Unable to delete RBS Blob data from File system even deleting from SharePoint 2010

Consider fallowing scenario that you have SQL Server 2008 R2  RBS enabled, and SharePoint Server 2010 RBS installed servers . You have some files that stored in SharePoint document library whichs streams are stored in RBS and even you deleted this files form SharePoint Document Library you noticed that the Blob data in file system still remaining.

Usually this is not a problem it is by design issue , because purpose of data recovery ,performance consideration, data integrity and safety the deleted files in real are not deleted immediately. So many systems are designed like this as SharePoint and also RBS included. In that kind of systems as a manner of being on the safe side they are just mark the files are deleted and than runs some background process later for deleting files according when some thresholds or limits are exceeded. If you what to find out this issue is a real problem you have to disable or make shut down this functionalities and after doing this still the blob files are remain on file system then you can say that you have a real problem.

On SharePoint side First thing you should check that the feature of Recycle Bin.

Recycle Bins are used to help users protect and recover data.Microsoft SharePoint Server 2010 supports two stages of Recycle Bins: the first-stage Recycle Bin and second-stage Recycle Bin.When a user deletes an item, the item is automatically sent to the first-stage Recycle Bin. By default, when an item is deleted from the first-stage Recycle Bin, the item is sent to the second-stage Recycle Bin. A site collection administrator can restore items from the second-stage Recycle Bin.You turn on and configure Recycle Bins at the Web application level. By default, Recycle Bins are turned on in all the site collections in a Web application. This article describes how to configure Recycle Bin settings for a Web application.”

For more information and usage recommendations about SharePoint Server 2010 Recycle Bins, see Plan to protect content by using recycle bins and versioning (SharePoint Server 2010).

In that Point you have two option to bypass this feature that 1) you can totally disable Recycle Bin from Central Administrations site by CA-> Manage Web Application -> Select web Application which you decided to disable Recycle bin feature -> on Ribbon Menu Select General Settings and set “Recycle Bin” property as “Off”
2) or when you delete a file you can clear Site (First-stage) and Site Collection (Second-Stage) Recycle bins.

On SQL side in Content Database if you want to be sure and confirm deletion of the file you can use fallowing SQL .
1) Open SQL Server Management Studio
2) Select related Content Database and click “New Query”
3) Select * from AllDocs where ListID='<GUID>’
*** You can find List Guid on Browser Address bar when you open Library Settings page of a document libarary.
and check results for the file still is exists in that list .If you clear correct there should not be the related file is present on the results.

Even that you confirm that the file has been deleted from Content Database will still the Blob Data remains in File System where the blobs are stored. Becuase there is another mechanizm in SQL RBS side named “RBS Garbage Collector”

“SharePoint Server 2010 automatically marks unreferenced or deleted BLOB data for removal. SharePoint Server 2010 counts references to BLOBs by looking at the list of BLOB IDs stored by SharePoint Server 2010 in its content databases at the time of removal. Any BLOB references that are present in the RBS store tables but absent in the content database are assumed to be deleted by SharePoint Server 2010 and will be marked for removal. BLOBs that are not present in the content database and were created before the orphan cleanup time window, described later in this article, are also assumed to be deleted by SharePoint Server 2010 and will be marked for removal.

Because SharePoint Server 2010 tabulates BLOB references from the RBS columns of the content database, every RBS column must have a valid index before it can be registered in RBS.

The SQL Server RBS Maintainer tool removes the items marked by SharePoint Server 2010 for removal. You should schedule the clean-up tasks to be run during off-peak hours to reduce the effect on regular database operations.

RBS garbage collection is performed in the following three steps:

  • Reference scan.(RC)  The first step compares the contents of the RBS tables in the SharePoint Server 2010 content database with RBS’s own internal tables and determines which BLOBs are no longer referenced. Any unreferenced BLOBs are marked for deletion.
  • Delete propagation. (DP) The next step determines which BLOBs have been marked for deletion for a period of time longer than the garbage_collection_time_window value and deletes them from the BLOB store.
  • Orphan cleanup. (OC) The final step determines whether any BLOBs are present in the BLOB store but absent in the RBS tables. These orphaned BLOBs are then deleted”


We have talked about ThreshHolds . In RBS configuration we have 3 important threshold for clearing BLOB data.

delete_scan_period :Specifies the minimum amount of time that must pass between two reference scan garbage collection runs. The default value is 30 days
orphan_scan_period: Specifies the minimum amount of time that must pass between two orphan cleanup garbage collection runs. The default value is 30 days
garbage_collection_time_window : Specifies the minimum time that must pass between identifying a blob as having no references in the database and deleting the blob from the store. This guarantees the availability of BLOBs for the specified time in case a backup is restored. The default value is 30 days.

So according to default values , your BLOB files should be cleared after 30 days , if they are not referenced to any Content Database record.
You can get more information about all configuration thresholds about RBS with following article:

For testing immediate delete we can change these threashold .From SQL Server Management Studio:
1)Open SQL Server Management Studio
2)Select RBS enabled Content Database and click “New Query”
3) Execute following queries.
exec mssqlrbs.rbs_sp_set_config_value ‘garbage_collection_time_window’, ‘time 00:00:00’;
exec mssqlrbs.rbs_sp_set_config_value ‘delete_scan_period ‘, ‘time 00:00:00’;
exec mssqlrbs.rbs_sp_set_config_value ‘orphan_scan_period’, ‘time 00:00:00’;

Our job is not done yet:
The actual work of GC is done by the RBS Maintainer application. The maintainer is a console application that takes command line parameters such as the connection string to the database and the phases of GC to execute. This can be run from any machine that has access to the DB and the blob store(s). It can also be run from multiple machines simultaneously. You can schedule it using your favorite scheduler e.g. Windows Task Scheduler.

