I've been going through a ton of videos and this is by far the best and extremely well-explained video. NO BS! Just 'How to do'. Keep up the good work!
This is extremely well explained. I hope you keep making tutorial videos! To make it this simple, yet super-clear and fast (not wasting time) is a true skill that few people have.
Thanks for the tutorial man. I have a couple of follow-up videos suggestions: 1. An in-depth video on how to attach an external database 2. A video on how to keep docker and n8n up to date after installing on Google Cloud VM Thanks again. Subscribed.
Hi, thanks for the insightful tutorial. I have one question regarding the security settings. I noticed that secure cookies are disabled in your configuration. Since the reverse proxy handles SSL/TLS termination for the custom domain, does this ensure that the connection remains fully secure even with secure cookies turned off? Thanks again for the video. Cheers!
@Tejesh-r3g It also depends on image you choose for ec2 but mostly it should work. If you get stuck somewhere, just ask chatgpt to give equivalent commands for your ec2.
Well explained. Thank you very much. I successfully did the tutorial and ran the n8n in my subdomain. Question: can I install ollama and openwebui in the same instance to integrate with the n8n?
@zackiLA I'm glad it was helpful. Yes, of course you can install other apps/models on the same instance. But ollama will require high VM configurations. So, check the requirements for the specific model you want to install before proceeding.
@lobos009 You can follow the steps here: github.com/futurminds/n8n-self-hosting/blob/main/README.md#:~:text=your%2Ddomain.com.-,How%20to%20update%20n8n%3A,-You%20can%20follow
I'm running into issues while trying to install in a oracle machine. Could you explain how to do it on oracle? Don't forget to tell people that they should allow used ports on its oracle firewall. My goal is to use oracle and cloudflare but the setup page never loads
So one of the advantage of Home Assistant, Huggin, Node Red and now n8n is that they're open source.. unlike Zapier , IFTTT and Make Integromat? What about costs associated with APIs and hosting.. is there any low-cost hobby tier? I remember OffCloud and IBM BlueMix was like this, but not anymore. I don't mind a pay-per-use model, but signing up for a monthly commitment wouldn't be worth it in my case. I think eventually, I'd like to setup self-hosting / DuckDNS / Proxmox but don't want to tackle too much at once.. for now, I think Google Cloud VM seems like it's the easiest way to get started learning the APIs (ex. Amadeus, Twilio), and AI tools (Custom GPT, AI Agents, Claude's Personal Projects)?
@bennguyen1313 These open source projects can save a lot of costs for applications that have heavy usage. Self hosting those requires paying for the server and not for the number of workflow executions. You can explore ec2 instances on aws or compute VMs on google cloud and get started with as low as $10 per month. And if it's for learning/experimental purpose the free google compute VM (demonstrated in this video) is good enough.
Great tutorial. Only issue is that you should not include the protocol () in the docker run command for the value of 'N8N_HOST'. I ran into an issue where N8N was appending the protocol to the value when generating a webhook URL on my VM.
@nabi188 yes, you can follow the steps here to upgrade n8n without losing data: github.com/futurminds/n8n-self-hosting?tab=readme-ov-file#how-to-update-n8n
Awesome tutorial! I'm a beginner on that, so I'm just wondering about the external database needs. If I want to connect an external database to store the datas like Supabase Firebase, to sotre users datas, is the 30GB enough just for my workflows inside the VM engine? I'm thinking something like up to 3 small workflows, just for an MVP validation.
@jhonydalcin4988 How many users' data do you plan to store? Basically what's the scale of the system? How many workflows do you expect to run daily? Most of the time, 30 GB is more than enough for MVPs.
great content man! however, I'm facing an issue regarding the webhook data not showing in UI, any idea how to resolve this? Ps- the data is available in the execution, just doesnt show in UI, also cannot stop execution
"the data is available in the execution, just doesnt show in UI, also cannot stop execution" -> I'm not clear about what issue you are seeing. But check if this is the expected behaviour.
