Disk Cleanup & Mount_Cache

[kmfyhr]

Over the past few months I've noticed a steady decline in free space on the C:\ drive of my Veeam server. I'm running Server 2008R2, Veeam 6.5.0.109, and have a 70 GB C:\ drive.

After some initial investigation a few months ago I attributed the usage to the files found in C:\ProgramData\Veeam\Backup, as this directory had grown to nearly 10 GB in size. I did some reading on here, did some cleanup, and then applied windows compression to this directory. As of today, the reported size of the directory is 8.38 GB, and size on disk is down to 2.00 GB, good results!

Keeping an eye on this, I noticed free space was being reduced at a rate faster than this newly compressed folder was growing, so something else must be to blame. After combing through the C:\drive one level at a time looking for the large location, I made it down to the following location: C:\Windows\SysWOW64\config\systemprofile\AppData\Local

Inside the Local directory I found MANY (deleted before I checked on how many) mount_cache{...} files. Some searching on the forum gave way to the fact that these are remnants of FLR sessions, and SHOULD be deleted automaticly at the close of each FLR session. I had enough of these files that they were using 44 GB (yes GIGAbytes) of my C:\ drive.

What may be unique about my usage of FLR, is that on a nightly basis, I run multiple powershell scripts that, mount the most recent backup of a given VM, and copy some of the data over to a network location, that data being -1 days old is used in a test/dev environment that day, so it keeps our test data always ALMOST real time.

I'm wondering if since I'm using powershell to do these FLR jobs, maybe the auto-delete of the mount_cache files is not happening as expected (as it presumably does when using the GUI for FLR) When I did a quick once-over of the files before deleting them, the number of files created each night, and the time stamps looked to match up perfectly with my scheduled jobs, so they seem to obviously be the source of these files.

An example of my one of my nightly scripts follows: Maybe I'm doing something wrong, or could be doing something better?

asnp VeeamPSSnapIn
$Copy_LAMS_to_ORATEST9 = Get-VBRBackup | where {$_.JobName -eq "orasrvr9"} | Get-VBRRestorePoint | sort CreationTime -Descending | where {$_.VMName -eq "orasrvr9"} | select -First 1 | Start-VBRWindowsFileRestore
$file = "c:\veeamflr\orasrvr9\volume1\oracle\oradata\lams\”
roboCopy $file \\oratest9\f$\oracle\oradata\lams\ /COPYALL /R:1 /W:1
Stop-VBRWindowsFileRestore $Copy_LAMS_to_ORATEST9
timeout 5
schtasks /Run /S oratest9 /TN STARTUP_OPEN_LAMS_FROM_FLASH
timeout 5

[veremin]

Hi, Kenny.

As a first investigation step it might be worth confirming that it is automatic FLR execution that causes this issue. So, you might want to run two different FLR session (one started via GUI, the other via PS) and see whether the files are left by PS or GUI session.

As to the workaround, you can add a few lines to the end of the script that will clear out given cache files once the FLR session is finished.

Thanks.

[foggy]

According to QC, there's a known issue with FLR cache files not being removed automatically after FLR session is finished (will be addressed in one of the next updates).

[D.Jurcka]

Hello Kenny Fyhr,

I seem to have the same problem as you do/did. I have a small question about the situation, what is a safe way to delete the mount_cache files?
Can you safely delete these files via the Windows Explorer or are there ways to do this via the GUI/PS

I hope to hear from you soon.

Yours sincerely
Dwayne Jurcka

[foggy]

Dwayne, you can safely remove them manually via Windows Explorer or any other file managing utility.