 Maintainer also takes an optional parameter to limit the amount of time it is run

RBS requires you to define a connection string to each database that uses RBS before you run the RBS Maintainer. This string is stored in a configuration file in the <RBS installation path>\Microsoft SQL Remote Blob Storage 10.50\Maintainer folder that is ordinarily created during installation. The RBS Maintainer can be run manually by executing the Microsoft.Data.SqlRemoteBlobs.Maintainer.exe program together with the parameters that are listed in the following table.
When you run Maintainer from Command Prompt you can trace the operation logs in cmd window:

1) On Sql server open CMD prompt as Administrator and navigate to the path “C:\Program Files\Microsoft SQL Remote Blob Storage 10.50\Maintainer”

2)Execute the command
Maintainer.exe -connectionstringname  RBSMaintainerConnection -operation GarbageCollection ConsistencyCheck ConsistencyCheckForStores -GarbageCollectionPhases rdo
-ConsistencyCheckMode r -TimeLimit 120

You can get more information about Maintainer.exe parameters
for  schedule an RBS Maintainer task please read following arcile:

After you run RBS Maintainer , RS and DP phase completed the blob records will be cleared ! no not yet 🙂 . This operation is takes much 2 or 3 mintues and depends on how much data you have.

Image Source:http://blogs.technet.com/b/pramodbalusu/archive/2011/07/09/rbs-and-sharepoint-2010.aspx


FILESTREAM GC runs as part of the database checkpoint process. This is what causes some confusion – an old FILESTREAM file will not be removed until after it is no longer needed AND a checkpoint runs. 

In Simple recovery mode, you may run following command
In  Full recovery mode, two transaction log with CHECKPOINT are needed

“Forces the FILESTREAM garbage collector to run, deleting any unneeded FILESTREAM files. A FILESTREAM container cannot be removed until all the deleted files within it have been cleaned up by the garbage collector. The FILESTREAM garbage collector runs automatically. However, if you need to remove a container before the garbage collector has run, you can use sp_filestream_force_garbage_collection to run the garbage collector manually

USE <Content Database>;
EXEC sp_filestream_force_garbage_collection @dbname =  N'<Content Database>’;

And finally if still your BLOB data is not cleared than you may create a Case for Microsoft Support 🙂

Two kind of Garbage Collector

when working on client machine you can use server type garbage collector.But it uses more memory.By default in server OS uses Server type garbage collector. You can configure  in your web config like

      <gcServer enabled="true"/>

The CLR has two different GCs: Workstation (mscorwks.dll) and Server (mscorsvr.dll). When running in Workstation mode, latency is more of a concern than space or efficiency. A server with multiple processors and clients connected over a network can afford some latency, but throughput is now a top priority. Rather than shoehorn both of these scenarios into a single GC scheme, Microsoft has included two garbage collectors that are tailored to each situation.

Server GC:

  • Multiprocessor (MP) Scalable, Parallel
  • One GC thread per CPU
  • Program paused during marking

Workstation GC:

  • Minimizes pauses by running concurrently during full collections

The server GC is designed for maximum throughput, and scales with very high performance. Memory fragmentation on servers is a much more severe problem than on workstations, making garbage collection an attractive proposition. In a uniprocessor scenario, both collectors work the same way: workstation mode, without concurrent collection. On an MP machine, the Workstation GC uses the second processor to run the collection concurrently, minimizing delays while diminishing throughput. The Server GC uses multiple heaps and collection threads to maximize throughput and scale better.

Source: MSDN

Object size in memory c#

Unfortunately there is no direct way to do it. Speacialy for the managed code. May you can use memorystream and serialization for getting an idea but this are not actual size of your object. You may use some solution below for calculation actual size.

One way is to use the GC.GetTotalMemory method to measure the amount of memory used before and after creating your object. This won’t be perfect, but as long as you control the rest of the application you may get the information you are interested in.
Source :http://stackoverflow.com/questions/605621/how-to-get-object-size-in-memory

Using Strike or SOS Debugging Extension (SOS.dll)
use : DumpHeap [-stat] [-min <size>][-max <size>] [-thinlock] [-mt <MethodTable address>] [-type <partial type name>][start [end]]

Displays information about the garbage-collected heap and collection statistics about objects.
The DumpHeap command displays a warning if it detects excessive fragmentation in the garbage collector heap.
The -stat option restricts the output to the statistical type summary.
The -min option ignores objects that are less than the size parameter, specified in bytes.
The -max option ignores objects that are larger than the size parameter, specified in bytes.
The -thinlock option reports ThinLocks. For more information, see the SyncBlk command.
The -mt option lists only those objects that correspond to specified the MethodTable structure.
The -type option lists only those objects whose type name is a substring match of the specified string.
The start parameter begins listing from the specified address.
The end parameter stops listing at the specified address.

!dumpheap -stat
Note that !dumpheap only gives you the bytes of the object type itself, and doesn’t include the bytes of any other object types that it might reference

More Info:

3rd party solutions

Using .Net Memory Profiler (Easy way)

using CLR Profiler (free)

ANTS Memory Profiler

Some people have confused the System.Runtime.InteropServices.Marshal.SizeOf() service with this API.  However, Marshal.SizeOf reveals the size of an object after it has been marshaled.  In other words, it yields the size of the object when converted to an unmanaged representation.  These sizes will certainly differ if the CLR’s loader has re-ordered small fields so they can be packed together on a tdAutoLayout type.
source : http://blogs.msdn.com/cbrumme/archive/2003/04/15/51326.aspx

code bye…