How to Install InfluxDB on Ubuntu 26.04 (Step-by-Step)

InfluxDB is one of the most popular time series databases, built for data that arrives with a timestamp: server metrics, IoT sensor readings, application performance data and event logs. This guide installs InfluxDB on Ubuntu 26.04 LTS using the official InfluxData APT repository, walks through first-time setup, secures it with tokens and TLS, and connects it to Telegraf (data collection) and Grafana (dashboards).

Table of Contents

1. Which InfluxDB Version Should You Install?
2. Architecture of This Lab
3. Prerequisites
4. Step 1: Update the System
5. Step 2: Add the InfluxData APT Repository
6. Step 3: Install and Start InfluxDB
7. Step 4: Initial Setup (User, Organization, Bucket, Token)
8. Step 5: Write and Query Your First Data
9. Step 6: Firewall Configuration
10. Step 7: Enable HTTPS
11. Step 8: Connect Telegraf and Grafana
12. Alternative: InfluxDB 3 Core
13. Backup and Restore
14. Useful File Locations and Commands
15. Troubleshooting
16. FAQ
17. Conclusion

---

1. Which InfluxDB Version Should You Install?

InfluxDB currently exists in two generations, and they are different products with different query languages and APIs. Pick the one that matches your goal before installing anything.

Aspect	InfluxDB 2.x (OSS)	InfluxDB 3 Core
Status	Mature, widely deployed	Newer engine, evolving quickly
Storage engine	TSM engine	Apache Arrow / Parquet based, object-store backed
Query languages	Flux and InfluxQL	SQL and InfluxQL
Data model	Organizations, buckets	Databases, tables
Default port	8086	8181
Packaging	APT package influxdb2	Install script, tarball or Docker image
Web UI	Built-in	Not built in (use separate tooling)
Ecosystem maturity	Extensive tutorials and integrations	Growing
Choose it when	You want a proven, well-documented setup with a UI	You prefer SQL and the newer architecture

This guide’s main path installs InfluxDB 2.x through APT because it is the most established route on Ubuntu. Section 12 covers InfluxDB 3 Core as an alternative.

2. Architecture of This Lab

  ┌───────────────┐   metrics    ┌──────────────────────────────────────────┐
  │ Servers, IoT, │ ───────────► │  Ubuntu 26.04 LTS                        │
  │ apps          │  (Telegraf)  │                                          │
  └───────────────┘              │   ┌──────────────┐    ┌───────────────┐  │
                                 │   │   Telegraf   │──► │   InfluxDB 2  │  │
  ┌───────────────┐              │   │  (collector) │    │   :8086       │  │
  │ Browser       │ ── HTTPS ──► │   └──────────────┘    │  bucket:      │  │
  │ (Influx UI)   │              │                       │  server_stats │  │
  └───────────────┘              │   ┌──────────────┐    └───────┬───────┘  │
                                 │   │   Grafana    │ ◄──────────┘          │
  ┌───────────────┐              │   │   :3000      │   Flux / InfluxQL      │
  │ Dashboards    │ ◄─────────── │   └──────────────┘                       │
  └───────────────┘              └──────────────────────────────────────────┘

3. Prerequisites

Requirement	Recommendation
OS	Ubuntu 26.04 LTS (fresh or existing)
CPU / RAM	2 vCPU and 4 GB RAM minimum for a lab; more RAM as series count grows
Disk	SSD strongly recommended; plan for retention and growth
Access	A user with sudo privileges
Network	Outbound HTTPS to repos.influxdata.com

Ubuntu 26.04 is the current LTS release. If a package or repository behaves unexpectedly on a brand new release, check the troubleshooting table in Section 15 first.

