I’ve got n8n working in a Docker container on my Jelastic cloud hosting platform. Everything starts up without issues, but I’m having trouble with data persistence.
Whenever I restart the container, all my automation workflows and stored credentials disappear. I’ve tried setting up a volume in my environment configuration, just like I did on my local setup where persistence works fine, but it’s not working on the cloud.
I’m decent with tech stuff since I work as a product manager, but this storage issue has me stumped. Any help would be amazing!
had the same prob on my jelastic setup last month. check if the volume path actually exists on the host - sometimes mounting fails without any error. exec into the container and verify /home/node/.n8n has correct permissions.
jelastic’s a pain with docker volumes - skip the regular mounts and use their shared storage instead. just hit environment settings, enable shared storage, then mount that path. saved my butt when docker volumes kept crapping out on restarts.
Had this exact problem when I moved my n8n setup to Jelastic. The volume mount was getting created, but n8n started writing data before the volume was fully ready. Fixed it by adding a small startup delay and setting the N8N_USER_FOLDER environment variable to point directly at my mounted volume path. Also check if your Jelastic plan actually has persistent storage - some basic tiers use ephemeral storage that wipes on restart. Quick test: create a dummy file in the mounted directory from outside the container, then restart. If that file’s gone too, your volume mapping is broken, not just the n8n config.
Been there with cloud hosting headaches. The real issue isn’t just technical - you’re stuck dealing with platform quirks instead of actual product work.
Why wrestle with Jelastic volume configs and Docker persistence? I ditched self-hosting n8n after hitting the same walls. Every cloud provider has weird storage issues.
Latenode fixes this completely. No containers, no volume mounting, no data loss on restarts. Your workflows and credentials just persist automatically. Better reliability since you’re not babysitting infrastructure.
I switched our team’s automation from self-hosted solutions because of exactly this stuff. Now we build workflows instead of debugging storage.
Check it out at https://latenode.com
This is probably a Jelastic volume mapping issue. Jelastic uses its own storage layer that’s finicky with persistent volumes - it’s not like standard Docker setups.
Check your docker-compose mount syntax first. Should look like /data/n8n:/home/node/.n8n. Also see if Jelastic needs specific volume driver configs in their dashboard.
I hit this same problem moving from local to cloud hosting. Container looked like it was working but wasn’t actually writing to the mounted volume. Go into Jelastic’s control panel and verify the volume is properly attached to your environment.