I followed these steps and could create successfully, however when I am testing web hook calls, the execution shows the test execution as success however the front end loader keeps going with Waiting for Trigger and the page stays the same. Any help?
great video! However my webhook data doesnt show in UI, how can this be resolved! Ps- the data is available in the execution, I also cannot stop the execution
@IvySaskia you can follow the instructions here to update the version: docs.n8n.io/hosting/installation/docker/#updating Important: Take a backup of ~/.n8n:/home/node/.n8n Ensure that your n8n instance is using a persistent volume or a mapped directory for its data storage. This is crucial because the workflows, user accounts, and configurations are stored in the database file (typically database.sqlite), which should be located in a directory that remains intact even when the container is removed. In your docker-compose.yml, you should have something like this: ``` volumes: - ~/.n8n:/home/node/.n8n ``` This mapping ensures that the .n8n directory on your host machine is used for data storage, preserving your workflows and configurations across container updates. When you stop and remove the n8n container, you are only deleting the container instance itself, not the data stored in the persistent volume. As long as the volume is correctly configured, your workflows and accounts should remain unaffected. But to avoid any chance of data loss you should take a backup of ~/.n8n:/home/node/.n8n before removing the container.
@@FuturMinds ok... when I use on my desktop docker, the data does not lose, but on this deploy how do I make the backup? I restart with the new image (making pull the new iamge) , I got two issues. 1 Data is lose .. 2 the new version on the container is not updated.
@IvySaskia On this link: docs.n8n.io/hosting/installation/docker/#updating, you will find that you first need to stop your container and then remove it. Then update the image and spin up a new container. To create a backup, you can copy ~/.n8n:/home/node/.n8n to your local or another directory on the same VM even before deleting the container. And then, after updating and spinning up a new container, if you see the data getting lost, you can replace ~/.n8n:/home/node/.n8n with the one you saved earlier. This should work fine.
you have been mentioned as a no code guru online..in some vedios ....but my no code experience with this vedio is bad ....not completely working for no-code self host n8n dudes out there.... if u do it again with a free duck dns .....walkthrough vedios you may get better likes...& it will great... & may be the only no-code n8n self host free installation vedio which breaks the internet ....
@Phongyewtong It depends on how complex your workflows are and how many workflows will run in parallel. If it's just a very few workflows running in parallel, I don't see any problem. If you are not looking for high execution speed, you can start with this spec and then upgrade if required.
@hashtechnovation3914 did you mean aws ec2? AWS Lambda is a serverless compute service designed to run code without managing servers, but it's not ideal for self-hosting applications like n8n due to its constraints around long-running tasks, storage, and network access. n8n requires a persistent environment, such as a VM or container, which AWS Lambda is not designed for.
I think it's not working anymore, I followed all the instructions and got stuck at step 2. Even though docker is active, I can't access N8N with the external IP.
@Docente_Digital_MX Did you see any errors while running 'docker run'? You can run 'docker ps' and see if container is in running state. Instructions are correct since they are working for everyone. See if you missed any step in the doc.
@@FuturMinds CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES f9ac7baebbb0 n8nio/n8n "tini -- /docker-ent…" 2 hours ago Up About an hour 0.0.0.0:5678->5678/tcp, :::5678->5678/tcp n8n No, no errors during installation.
@@FuturMinds HEY THANKS FOR REPLY, thanks for tutorial as well , completed my n8n installation but when i suspend my machine all things have gone also assigned with new external ip
Yes since the n8n data is stored on that same VM you terminated, it is lost. If you want your data to be independent of VM, you'll need to configure external DB.