4. Step 1: Update the System

sudo apt update && sudo apt upgrade -y
sudo apt install -y curl gnupg ca-certificates

Set the correct timezone, because time series data depends on accurate clocks:

timedatectl status
sudo timedatectl set-timezone Asia/Jakarta     # adjust to your region
sudo timedatectl set-ntp true

5. Step 2: Add the InfluxData APT Repository

Download the signing key, verify its fingerprint, and store it in the keyring. The one-liner only writes the key if the fingerprint matches.

curl --silent --location -O https://repos.influxdata.com/influxdata-archive.key

gpg --show-keys --with-fingerprint --with-colons ./influxdata-archive.key 2>&1 \
  | grep -q '^fpr:\+24C975CBA61A024EE1B631787C3D57159FC2F927:$' \
  && cat influxdata-archive.key \
  | gpg --dearmor \
  | sudo tee /etc/apt/keyrings/influxdata-archive.gpg > /dev/null

Add the repository:

echo 'deb [signed-by=/etc/apt/keyrings/influxdata-archive.gpg] https://repos.influxdata.com/debian stable main' \
  | sudo tee /etc/apt/sources.list.d/influxdata.list

sudo apt update

Confirm the package is visible:

apt-cache policy influxdb2

If the fingerprint check fails, the key file was not written. Do not continue. Re-download the key and compare the fingerprint against InfluxData’s documentation, because the signing key has been rotated in the past.

6. Step 3: Install and Start InfluxDB

sudo apt install -y influxdb2 influxdb2-cli

• influxdb2 is the server (influxd).
• influxdb2-cli provides the influx command-line client.

Enable and start the service:

sudo systemctl enable --now influxdb
sudo systemctl status influxdb --no-pager

Check the health endpoint:

curl -s http://localhost:8086/health

Expected output (abridged):

{"name":"influxdb","message":"ready for queries and writes","status":"pass", ...}

7. Step 4: Initial Setup (User, Organization, Bucket, Token)

InfluxDB 2 organizes data as organization → bucket → measurement. A bucket is a named store with a retention policy, and an API token controls access. First-run setup creates the initial admin user, organization, bucket and operator token.

Option A: CLI setup (recommended for servers)

influx setup \
  --username admin \
  --password 'ChangeMe_UseALongPassword!' \
  --org bckinfo \
  --bucket server_stats \
  --retention 30d \
  --force

The command prints the details and stores an active configuration profile for the influx CLI. The token is saved to ~/.influxdbv2/configs. Confirm it:

influx config list

Option B: Web UI

Open http://<server-ip>:8086 in a browser, click Get Started, and fill in the same values.

Create a least-privilege token

Avoid using the operator token for applications. Create a token scoped to a single bucket:

influx bucket list                                   # note the bucket ID
influx auth create \
  --org bckinfo \
  --description "telegraf-write" \
  --write-bucket <BUCKET_ID>

Store each token in a secrets manager or a root-only file. It cannot be recovered in full after you leave the screen in some workflows, so save it immediately.

8. Step 5: Write and Query Your First Data

InfluxDB ingests line protocol: measurement,tag=value field=value timestamp.

influx write \
  --bucket server_stats \
  --org bckinfo \
  --precision s \
  'cpu,host=web01,region=jkt usage=42.5'

Write a few more points, then query them with Flux:

influx query --org bckinfo '
from(bucket: "server_stats")
  |> range(start: -1h)
  |> filter(fn: (r) => r._measurement == "cpu")
'

Or open Data Explorer in the web UI to build the same query visually.

Concept	Meaning	Example
Measurement	Like a table name	cpu
Tag	Indexed metadata, string values	host=web01
Field	The actual measured values, not indexed	usage=42.5
Timestamp	When the reading happened	nanoseconds since epoch

Design tip: put values you filter or group by (host, region) in tags, and the numbers you measure in fields. Very high-cardinality tags such as unique request IDs can hurt performance.

9. Step 6: Firewall Configuration

If you only use InfluxDB locally, no firewall change is needed. To accept remote writes and UI access, allow port 8086 from trusted addresses only.

sudo ufw allow OpenSSH
sudo ufw allow from 10.10.10.0/24 to any port 8086 proto tcp
sudo ufw enable
sudo ufw status numbered

By default influxd listens on all interfaces at :8086. To restrict it to localhost (for example, when a reverse proxy sits in front), set the bind address in the config file in the next step.

