Knowledge Base

Recover a file on research storage

How to recover a file on research storage

Files that have been accidentally deleted or a previous version of them can be recovered from a snapshot. The way a file is recovered depends on the underlying file-system (Windows/Linux) and in the case of Linux, the storage level. However if you want to manage revisions of a file, for example to record code changes, then you should use a repository service such as our gitlab service and not rely on the file system snapshots.

Windows Shares

For files on windows research shares, which are usually mapped from \\\research, you can use previous versions which are also sometimes called shadow copies. This does not apply to shares found under \\\research-nfs. Files in these shares need to be recovered in same way as a Linux Export (see below).

Linux Exports

The way you recover a file under /storage on Linux computers, depends on the level of the storage (gold or silver). The approach is similar in both cases: go to the appropriate snapshot directory and use the copy command “cp” to copy the file you want to recover into the normal storage. The difference between gold and silver storage is the location and the name of the snapshots.


For gold level storage the snapshots are located in the .zfs/snapshot directory at the top of the export (/storage/gold/<export>. For example in the case of the export act-data2

Inside each snapshot is a complete copy of all the files and directories in the export as they were when the snapshot was taken. So if you want to recover a file in a sub-directory  you need to use the change directory command “cd” to navigate to the sub-directory in the snapshot or give the full path-name to the “cp” command.


Because the snapshots appear in the directory structure care needs to be taken when using recursive commands such as e.g. “find” or “du” at the top level of the export.

The current schedule for snapshots on gold storage is

Time Local Remote
Every hour, on the hour Keep 25 snapshots Keep 50 snapshots
04:30 every day Keep 90 snapshots Keep 90 snapshots


On silver storage the snapshots are available under .snapshot in every directory, except in some cases the top of the export. The .snapshot directory does not appear unless explicitly referenced so it will NOT appear in the output of the list command “ls -a” or when using recursive commands such as “find” or “du” unless it has been recently accessed.

Inside each snapshot is a complete copy of all the files and sub-directories in the relevant directory as they were when the snapshot was taken. So to recover a file from a snapshot you can use the copy command “cp”:

The current snapshot schedule for silver storage is

Name Frequency Keep
Hourly Every  hour 24
Daily Every  day 7
Weekly Every week on Sun  4
Monthly Every month on  3

UNIX Home Directories

Files in the UNIX home directories /home/users/<username> can be restored in the same way as silver storage.


RACC updates

We have changed the default partition to be the 'limited'. Users submitting jobs longer than 24h need to explicitly request another partition, e.g. 'cluster'. Jobs submitted with time limit exceeding the partition limit will be rejected.   Purchasing RACC nodes owned by projects can  be requested using the Self Service Portal form,  see /
Read More
Suggest Content…

Related articles

Request Project Support from the Academic and Research Computing Team

Use the Reading Academic Computing Cluster (RACC)

Use the Linux Desktop Service (NX with NoMachine)

Transfer files in and out of the university

Connect from outside of campus

Manage my web presence

Contact IT