafana

With other data sources it is possible to alter the time interval using a variable.

The interval could be changed to for example 1 day, 1 week, 2 week, 1 month.

Is this possible with Grafana cvs plugin? If not, will it be made available?

Just tried the plugin with the covid test dataset documented on the front page.

After loading the dataset and going to "Explore" in grafana, then selecting the data, the following error appears:

record on line 57: wrong number of fields

Not sure if there's a parsing issue or if the csv format is bad but wanted to share the error.
Note, loading the natural gas sample csv works.

From: #6 (comment)

On Windows 10, importing files with backslashes in the path fails, e.g. c:\temp\file.csv, whereas importing files with forward slashes succeeds, e.g. c:/temp/file.csv

Confirmed in v0.3.0: #6 (comment)

More details:

I added the filepath.ToSlash that looks like what we need, but it seems like the problem persists.

I need to get a development environment set up on a Windows machine to test this, so not sure when I'll be able to look into this.

If you have a Go environment set up on your Windows machine, I'd appreciate help testing this.

first

bash-5.0$ grafana-server -v
Version 6.7.2 (commit: 423a25fc32, branch: HEAD)

then

bash-5.0$ grafana-cli plugins install 'marcusolsson-csv-datasource'
Error: ✗ Failed to find requested plugin, check if the plugin_id is correct: 404 not found error

whereas

bash-5.0$ grafana-cli plugins install grafana-piechart-panel
installing grafana-piechart-panel @ 1.6.1
from: https://grafana.com/api/plugins/grafana-piechart-panel/versions/1.6.1/download
into: /var/lib/grafana/plugins

✔ Installed grafana-piechart-panel successfully 

Restart grafana after installing plugins . <service grafana-server restart>

I'm keen on trying out the csv data source in grafana as I have an endpoint which can produce csv data if you add the

Accept: text/csv

http header. (If the header is missing it complains)

As of the moment the plugin seem to ignore the custom header section in the config, it might be an option to add the mime type text/csv as a default to the performed http request, as this anyway might always be the expected mime type for this plugin.

This is not a big issue, but may be easy to address. My csv was failing and it took me a bit of time to realise that I had got a 'space' at the end of each line. If this was identified in the fail error message someway it would have alerted me to my formatting error.

I have installed the plugin and when I add a CSV datasource it is looking for an http url.

I originally was hoping to access the CSV files locally on the server give that the section header is http I am guessing file:// URI schemes wouldn't be supported in the URL field.

Is there another server component to this plugin kind of like CSVServer or is this just the URL to call that responds with CSV data in the body of the response?

Hi, I'm running Grafana inside a Jail on my TrueNAS server and wanted to try this plugin out. But in the log I see the following error:

t=2021-01-18T17:39:51+0100 lvl=eror msg="Failed to start plugin" logger=plugins.backend pluginId=marcusolsson-csv-datasource error="fork/exec /var/db/grafana/plugins/marcusolsson-csv-datasource/gpx_csv-datasource_freebsd_amd64: no such file or directory"

And indeed there is no such file, so I figure this hasn't been made available yet for FreeBSD. Do you plan on making that file so I can test this?

Thank you for your reply.

Hello Marcus!

Is it possible to compile this plugin for ARMv7 (raspberry pi)?

Thx and BR
Rainer

Hello-

While running Grafana locally, via chrome browser, I would like to download the ZIP and install on my PC. Where exactly can I find my plugins directory, and where should I un-zip the file? Your instruction say:

Alternatively, you can manually download the .zip file and unpack it into your grafana plugins directory.

Thanks.

Feature Request: Add support for custom parameters in the panel configuration.
Thanks!

Hi!
I'm tying to "Explore" my very simple csv file using this Dropbox URL: https://www.dropbox.com/s/xqprwqdu6z25uy4/temp.csv?dl=0
While creating the Data Source with the given URL, it says that everything is fine and tested.
But when calling "Explore", the only thing I get is
record on line 3: wrong number of fields

Hi,

I have a CSV that only has two fields and two rows. I have placed this csv on a public git repo and confirmed that I am able to access it manually without a problem.

I added the url and tested, which resulted in success. However, in the dashboard it keeps saying
record on line 74: wrong number of fields
I also tried your example csv from John Hopkins data. That also results in the same error.

when i use CSV datasource to set some variables，it does not work