Quantcast
Channel: File Services and Storage forum
Viewing all 13580 articles
Browse latest View live

Windows Server 2019 Dedupe problem

$
0
0

eventlogger

error1

The Data Deduplication Service service failed to start due to the following error:
A privilege that the service requires to function properly does not exist in the service account configuration. You may use the Services Microsoft Management Console (MMC) snap-in (services.msc) and the Local Security Settings MMC snap-in (secpol.msc) to view the service configuration and the account configuration.

error2

\The Data Deduplication Service service terminated with the following service-specific error:
%%2153141004

trying to start the service

msg from server manager 




Windows 2019 Deduplication error (2153141004)

$
0
0

After an in-place upgrade from 2016 to 2019, dedup service refuses to start with the following error. I can still access data inside the drives, but dedup doesn't work and I get errors stating the same on the server management console for files and storage.

This is the only related event error:

Event 7024

The Data Deduplication Service service terminated with the following service-specific error: 
%%2153141004

I have the server patched to the latest. Is this a known issue?

Thanks!

ClusterFaultDomain -FaultDomainType StorageNode S2D Windows 2019

$
0
0

Hello.

Is any one known where FaultDomainType =  StorageNode fits ? Because i cano't faind information about that type ? This type is parent of node or chassis ?


Moving DFS Share on new drive with same drive letter

$
0
0

Hello Community!

Right now I'm facing the Tast that I want to move some dfs shares to a new disk on the same server because the old one is getting to small. The plan is to move all the data to the new disk, disable the dfs share service, change the drive letters so that the new disk has the same drive letter as the old disk, restart the server and with it the dfs service.

The question that I have now is, if I have to do anything else regarding the dfs shares when the new drive has the same drive letter as the old one had?

If so, what are the steps necessary?

Hope you guys can help me out here.

Thank you all!

Change ClusterFaultDomain after create s2d cluster windows 2019

$
0
0

Hi.

Is possible to change ClusterFaultDomain(in windows 2019) after create cluster. ?

Because i fount information that it's recommend to create that fault domains (chassis, rack, site etc. ) before creation s2d cluster.  And create s2d cluster creator ask me if i want user that fault domains. But what if i have already operational cluster and i want to change ClusterFaultDomain from node to rack ?


How do I grant access to a clustered file server's admin share?

$
0
0

I've set up a clustered file server and would like to grant access to the clustered file server's admin share. From my understanding, admin share permissions can't be modified. Is there any way to grant access to a cluster admin share without adding the user/group to BUILTIN\Administrators. Methods to modify the SMB share fail with "The request is not supported" when applied to an admin share.

Name ScopeName       AccountName              AccessControlType AccessRight
---- ---------       -----------              ----------------- -----------
D$   *               BUILTIN\Administrators   Allow             Full
D$   *               BUILTIN\Backup Operators Allow             Full
D$   *               NT AUTHORITY\INTERACTIVE Allow             Full
D$   FileServer      BUILTIN\Administrators   Allow             Full       <--- 


ReFS Block Cloning with Data Deduplication Enabled Drive

$
0
0

Hello,

Veeam this year added an experiential feature that allows Veeam to use ReFS Block Cloning on Deduplicated files when the repository is on a 2019 server.

Before I look into enabling this, I wanted to get some clarity into this interoperability between ReFS and Data Deduplication since I can't find much online about this.

From my knowledge I feel having chunked data being written to ReFS that can then be deduplicated and further chunked might cause some problems.

Any insight into this would be helpful, thanks!!

Measuring S2D Storage Performance

$
0
0

Hello,

I am trying to measure a performance of storage on 2 node cluster using Windows 2019 Server DC.

Each node specs:

  • 2x Xeon 2650v4
  • 512GB of memories (DDR4)
  • 1x SAS HDD 256GB for Operating System
  • 1x SSD SAS 800GB for Storage Spaces Direct (Cache)
  • 10x HDD SAS 500GB for Storage Spaces Direct (Capacity)
  • NIC: Mellanox Connectx4 Lx (RDMA Enabled)

I am using VMFleet and DiskSPD and have 20 VMs per node. Using the following settings per each test VM (Windows Server Core 2019 eval):

set-vmfleet.ps1 -ProcessorCount 2 -MemoryStartupBytes 4GB -MemoryMinimumBytes 4GB -MemoryMaximumBytes 4GB

When I launch the VMFleet and look at the metrics: 

.\start-sweep.ps1 -b 4 -t 2 -o 40 -w 0 -d 300

I am getting terrible performance as seen below. What am I doing wrong? How come other's with similar setups are showing 200K+ IOPS and I am seeing such poor results. The IOPS ramp up to 50-70K total and then drop down to below 5K and hover around 500-2500. Regardless of different sweeps performed.

Total IOPS on both nodes are around 3K, same for reads, 42-70 Write. BW (MB/s is 3 for reads, 1 for writes). Latency is 0.002 for reads and 0.384 for writes.

Any ideas?




Windows Server 2019 SMB Share

$
0
0

Hello,

please let me know if anyone can help me with this.

i srote this to the german forum already but i got no reply after 2 Weeks so i hope its okay that i ask the same question here.

Ive installed a Windows Server 2019 with the Feature "SMB1.0/CIFs File Sharing Support" plus the Services SSDP Discovery and the UPnP Device Host.

