I had a large number of executions that were created before Feb 2. Many of them were in a “Waiting” state at that time.
Now, when I check the Executions list, none of the executions from before Feb 2 are showing up. It looks like they have completely disappeared from the history and are not loading at all also not in the n8n api.
Has anyone else faced this issue?
Is there any way to restore or view older executions?
Although you could go down looking for logs in the docker instance if you are self hosted on docker, i think this happened after an update, i do not think so you can restore them by any chance within the UI. I Recommend backing up your workflows according to their versions and just restore them according to the version name.
Most likely: Your execution data has been pruned (deleted) by n8n’s automated retention policy.
Today is Feb 9. Your missing data is from “before Feb 2,” which is exactly 7 days ago (168 hours). The default setting for EXECUTIONS_DATA_MAX_AGE in many configurations is often set to roughly one week (or 14 days, depending on the version/hosting) to prevent the database from crashing due to size.
Why “Waiting” executions disappeared:
While n8n tries to preserve active executions, if the “Waiting” executions were interrupted (e.g., a server restart) and failed to resume properly, they may have been marked as “crashed” or “unknown” by the system. Once in that state, the pruner treats them as old data and deletes them along with the completed ones.
Next steps:
• Check your environment variables (or .env file) for EXECUTIONS_DATA_MAX_AGE. It is likely set to 168 (hours).
• CheckEXECUTIONS_DATA_PRUNE_MAX_COUNT. If you have a high volume of executions, you might have hit the limit (e.g., 10,000), causing n8n to delete the oldest ones regardless of age to make room for new ones.
• Database Recovery: If you are self-hosting and have a backup of your database.sqlite (or Postgres dump) from before today, you can restore it to see the data. Without a backup, pruned data is permanently deleted from the disk to save space.
Quick questions:
• Are you using the default SQLite database? (If so, the file size grows quickly, and n8n prunes aggressively to keep it stable).
• Did you restart your n8n instance recently? (Restarting can sometimes clear “Waiting” nodes if they weren’t saved correctly to the database).
We recently identified and addressed a critical security vulnerability affecting several n8n versions. Because of this, we have taken action to schedule an automatic version update of impacted instances.
We are also preparing advisory information for additional vulnerabilities that will be published next week. The scheduled upgrade will address these as well.
You can also start the upgrade manually from your Cloud dashboard: https://*******.app.n8n.cloud
For full details, please refer to the advisory information on GitHub:
I was reading the doc and found that for the starter plans n8n keeps the executions for only 1 week or upto 2500 executions
that’s why i am not able to get older executions
I am just wondering if N8N also deletes the executions which are in waiting stage
Edited
I read the documentation and found n8n do not delete running or waiting executions on community edition not sure about the cloud edition also my waiting exicutions are not deleted