@@FuturMinds hello , i am able to run n8n over google cloud through your video but i got a problem like no one having, any workflow get stuck at executing and when i stopped it got executed successfully, even in chat model i have to stop after every asking question to able to run properly,. Need help please
Just got the n8n first page after so many iterations. Noe only security part is remaining. I am completely new to networking, I missed setting up a firewall port 5678. Additionally, you won't get SSL certificate with barely IP address. You should have a domain name for it. I learnt many things on networking with iteration (from youtube and Chatgpt)
@consig1iere294 You will need a domain to use certbot which is used to enable ssl on your server. I can skip setting up ssl but then you will see warnings in your browser and traffic won't be encrypted and it's risky. Instead, I would suggest you to buy a really cheap (
Huge thanks for helping me get this done. Your channel will be huge with this type of value. Keep going!
Appreciate the kind words! Glad it helped.
I've been going through a ton of videos and this is by far the best and extremely well-explained video. NO BS! Just 'How to do'. Keep up the good work!
@NimeshMadushanka Thanks for watching and I'm glad you found the video useful!
This is extremely well explained. I hope you keep making tutorial videos! To make it this simple, yet super-clear and fast (not wasting time) is a true skill that few people have.
Thanks @pressrepeat2000 This means a lot :)
You are a legend! It worked smoothly
please make a video for installation on oracle free tier, thanks!!!!!!!!
Thanks for the tutorial man. I have a couple of follow-up videos suggestions:
1. An in-depth video on how to attach an external database
2. A video on how to keep docker and n8n up to date after installing on Google Cloud VM
Thanks again. Subscribed.
@diegocs8 Thanks for the video suggestions. Those are good topics. I'll try to cover them soon.
Hello, your guide is outstanding. Thank you for sharing it. Best regards, Oren
Glad it was helpful!
Thank you. It is really helpful! Do you recommend I host Dify and N8N on one server or separately?
Great explanation. New subscriber :)
Awesome! Thanks @torrquemado1980
Great, thanks for sharing! What happens if the server gets rebooted or so, will the work flows be retrieved or lost?
Hi, thanks for the insightful tutorial. I have one question regarding the security settings. I noticed that secure cookies are disabled in your configuration. Since the reverse proxy handles SSL/TLS termination for the custom domain, does this ensure that the connection remains fully secure even with secure cookies turned off? Thanks again for the video. Cheers!
nice.. New subscriber :)
Awesome! Thanks for subscribing.
Is this same for AWS EC2 , for commands
@Tejesh-r3g It also depends on image you choose for ec2 but mostly it should work. If you get stuck somewhere, just ask chatgpt to give equivalent commands for your ec2.
while running the docker its not showing the cointainer id
how to resolve this?
Server is not working I get nothing when I do the sudo docker ps command. IS there a way to clear out the server and start over?
Well explained. Thank you very much. I successfully did the tutorial and ran the n8n in my subdomain.
Question: can I install ollama and openwebui in the same instance to integrate with the n8n?
@zackiLA I'm glad it was helpful. Yes, of course you can install other apps/models on the same instance. But ollama will require high VM configurations. So, check the requirements for the specific model you want to install before proceeding.
Great tutorial. Any tutorial on how to safely update n8n to latest version?
@lobos009 You can follow the steps here: github.com/futurminds/n8n-self-hosting/blob/main/README.md#:~:text=your%2Ddomain.com.-,How%20to%20update%20n8n%3A,-You%20can%20follow
I'm running into issues while trying to install in a oracle machine. Could you explain how to do it on oracle? Don't forget to tell people that they should allow used ports on its oracle firewall. My goal is to use oracle and cloudflare but the setup page never loads
So one of the advantage of Home Assistant, Huggin, Node Red and now n8n is that they're open source.. unlike Zapier , IFTTT and Make Integromat?
What about costs associated with APIs and hosting.. is there any low-cost hobby tier? I remember OffCloud and IBM BlueMix was like this, but not anymore. I don't mind a pay-per-use model, but signing up for a monthly commitment wouldn't be worth it in my case.
