influxdata/influxdb

Scalable datastore for metrics, events, and real-time analytics

Patrick Callahan

@therealpxc

can I use capturing groups in InfluxDB's regexes, to put something I can use later in the query or create a virtual/temporary field or tag based on what gets captured?

I've got some data that's in kind of a bad format; some things are stored in the name

I'd like to pull that stuff out so that I can use it to clean up some output in Grafana

Patrick Callahan

@therealpxc

it does appear to accept named capturing groups in the regex themselves, but idk if it actually does anything with them

Shan Desai

@shantanoo-desai

Hi all, What is an optimal way to dump series back into the db after changing a field value? A detailed question is available on SE and is open for bounty if you are SE user

HarshaHegde1994

@HarshaHegde1994

if we dont take the backup of influxdb , but if we have whole /var/lib/influxdb folder can we restore the database

Ryan Watts

@rwatts3

Greetings, is it possible to setup an ingestion pipeline with influxdb.

for example if I would like to ingest data from an http endpoint and schedule that ingestion.
For example pulling data from google analytics http endpoint and having that data scaled to pull daily

asyousef

@asyousef

Hello. I'm building an app using Python which should collect some metrics from network devices and store it in influxdb. I need to understand which way is better and why? Send the data directly to influxdb or send it to telegraf instead.

Jagdeep Singh

@jagdeepgithub

Hey guys, I'm setting up promethues remote storage to influx.. got stuck at issue and there is an open Pr to fix this issue: influxdata/influxdb#10591 can someone please take a look and merge this?

Konstantin Bläsi

@konstantinblaesi

do you typically encode the unit of the field value into the field key for example "energy_total_kWh" or do you use tag key/values for example "energy_total_unit" => "kWh" ?

Konstantin Bläsi

@konstantinblaesi

tags make more sense to me if the unit can change (micro, milli, kilo, mega, giga, ...) :D

but I wonder if that's of any use to tools like grafana? Should I just make sure values all use the same unit and adjust them as needed?

Poul H. Sørensen

@poulhs

I perfer to use the basic SI units without any prefix, energy_total 10000000 (Wh). I have previously seen 10 kMB (10 000 Mega Bytes) because the measurement was "MB" and not "B" - to make it worse (it was an old system) and I think it should have been 10 kiMiB...
10 737 418 240 Bytes (10 240 MiB).
It may cause trouble when presenting and/or calculating stuff with different "prefix" - (swap in Bytes, memory i MB and disk space in GiB)
my .1€ : go for SI standard units and let big data handle the rest...

Poul H. Sørensen

@poulhs

And I would use the field to represent the measurement, whether it is "energy_total", "energy_total_kWh" or "energy_total_TWh"

Sebastien Leger

@regel

Open source AI for time series data, integrated with influxdb: https://github.com/regel/loudml
The project is MIT licensed, your contributions are welcome.

Sebastien Leger

@regel

Star the project if you like ;)

Ciprian Larion

@larryon

Is there a room for Chronograf? I have a question about it

Dilan Bellinghoven

@dbellinghoven

Anybody know when 1.8 is planning on being released?

Raynaldo Rivera

@R4stl1n

Is anyone online having a werid issue with the influxdb go client

query executes fine in chronograf and from the cli but when executing through the client i get this

error parsing query: invalid duration

Armel Abdelkbir

@ArABd_gitlab

Hi everyone,

I’ve nearly pulled out all my hair and on my 10th reinstall on Ubuntu. I am trying to set up a Grafana - Telegraf - Influx dB dedicated server.

The first step is to install InfluxDB. However after a vanilla install of Ubuntu, I cannot get Influx DB (having installed via the Influx DB website Sudo apt install influxdb, then the curl update).

I get the error

Failed to connect to http://localhost:8086: Get http://localhost:8086/ping: dial tcp 127.0.0.1:8086: getsockopt: connection refused Please check your connection settings and ensure ‘influxd’ is running.

Systemctl status Influxdb shows that Influx is currently running fine?

ubuntu 16.04 and influxdb 1.7.4

bobcroft

@bobcroft

@ArABd_gitlab It may not be of help to you but I installed Home Assistant (Hass.io) on a vanilla install of Ubuntu 18.04.2 . Hass.Io installs itself in Docker containers, then once installed you can install Influx and Grafana as Addons. Although I haven't do it myself I believe you can install Telegraph as well.
I am no expert in Linux or docker but the Hass.IO system works brilliantly well and is very well supported. It may be a method you could adopt to achieve your goals.