10. Step 7: Enable HTTPS

Never send tokens over plain HTTP across a network. For a lab you can generate a self-signed certificate; for production use a certificate from your CA or Let’s Encrypt.

sudo mkdir -p /etc/influxdb/tls
sudo openssl req -x509 -nodes -newkey rsa:4096 -days 365 \
  -keyout /etc/influxdb/tls/influxdb.key \
  -out /etc/influxdb/tls/influxdb.crt \
  -subj "/CN=influx.example.local"

sudo chown influxdb:influxdb /etc/influxdb/tls/influxdb.*
sudo chmod 600 /etc/influxdb/tls/influxdb.key

Create or edit /etc/influxdb/config.toml:

http-bind-address = ":8086"
tls-cert = "/etc/influxdb/tls/influxdb.crt"
tls-key  = "/etc/influxdb/tls/influxdb.key"

Restart and test:

sudo systemctl restart influxdb
curl -k https://localhost:8086/health

Update the CLI profile so it uses HTTPS (add --skip-verify only for self-signed lab certificates):

influx config set --active --host-url https://localhost:8086

11. Step 8: Connect Telegraf and Grafana

Telegraf (collect system metrics)

Telegraf comes from the same repository:

sudo apt install -y telegraf

Edit /etc/telegraf/telegraf.conf (or a file in /etc/telegraf/telegraf.d/):

[[outputs.influxdb_v2]]
  urls = ["https://localhost:8086"]
  token = "$INFLUX_TOKEN"
  organization = "bckinfo"
  bucket = "server_stats"
  insecure_skip_verify = true      # lab only, remove when using a trusted certificate

[[inputs.cpu]]
  percpu = true
  totalcpu = true
[[inputs.mem]]
[[inputs.disk]]
  ignore_fs = ["tmpfs", "devtmpfs", "overlay"]
[[inputs.net]]

Provide the token through an environment file rather than hard-coding it:

echo 'INFLUX_TOKEN=<your-write-token>' | sudo tee /etc/default/telegraf
sudo chmod 600 /etc/default/telegraf
sudo systemctl enable --now telegraf
sudo journalctl -u telegraf -n 20 --no-pager

Grafana (dashboards)

Install Grafana from its official repository (see Grafana’s documentation for the current repository instructions), then add a data source:

Setting	Value
Type	InfluxDB
Query language	Flux
URL	https://localhost:8086
Organization	bckinfo
Token	A read-only token for the bucket
Default bucket	server_stats

Create a read-only token for Grafana:

influx auth create --org bckinfo --description "grafana-read" --read-bucket <BUCKET_ID>

12. Alternative: InfluxDB 3 Core

If you prefer SQL and the newer engine, InfluxDB 3 Core runs as a single binary. The project is evolving fast, so treat the commands below as a starting point and confirm them against InfluxData’s current documentation.

Docker method (requires Docker; see Docker on Ubuntu 26.04 (verify slug)):

docker run -d --name influxdb3 --restart unless-stopped \
  -p 8181:8181 \
  -v influxdb3_data:/var/lib/influxdb3 \
  influxdb:3-core \
  influxdb3 serve \
    --node-id node0 \
    --object-store file \
    --data-dir /var/lib/influxdb3

Script method:

curl -O https://www.influxdata.com/d/install_influxdb3.sh
less install_influxdb3.sh        # always review a script before running it
sh install_influxdb3.sh

Typical first steps afterwards:

influxdb3 create token --admin                        # save the printed token
influxdb3 write --database sensors --token "$TOKEN" 'home,room=kitchen temp=22.1'
influxdb3 query --database sensors --token "$TOKEN" "SELECT * FROM home"

Note that InfluxDB 3 Core listens on port 8181 by default, has no built-in web UI, and uses different tokens, databases and query semantics from InfluxDB 2. Do not mix the two sets of tooling.

