Help Portal

Octeth Installation ​

On the server, create the Octeth directory and unzip the Octeth package:

bash
mkdir -p /opt/octeth
unzip /opt/oempro-rel-v5.9.3.zip -d /opt/octeth

Make installation tool executable:

bash
chmod +x /opt/octeth/cli/octeth.sh

Run installation tool:

bash
/opt/octeth/cli/octeth.sh install:start

Installing from a script, a CI job or an AI agent

install:start and install:reset can run with no prompts at all using --yes and the accompanying value flags. See Unattended installation in the CLI tool reference.

The installer will check server requirements and then ask you a few questions:

  • Application URL: http://203.0.113.10/
  • Full Name: John Doe
  • Email Address: admin@example.com
  • Username: admin
  • Password: YourSecurePassword123
  • EULA Agree: Y
  • License Key: XXXX-XXXX-XXXX-XXXX-XXXX-XXXX-XXXX-XXXX-XXXX (can be left blank and added later in .oempro_env)

Once you confirm, installation utility will start installing Octeth on your server.

The installation process will take several minutes. Please be patient.

Once it's completed, you will see a success message like:

text
╔════════════════════════════════════════════════════════════════════════════╗
║                                                                            ║
║  ✓ Installation Successfully Completed!                                    ║
║                                                                            ║
║  🔐 Admin Login Details:                                                   ║
║  ────────────────────────────────────────────────────────────────────────  ║
║  URL:      http://203.0.113.10/app/admin/                                  ║
║  Username: admin                                                           ║
║  Password: YourSecurePassword123                                           ║
║                                                                            ║
║  📚 Helpful Resources:                                                     ║
║  ────────────────────────────────────────────────────────────────────────  ║
║  • Octeth Website:      https://octeth.com/                                ║
║  • Octeth Client Area:  https://my.octeth.com/                             ║
║  • Octeth Help Portal:  https://help.octeth.com/                           ║
║  • Octeth Dev Portal:   https://dev.octeth.com/                            ║
║                                                                            ║
║  🚀 Getting Started:                                                       ║
║  ────────────────────────────────────────────────────────────────────────  ║
║  Visit https://octeth.com/getting-started/ to learn more about:            ║
║  • Configuring your Octeth installation                                    ║
║  • Setting up email delivery                                               ║
║  • Creating your first campaign                                            ║
║  • Best practices and optimization tips                                    ║
║                                                                            ║
║  🔒 Security Reminder:                                                     ║
║  ────────────────────────────────────────────────────────────────────────  ║
║  • Change the default admin password after first login                     ║
║  • Enable SSL certificates for enhanced security                           ║
║                                                                            ║
║  💪 Thank you for using Oempro!                                            ║
║                                                                            ║
╚════════════════════════════════════════════════════════════════════════════╝

Open your web browser and visit admin login URL:

http://203.0.113.10/app/admin/

Enter your admin username and password. They are shown at the success message as above.

In the meantime, you can check the health status of the Octeth system by running this command:

bash
/opt/octeth/cli/octeth.sh health:check

You should see a success screen like:

text
ℹ Using API endpoint: http://203.0.113.10/api/v1/system-health-check
ℹ Using API key: sk_xxxxxxxx... (truncated for security)

ℹ Checking system health...

✓ Health check completed successfully (HTTP 200)

System Health Status:

┌─────────────────────┬────────────────────────────────────┐
│ Service             │ Status                             │
├─────────────────────┼────────────────────────────────────┤
│ MySQL               │ ✓ OK                               │
│ Redis               │ ✓ OK                               │
│ RabbitMQ            │ ✓ OK                               │
│ ElasticSearch       │ ✓ OK                               │
│ ClickHouse          │ ✓ OK                               │
│ Session             │ ✓ OK                               │
│ System Container    │ ✓ OK                               │
│ Vector              │ ✓ OK                               │
│ HAProxy             │ ✓ OK                               │
│ Cron                │ ✓ OK                               │
│ Supervisor          │ ✓ OK                               │
│ Send Engine         │ ✓ OK                               │
│ Admin Frontend      │ ✓ OK                               │
│ User Frontend       │ ✓ OK                               │
│ File Permissions    │ ✓ OK                               │
└─────────────────────┴────────────────────────────────────┘

Raw JSON response (for debugging):
{"Checks":{"MySQL":"OK","ClickHouse":"OK","Elasticsearch":"OK","RabbitMQ":"OK","Redis":"OK","Session":"OK","SystemContainer":"OK","Vector":"OK","Haproxy":"OK","Cron":"OK","Supervisor":"OK","SendEngine":"OK","AdminFrontend":"OK","UserFrontend":"OK","FilePermissions":"OK"}}