I think eventually, I'd like to setup self-hosting / DuckDNS / Proxmox but don't want to tackle too much at once..
for now, I think Google Cloud VM seems like it's the easiest way to get started learning the APIs (ex. Amadeus, Twilio), and AI tools (Custom GPT, AI Agents, Claude's Personal Projects)?
@bennguyen1313 These open source projects can save a lot of costs for applications that have heavy usage. Self hosting those requires paying for the server and not for the number of workflow executions.
You can explore ec2 instances on aws or compute VMs on google cloud and get started with as low as $10 per month.
And if it's for learning/experimental purpose the free google compute VM (demonstrated in this video) is good enough.
Great tutorial. Only issue is that you should not include the protocol () in the docker run command for the value of 'N8N_HOST'. I ran into an issue where N8N was appending the protocol to the value when generating a webhook URL on my VM.
Great video! How can automatically pull the latest N8N version without workflows getting lost? Do you have a video on that?
Thanks @JpresValknut. Yes, this one: th-cam.com/video/x3pYHosEUnA/w-d-xo.html
@@FuturMinds Thanks. You got one for GCP as well?
@JpresValknut Version update process will remain same irrespective of where the server resides. So, you can follow the same steps.
@@FuturMinds I did it in the Google Shell however there was some errors that I encountered that's why I asked
Thank you for your tutorial. But I wonder if I can upgrade n8n later without losing data?
@nabi188 yes, you can follow the steps here to upgrade n8n without losing data: github.com/futurminds/n8n-self-hosting?tab=readme-ov-file#how-to-update-n8n
Awesome tutorial! I'm a beginner on that, so I'm just wondering about the external database needs. If I want to connect an external database to store the datas like Supabase Firebase, to sotre users datas, is the 30GB enough just for my workflows inside the VM engine? I'm thinking something like up to 3 small workflows, just for an MVP validation.
@jhonydalcin4988 How many users' data do you plan to store? Basically what's the scale of the system? How many workflows do you expect to run daily? Most of the time, 30 GB is more than enough for MVPs.
great content man!
however, I'm facing an issue regarding the webhook data not showing in UI, any idea how to resolve this?
Ps- the data is available in the execution, just doesnt show in UI, also cannot stop execution
"the data is available in the execution, just doesnt show in UI, also cannot stop execution" -> I'm not clear about what issue you are seeing. But check if this is the expected behaviour.
Not for windows ? Only lunix ?
I followed these steps and could create successfully, however when I am testing web hook calls, the execution shows the test execution as success however the front end loader keeps going with Waiting for Trigger and the page stays the same. Any help?
great video!
However my webhook data doesnt show in UI, how can this be resolved!
Ps- the data is available in the execution, I also cannot stop the execution
same issue here!
same issue here! Did you manage to fix it?
same
Same
Managed to solve the issue thank you
How install n8n queue mode? For scaling n8n
@rukhanzakaria Are you setting up multiple n8n instances? If yes, you can follow this: docs.n8n.io/hosting/scaling/queue-mode/
After multiple failed attempts, I'm really feeling despondent. Somehow I just don't get n8n to launch from the Google cloud server😢
@asampson628 What issue are you facing? I'll try to help.
Where you able to launch it??
When a new version of n8n is up, how do u update on the deployed one?
@IvySaskia you can follow the instructions here to update the version: docs.n8n.io/hosting/installation/docker/#updating
Important: Take a backup of ~/.n8n:/home/node/.n8n
Ensure that your n8n instance is using a persistent volume or a mapped directory for its data storage. This is crucial because the workflows, user accounts, and configurations are stored in the database file (typically database.sqlite), which should be located in a directory that remains intact even when the container is removed.
In your docker-compose.yml, you should have something like this:
```
volumes:
- ~/.n8n:/home/node/.n8n
```
This mapping ensures that the .n8n directory on your host machine is used for data storage, preserving your workflows and configurations across container updates.