Raynaldo Rivera

@R4stl1n

quick question

Is there any chance that when writing large data segments, a application can grab the incomplete batch write data in the middle of the transaction?

Or is influx working like other sql databases where the data isn't available until after the full batch points is written?

lmagdanello

@lmagdanello

Hi guys, can someone help me? I have InfluxDB - Telegraf - Grafana setup on a high avabiliaty cluster, with Lustre Filesystem on them. My telegraf agent is collecting filesystem shared stats, but on the database on InfluxDB there are none of those data.

I'm using inputs.lustre2 for this

Vitaly Zhuravlev

@v-zhuravlev

Hi, guys, can't get it to work. What is the best practice to find when was the last time measurement had no data?

using InfluxQL

Like metrics are coming every 10 seconds. I need to find 'gaps' in my time series

Konstantin Bläsi

@konstantinblaesi

any idea how to make use of this PR to support month/year tags in influxdb ? influxdata/telegraf#5895

switlen

@switlen

Hi ,guys,when I use InfluxDB,I meet a problem ,I have created a measurement,and it has stored many production data,but now I want to change the measurement name,I find the ALTER command does not work well,my InfluxQL is ALTER measurement "my measurement name" rename to "another Name",has anyone tell me what I should do to get expectancy?thank you very much

Konstantin Bläsi

@konstantinblaesi

If I want to extend an existing measurement with two tags year and month is there already a tool that can do that? Either update data or write a new measurement with the same data + those tags calculated usiing the timestamp?
I know telegraph has recently merged a patch that can do this, but a) I already have months of data and b) my current setup is a nodejs app listening to mqtt and storing the values to influxdb
referring to this PR influxdata/telegraf#5895

Rafael Gonzaga

@RafaelGSS

Hello everyone! I have some doubt about usage of InfluxDB... I already read the documentation getting started, maybe us can help me:

I have the following JSON data (I thinking change my current database MongoDB to InfluxDB):

{
    datetime: "2018-01-01 00:11:11",
    latitude: -3.3333,
    longitude: -4.5555,
    locales: [
      {
          id: 1,
          name: "Any Name"
      }
    ]

I'll save all data based on this schema. My question is:

How I can save the array of locales based on example above ? Is like relational data, by join ?
The InfluxDB or any time series database is recommended to use with this data scheme ? Are data of lightning in some country (maybe can has 1.2bi of data)

Thanks so much!

kalidasya

@kalidasya

@RafaelGSS influxdb is not a nosql database

dupondje

@dupondje

No Telegraf dev in here? :)

Olav Rossøy

@OlavRossoy

is there any way to make a select with a group by time (days) with a tz return only the date portion of the timestamp?
ie have
SELECT mean("value") AS "value" FROM some_data GROUP BY time(1d) tz('Europe/Berlin')
return dates like '2019-07-05' instead of '2019-07-04T22:00:00Z'

its kind of awkward having to transfer dates to local timestamps to be able to query time ranges correctly for aggregated values created with CQs etc

Tushar Nerkar

@tusharnerkar8_gitlab

Our data looks like this and there are 20000 nodes

node     status             time
node-1  status: down      1 hr before
node-1     status: up        now

select count(*) from node_health where status = 'down' and $timeFilter group by node

I want the above query to return 0 but it doesn't.
I think I can solve this by following query.

select count(*) from (select last(status) from node_health where $timeFilter group by node) where status = 'down'

But I wanted improve the performance by adding the status where clause . My query is real slow takes 50 seconds to load.
How to make this work without subquery so as I can improve the performance of the query.

mgiammarco

@mgiammarco

Hi, I try to send an alert from influxdb to alerta but I get always: unsupported protocol scheme "" when I click on "send test alert" button no matter what I do

I have also launched tcpdump and I am pretty sure that kapacitor/chronograf does not even try to send a POST to alerta

can you help me?

thanks

zoe56

@kongkong56

what meaning difference between mean and last

Poul H. Sørensen

@poulhs

@mgiammarco Sounds to me like you have a misconfigured setup

[alerta]
  enabled = true
  url = "https://alerta.api.example"
  token = "GuvfVfGurFrpergNyregnNCVgbxra"
 ...

And now a telegraf question:
Hi - We need to monitor Horizon DaaS.
I hope that someone have some exeperiences with this.
A more general solution would be a CIM plugin (which I have not found)
Any ideas?

Poul H. Sørensen

@poulhs

influxdata/telegraf#6582

Where communities thrive

People

Repo info

Activity