ℹ Health check completed

Container resource limits ​

Octeth runs as 14 Docker containers. Each one can be given a CPU and a memory cap through .oempro_env:

bash
<SERVICE>_CPU_LIMIT        # CPU cores, e.g. 1.5 - a ceiling, not a pinning
<SERVICE>_MEM_LIMIT        # hard memory cap - exceeding it OOM-kills the container
<SERVICE>_MEM_RESERVATION  # soft reclaim hint under host memory pressure

0 means unlimited, and it is the default for 12 of the 14 services. The full key list and per-service sizing guidance is in the Container Resource Limits entry of Octeth configuration.

What the installer does for you ​

Two services — the send engine (SENDENGINE_*) and the link proxy (LINK_PROXY_*) — ship with a real CPU limit of 3.0 rather than unlimited, because that was their historical hardcoded value.

install:start sizes those two down to your server automatically. On a fresh install it sets them to (CPU cores − 1), with a floor of 1.0, whenever that is lower than the shipped value:

Server vCPUsSENDENGINE_CPU_LIMIT / LINK_PROXY_CPU_LIMIT
11.0
2 (minimum spec)1.0
32.0
4 (recommended spec) or more3.0 (unchanged)

One core is deliberately left for the other containers — MySQL is by far the heaviest consumer in the stack.

A CPU limit above your core count stops the container from starting

Docker validates the CPU limit against the number of cores the host really has and refuses to start the container if it is exceeded:

Error response from daemon: range of CPUs is from 0.01 to 2.00,
as there are only 2 CPUs available

If you edit <SERVICE>_CPU_LIMIT by hand, or you move an existing .oempro_env to a smaller server, keep every CPU limit at or below the server's core count.

Memory has no such constraint: a <SERVICE>_MEM_LIMIT larger than the server's RAM is accepted and simply never takes effect, and <SERVICE>_MEM_RESERVATION is only a hint to the kernel — it never reserves memory and never prevents an out-of-memory kill.

Applying a change ​

Resource limits take effect when a container is created, so after editing .oempro_env:

bash
/opt/octeth/cli/octeth.sh docker:up

Confirm what actually got applied:

bash
docker inspect oempro_sendengine \
  --format 'Memory={{.HostConfig.Memory}} NanoCpus={{.HostConfig.NanoCpus}}'

Running docker compose by hand

The Octeth CLI passes --env-file .oempro_env for you. There is no root .env file, so if you call docker compose directly you must add --env-file .oempro_env yourself — otherwise your limits are ignored and the compose file's built-in defaults are used.

Upgrading an existing installation ​

./cli/octeth.sh upgrade adds any resource-limit keys that your .oempro_env does not yet have, using the shipped values (3.0 CPU for the send engine and link proxy, 0 for everything else). It does not re-run the installer's host sizing, and it never changes a key you have already set. On a server with fewer than 4 cores, review SENDENGINE_CPU_LIMIT and LINK_PROXY_CPU_LIMIT after your first upgrade.

Sizing the send engine on a dedicated MTA server ​

./cli/sendengine-deploy-to-server.sh copies the app server's .oempro_env to the MTA server, so SENDENGINE_CPU_LIMIT, SENDENGINE_MEM_LIMIT and SENDENGINE_MEM_RESERVATION arrive sized for the app server — usually too conservative for a box that does nothing but send. Review them on the MTA server before the first start:

bash
ssh root@<MTA_SERVER_IP>
cd /opt/oempro
vi .oempro_env

Then bring the containers up, always passing --env-file:

bash
docker compose --env-file .oempro_env -f docker-compose.mta.yml build
docker compose --env-file .oempro_env -f docker-compose.mta.yml up -d --scale oempro_sendengine=2

Without --env-file .oempro_env your limits are ignored and the defaults inside docker-compose.mta.yml (3.0 CPU / 6G) are used instead.

The limits are per container

docker-compose.mta.yml runs 2 send-engine replicas by default, and --scale raises that. The server must be able to satisfy SENDENGINE_CPU_LIMIT × replicas — with the default 3.0 and 2 replicas that is 6 cores. Exceeding the core count on any single container makes that container refuse to start.

docker-compose.mta.yml is used on its own (-f docker-compose.mta.yml). It is not an overlay for docker-compose.yml, and the two cannot be combined.

Any questions? Contact us.

©Copyright 50SAAS LLC. All rights reserved.