influxdata / awesome Goto Github PK

I’d like to suggest that the --rate-limit option be mentioned in this section on csv uploading: https://awesome.influxdata.com/docs/part-3/writing-and-querying-data/#writing-raw-csv-data

For all but the smallest csv files, uploading one is likely to run into write rate limits. I'm suggesting always using --rate-limit with influx write to stay under the limits.

https://awesome.influxdata.com/docs/part-2/querying-and-data-transformations/#windowing-and-aggregatewindow

The following query would return two tables with the sum for all the points in the series within a 90 min window:

But then the example only contains 1 table

Folks,
The very first paragraph on: https://awesome.influxdata.com/docs/part-2/input-format-vs-output-format/ is partially garbled. See sentence 5 within:

The InfluxDB input format is line protocol. The InfluxDB output format is Annotated CSV. The input format is different from the InfluxDB persistence. The Annotated CSV output can match the InfluxDB persistence format with simple Flux queries. However, can add Flux transformations to your query such that the Annotated CSV output doesn’t reflect the InfluxDB persistence format. Understanding these subtle differences is critical for good schema design and for using InfluxDB optimally.

In essence the final sentence tells me to worry about a concept remaining unexplained in the preceding sentences :-|

Also as a new InfluxDB cloud user I do not currently understand "InfluxDB persistence format". Is this still important for InfluxDB cloud users? Else rewriting of this section to help readers pay extra attention to critical subtleties is appreciated :-)

In this section: https://awesome.influxdata.com/docs/part-2/querying-and-data-transformations/#accessing-external-data-sources

The code for from():

sql.from(
  driverName: "postgres",
  dataSourceName: "postgresql://user:password@localhost",
  query:"SELECT * FROM TestTable"
)

I suggest that we demonstrate using import "influxdata/influxdb/secrets" throughout.

dictionaries are an important language feature, they should be covered in at leat moderate detail.

Make this content available under http://awesome.influxdata.com/

we can simplify downsampling with this

The current example uses the API directly, but users continue to ask for help with the exact same scenario using the client library.

stateDuration() is an important function. There is limited content related to it. I suggest we add a section that provides more details on how to leverage this function.

Include sum and unique sections.
Explain the difference between bare aggregators and selectors.

Suggested as this is a common recurring issue that users encounter. We should provide systematic guidance for this.

The site name is still being shown as book.github.io in places, we should change those to say "Time to Awesome" instead

Finish Readme and Home page
Include information on:

• how to run locally and contribute
• uses just-the-docs themes
• licensing
• links to community

We have some info in the introduction, perhaps some detailed walkthrough of exporting and importing dashboards would be helpful

HTML meta description should be changed from “A Jekyll theme for documentation” to something relevant to each doc page:

<meta name=“description” content=“A Jekyll theme for documentation” />
<meta property=“og:description” content=“A Jekyll theme for documentation” />

Note that the homepage has a valid description but other pages use the default.

1. What is the boundary for 'every'? i.e. it is not based on execution time, but windowing occurs based on the minute/hour hands of the clock.
2. What do _start, _stop, and _time mean? Can I change what _time means?
3. What happens if the last window is not complete. For example, I window by minute, but the last minute does not yet have data? i.e. _time is the _timestamp for the last bit of data?

Hey @mhall119 is there any desire to rename this repo from book.github.io to something like awesome or similar?

Not sure which version of the CLI made the change but --skip-header is now --skipHeader

I noticed it while working through the CSV upload in Part III:

Sometimes, downsampling tasks can fail. There are different ways to recover from this:

• use the CLI: influx task retry-failed
• query failed tasks in the system _tasks bucket and use the /tasks/.../retry endpoint to retry
• use a from() range 2-3 times as long as the aggregateWindow() so if a task run fails, it will recover the downsampled period the following task run

depending on your situation, you can use if, then, else, or dictionaries for this. here is an example with dictionaries:
import "array"
import "dict"

d = ["field1": 10.0, "field2":20.0]

array.from(rows: [{_time: 2020-01-01T00:00:00Z, _field: "field1", _value: 1.0},
{_time: 2020-01-02T00:00:00Z, _field: "field2",_value: 1.0}])
|> map(fn: (r) => ({r with _value: r._value * dict.get(dict: d, key: r._field, default: 0.0)}))

here is the same thing but using conditional logic:
import "array"

array.from(rows: [{_time: 2020-01-01T00:00:00Z, _field: "field1", _value: 1.0},
{_time: 2020-01-02T00:00:00Z, _field: "field2",_value: 1.0}])
|> map(fn: (r) => ({r with _value: if r._field == "field1"
then r._value* 10.0
else
if r._field == "field2"
then r._value * 20.0
else 0.0}))

tsia

Add a new section explaining InfluxDB's build-in secret management, how to use them from the CLI and Flux

How do you make strings that have line breaks, etc...?