Httptemp/cluster_shared not cleaned up

Seafile pro 9.0.16 on CentOS.
I just found out that the folder httptemp/cluster_shared holds 4.3 TB of data. The oldest file is from 9.2022. The httptemp folder itself is clean, but it seems, that the cluster_shared folder is not being cleaned up. The cleanup default should be 3 days.


Hello, this issue is due to the fact that the files in the cluster_shared directory are cleaned up according to the records in the WebUploadTempFiles table in the database.
This directory saves temporary files for resuming uploads, and you can manually delete files in this directory that exceed a certain period of time.
In the future, we will optimize the logic of cleaning up cluster_shared to solve this problem.

1 Like

Thanks for the information.