mirrors/n8n-docs
2
0
Fork 0
mirror of https://github.com/n8n-io/n8n-docs.git synced 2025-11-20 17:48:34 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Initial writeup

Browse Source
This commit is contained in:
Iván Ovejero 2023-11-16 12:58:16 +01:00
parent 894aa61025
commit aadde65e47
1 changed files with 40 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

40
docs/hosting/scaling/queue-mode.md
View File

@ -171,3 +171,43 @@ You can define the number of jobs a worker can run in parallel by using the `con
```bash
n8n worker --concurrency=5
```
## Multi-main setup
/// info | Feature availability
* Available on Self-hosted Enterprise plans
* If you want access to this feature on Cloud Enterprise, [contact n8n](https://n8n-community.typeform.com/to/y9X2YuGa){:target=_blank .external-link}.
///
Optionally, queue mode also supports running multiple `main` processes for high availability.
In a single-mode setup, the `main` process is responsible for:
- running the API,
- serving the UI,
- listening for webhooks,
- handling manual executions,
- running triggers and pollers,
- handle specific licensing tasks,
- pruning executions and binary data, etc.
In a multi-main setup, n8n runs the API, serve the UI, listen for webhooks, and handle manual executions on all `main` processes. n8n also automatically designates a leader `main` process, which is uniquely entitled to run triggers and pollers, handle specific licensing tasks, prune executions and binary data, etc. All non-leader `main` processes are known as followers.
The leader `main` process regularly reports to Redis, setting a quickly expiring key. If the leader ever fails to set the key, e.g. because it crashed or because its event loop has become overly busy, one of `follower` main processes will be designated as the new leader to take over the former leader's responsibilities. If the former leader later becomes responsive again, it will become a follower.
### Configuring multi-main setup
To deploy n8n in multi-main setup, ensure you have a valid license and are running in queue mode.
Then, for each main process, set the environment variable `N8N_MULTI_MAIN_SETUP_ENABLED` to `true` and set `N8N_PORT` to a different port.
If needed, you can adjust the leader check options:
| Via configuration file | Via environment variables | Description |
| ------ | ------ | ----- |
| `multiMainSetup.multiMainSetup:10` | `N8N_MULTI_MAIN_SETUP_KEY_TTL=10` | Time to live (in seconds) for leader key in multi-main setup. |
| `multiMainSetup.interval:3` | `N8N_MULTI_MAIN_SETUP_CHECK_INTERVAL=3` | Interval (in seconds) for leader check in multi-main setup. |
/// note | Keep in mind
We recommend running the multiple main processes behind a load balancer that supports **session persistence**, also known as **sticky sessions**.
///