When you stop and remove the n8n container, you are only deleting the container instance itself, not the data stored in the persistent volume. As long as the volume is correctly configured, your workflows and accounts should remain unaffected.
But to avoid any chance of data loss you should take a backup of ~/.n8n:/home/node/.n8n before removing the container.
@@FuturMinds ok... when I use on my desktop docker, the data does not lose, but on this deploy how do I make the backup? I restart with the new image (making pull the new iamge) , I got two issues. 1 Data is lose .. 2 the new version on the container is not updated.
@IvySaskia On this link: docs.n8n.io/hosting/installation/docker/#updating, you will find that you first need to stop your container and then remove it. Then update the image and spin up a new container.
To create a backup, you can copy ~/.n8n:/home/node/.n8n to your local or another directory on the same VM even before deleting the container. And then, after updating and spinning up a new container, if you see the data getting lost, you can replace ~/.n8n:/home/node/.n8n with the one you saved earlier.
This should work fine.
you have been mentioned as a no code guru online..in some vedios ....but my no code experience with this vedio is bad ....not completely working for no-code self host n8n dudes out there.... if u do it again with a free duck dns .....walkthrough vedios you may get better likes...& it will great... & may be the only no-code n8n self host free installation vedio which breaks the internet ....
can n8n run on this spec with a few workflow running 24/7?
@Phongyewtong It depends on how complex your workflows are and how many workflows will run in parallel. If it's just a very few workflows running in parallel, I don't see any problem. If you are not looking for high execution speed, you can start with this spec and then upgrade if required.
Please make video how to self host on aws lambda please
@hashtechnovation3914 did you mean aws ec2?
AWS Lambda is a serverless compute service designed to run code without managing servers, but it's not ideal for self-hosting applications like n8n due to its constraints around long-running tasks, storage, and network access. n8n requires a persistent environment, such as a VM or container, which AWS Lambda is not designed for.
Worked 01/2025
I think it's not working anymore, I followed all the instructions and got stuck at step 2. Even though docker is active, I can't access N8N with the external IP.
@Docente_Digital_MX Did you see any errors while running 'docker run'? You can run 'docker ps' and see if container is in running state. Instructions are correct since they are working for everyone. See if you missed any step in the doc.
@@FuturMinds CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
f9ac7baebbb0 n8nio/n8n "tini -- /docker-ent…" 2 hours ago Up About an hour 0.0.0.0:5678->5678/tcp, :::5678->5678/tcp n8n
No, no errors during installation.
@@FuturMinds Follow all the steps rigorously, but when I try to log in it tells me that it took too long to respond.
I did it, I needed to create a firewall rule
Awesome :)
Is there any way to host n8n on google cloud run
Not possible. n8n requires long running VM with some memory and space.
@@FuturMinds HEY THANKS FOR REPLY, thanks for tutorial as well , completed my n8n installation but when i suspend my machine all things have gone also assigned with new external ip
Yes since the n8n data is stored on that same VM you terminated, it is lost. If you want your data to be independent of VM, you'll need to configure external DB.
@@FuturMinds hello , i am able to run n8n over google cloud through your video but i got a problem like no one having, any workflow get stuck at executing and when i stopped it got executed successfully, even in chat model i have to stop after every asking question to able to run properly,. Need help please
Ok some how i managed to solve it thanks
Bro I could struggled 5 hr but could not suceceded
Just got the n8n first page after so many iterations. Noe only security part is remaining. I am completely new to networking, I missed setting up a firewall port 5678.
Additionally, you won't get SSL certificate with barely IP address. You should have a domain name for it.
I learnt many things on networking with iteration (from youtube and Chatgpt)
I am a beginner and was following through your tutorial great but got stuck at 5:20 Hostinger part. Is there a free way to do that?
@consig1iere294 You will need a domain to use certbot which is used to enable ssl on your server. I can skip setting up ssl but then you will see warnings in your browser and traffic won't be encrypted and it's risky. Instead, I would suggest you to buy a really cheap (