13. Backup and Restore

InfluxDB 2 includes a backup command that works against a running server.

sudo mkdir -p /backup/influxdb
influx backup /backup/influxdb/$(date +%F) --token "$INFLUX_OPERATOR_TOKEN"

Restore into a fresh bucket or instance:

influx restore /backup/influxdb/2026-09-21 --token "$INFLUX_OPERATOR_TOKEN"

Schedule a nightly backup with cron and copy it off the host:

sudo tee /etc/cron.d/influxdb-backup > /dev/null <<'EOF'
0 2 * * * root influx backup /backup/influxdb/$(date +\%F) --token "$(cat /root/.influx_token)" && find /backup/influxdb -mtime +7 -type d -exec rm -rf {} +
EOF

Also set sensible retention periods on buckets. Data that is never deleted is the most common reason a time series database fills its disk.

influx bucket update --id <BUCKET_ID> --retention 90d

14. Useful File Locations and Commands

Item	Location or command
Config file	/etc/influxdb/config.toml
Data directory	/var/lib/influxdb/
CLI profiles	~/.influxdbv2/configs
Service logs	sudo journalctl -u influxdb -f
Service control	`sudo systemctl status
List buckets	influx bucket list
List tokens	influx auth list
Server version	influxd version
Health check	curl -s http://localhost:8086/health

15. Troubleshooting

Symptom	Likely cause	Fix
E: Unable to locate package influxdb2	Repository not added or apt update not run	Recheck /etc/apt/sources.list.d/influxdata.list, then sudo apt update
NO_PUBKEY or signature error on apt update	Key missing, wrong path or outdated	Re-download influxdata-archive.key, verify the fingerprint, rewrite /etc/apt/keyrings/influxdata-archive.gpg
Installed an old influxdb package (1.x)	The distribution package name differs from influxdb2	Remove it and install influxdb2 from the InfluxData repository
Repository has no packages for the release	Vendor repo lagging on a brand new Ubuntu release	Confirm the stable main line is used (not the codename), or install the .deb from InfluxData’s download page, or use Docker
influx setup says the instance is already set up	Initial setup was already completed in the UI	Use existing credentials, or create a new user with influx user create
curl: (7) Failed to connect to localhost:8086	Service not running or crashed	sudo systemctl status influxdb and journalctl -u influxdb -n 50
Service fails after editing config.toml	Syntax error or unreadable certificate	Validate the TOML; check ownership influxdb:influxdb and mode on cert and key
401 unauthorized	Missing, wrong or insufficient token	Check influx auth list; use a token with the required read or write permission
x509: certificate signed by unknown authority	Self-signed certificate	Use --skip-verify for labs only, or install a trusted certificate
Remote clients cannot connect	Firewall or bind address	Allow port 8086 in ufw from the client subnet; confirm the server is not bound to localhost only
Disk usage keeps growing	No retention on the bucket	Set --retention on the bucket and confirm old shards are being dropped
Slow queries and high memory	High-cardinality tags	Move unique values into fields; review the tag design

16. FAQ

Is InfluxDB free?
InfluxDB 2 OSS and InfluxDB 3 Core are open source. InfluxData also offers commercial editions and a managed cloud with additional features such as clustering and enterprise support.

Should I use InfluxDB 2 or InfluxDB 3?
Choose InfluxDB 2 if you want the most mature ecosystem, a built-in UI and Flux. Choose InfluxDB 3 Core if you prefer SQL and are comfortable with a younger, fast-changing product.

How is InfluxDB different from Prometheus?
Prometheus pulls metrics and is designed mainly for monitoring and alerting. InfluxDB accepts pushed data, supports richer general-purpose time series storage, and suits IoT and event data as well as infrastructure metrics.

Can I run InfluxDB in Docker instead?
Yes. The official images exist for both InfluxDB 2 and 3. APT is simpler for a single dedicated server, while Docker fits containerized stacks.

How much RAM does InfluxDB need?
It depends mostly on the number of unique series (cardinality) and query load. Start with 4 GB for a lab and monitor memory as you add data.

How do I upgrade InfluxDB?
Run sudo apt update && sudo apt install --only-upgrade influxdb2 influxdb2-cli, and take a backup first.

17. Conclusion

You now have InfluxDB 2 running on Ubuntu 26.04 with a secured initial setup, scoped API tokens, HTTPS, Telegraf collecting metrics, Grafana ready for dashboards and a backup routine. From here, the highest-value next steps are defining retention policies per bucket, designing tags carefully to keep cardinality under control, and adding alerting.