Account Reset l Every 2 days l Business Plan

[Arina]

Hi Community,

On 7/18 I joined a monthly subscription on the Business Plan - using n8n version 1.102.3. We are working on a proof of concept and the account has been ‘reset’ (essentially everything wiped: my account information, workflows, everything) 3-4. The behavior started a few days before, 3-4 days before the trial ended and billing took effect.

The software is deployed on our Azure and we did not notice any weird behavior on the server.
It is so unfortunate, sometimes we are logged out while working and we lose work.

I wrote to n8n support 3x and so far no reply. They also don’t see to identify my username. Which in a way makes sense, since my account gets reset…

Really unsure what to do
Thanks for advice,
Arina

Describe the problem/error/question

What is the error message (if any)?

Please share your workflow

(Select the nodes on your canvas and use the keyboard shortcuts CMD+C/CTRL+C and CMD+V/CTRL+V to copy and paste the workflow.)

Share the output returned by the last node

Information on your n8n setup

• n8n version:
• Database (default: SQLite):
• n8n EXECUTIONS_PROCESS setting (default: own, main):
• Running n8n via (Docker, npm, n8n cloud, desktop app):
• Operating system:

[liam]

Hi. I escalated this to the support team.

They are swamped right now so unfortunately response times are delayed

In the meantime can you share some details on your setup? It’s unusual for licencing to be the cause of stuff to get wiped and is more often a configuration mistake.

Please share as much as you safely can (config files/.env files/etc), the support team may also ask for that.

[Arina]

Thanks Ilam, right now the software was deployed under an Azure blob container. We are doing a few POCs to see if we want to go for the enterprise package. Let me get more details from my Infra/IT team, not sure how much I can share though.

[King_Samuel_David]

@Arina I assume the main issues is persistant storage, ie if ure using sqlite and not saving the instance data and it crashed and it redeloys likely lose the data ie all workflows etcetc, if using postgres just make sure it’s saving and keeping the same enc key otherwise u wont see the data if the instance redeloys.

Have you enabled debug logs? or got any logs available, ask the DBA team if they can check around the times ure seeing or provide them and can take a further look.

Also happy to jump on a live call with you as it seems quite urgent getting ure POC ready.

Let me know I have years of experience in technical support over 10+ years. am free atm for the interium period while am doing afew bit so my time maybe limited here soon but am quite free now

Samuel

[Arina]

Hello Samuel,

Many thanks for the reply!
We looked at the logs and noticed this fatal failure:
FATAL ERROR: Reached heap limit Allocation failed - JavaScript heap out of memory

So indeed we increased the storage and from yesterday night until now it seemed to be stable.
What is your expertise? We can connect on Linkedin.
Thanks again!

[King_Samuel_David]

@Arina Thats it, a OOM, I would recommend if not doing regular backups too, or instance and data of n8n using export etc nighly maybe.

I would recommedning getting the build into a pipeline cicd style if not done already too.

Feel free to check my linkedin https://www.linkedin.com/in/samuelskidmore/

Worked more recently in the finance / enterprise architecture side doing tech support

Well am glad its resolved. Also I assume your use proper env, I did setup a grafana montiroing for n8n, which u can tag alerts too N8n + Grafana Full Node.js Metrics Dashboard (JSON Example Included!)

Btw did you get the workflows back after? it’s likely the encryption key in the n8n instance folder, just make sure to use the same otherwise n8n can’t read the db etc.

Hope all goes well and if you need more help digging into that OOM, happy to help, but am sure you identifed it now.

Samuel

[Arina]

Thank Samuel!