On an other Server i installed a Storage-Software and created a WORM Storage with and SMB Share.

My problem is i cant get on to this share with Windows Server 2019 but with all other OS (Windows Server 2016/2012R2 Windows 7/10) the connection works fine.

Is there anything i can do?
Windows Server 2019 is up to Date.

Best regards.

Server 2008 R2 sp1 Winsxs folder is 22gb and disk cleanup is only removing 1-3mb

$
0
0
Server 2008 R2 sp1 Winsxs folder is 22gb full and disk cleanup is only removing 1-3mb. What am I missing?

Storage Replica - ReplicationStatus is WaitingForDestination

$
0
0

Hi, 

I'm just setup Storage replica and leave it sync for a month. Simple setup, 1 AD server, 1 Source server and 1 Destination server.

Everything is fine until one day I  and turn it off Destination server for couple hours.

I power it on and leave it for couple hours, check the status it always said "ReplicationStatus  : WaitingForDestination" 

Storage replica status WaitingForDestination

Just wondering is there any command or best practice to re-sync? Or I have to remove replication and redo it again?

Thanks in advance

Expanding volumes more than 15 TB

$
0
0

Hi everybody. 

I have a question, Can I expand volumes more than 15 TB using NTFS? I have a Virtualdisk called IDE HDD1, inside this, i'm configuring a Volume G: called HDD1 with 10 TB, when I try to expand this volume more than 15 TB the following error occurs 

However If I delete the Volume, and try to create this volume using 23 TB, it works well.



Do you know if I can expand volumes more than 15 TB using Powershell?

It's recommended use volumes with 15 TB max? 

Thanks in advance


Swapping out disks in S2D cluster

$
0
0

Hello,

I have a 2 node S2D cluster with 6 CSVs - 3 per node (Windows Server 2019 DC on both nodes) where I would like to swap out the 4TB disks with 8TB disks without blowing it away. Is that something that is supported (I understand there is a risk of data loss at all times), but is there any MS documentation on this and how it should be performed? How can we achieve this?


Thank you!


Storage Replication on MS Filecluster running on Server 2019

$
0
0

We built a new MS Filecluster running on Server 2019 that will replace our old FileCluster.  The cluster has been stretched between two buildings using Storage Replication with one Server 2019 in each building. I have two issues:

1. I need to rename the Fileserver Role in the new File cluster so that it has the name of the Fileserver Role in the old File Cluster.  What is the procedure to do this?  Will it update the Fileserver role's computer object?  And will it update DNS?

2. Prior to installing the FileCluster service on each node, I manually synced the files from the old filecluster to one of the new Server 2019 servers.  Then I used SR to replicate the files over to the other Fileserver.  This worked fine.  Until I realized that I cant see these volumes as available disks when I installed the MS FileCluster Service. 

I then tried to break the SRpartnership.  I used the remove-srpartnership powershell command, but that didnt work.  I then tried the clear-srmetadata commands but that didnt work either because I cant access either the source volume or the replicated volume.  Disk Manager sees the LUNs, turns them online (on the associated node) and even gave it a drive letter, but I get "Access Denied" whenever I try to read them.  How can I access these volumes that were in a SR replication group? 


Storage Spaces Direct (S2D) ReFS Dedup Memory Requirements

$
0
0

I am trying to figure out what the memory requirements are for my situation.

I have the following setup in place:

  • S2D 4 Node Cluster
  • Each node has 4 3.8TB SSDs and 1 1TB SSD
  • 2 Volumes are in place (8TB and 16TB) using Mirror Accelerated Parity (20% mirror)
  • Both volumes have deduplication turned on

After a volume is under high use for a short period of time the memory on the node that owns the volume gets used up to 95%-98%

From what I have read online it looks like S2D wants 4GB of memory for every TB of flash but in this use case I am not using any cache (at least that is my understanding), everything is capacity.

https://docs.microsoft.com/en-us/windows-server/storage/storage-spaces/storage-spaces-direct-hardware-requirements

If I look at what is using the memory with sysinternals RAMMap I can see that it is tied to memory mapped files.  See screenshots.

I'm sure others have run into this just hoping to get a helping hand here.

Thank you!



Take care, Chris Hunt





S2D Available Storage to Expand Vol is Lower Than if Creating new Vol

$
0
0

Hi all,  I've been working on this for a little while and found no solutions, hoping the community can help.  The situation is this.  I have a Storage Spaces Direct (S2D) Cluster of 4 nodes with 4 3.8TB SSD drives in each.  When I create new volumes I can get 3 8TB volumes with a little left over going with the Mirror-Accelerated Parity (20%/80% mix).  Basically taking the defaults from Windows Admin interface.  If I remove, say VOL3, so that I can add space to VOL2 I am only able to add a measly 4TB.  Seems that I don't have access to the rest of the storage, I would expect that I could add the full 8.

If I try to expand VOL2 then I have access to 10.4TB on SSD

If I try to create VOL3 again I have access to 19.4TB on SSD

When I delete a volume from the Windows Admin interface am I supposed to run some extra commands?

I really appreciate any help anyone can provide, thank you!

- Chris


Take care, Chris Hunt


Server 2019 - Dell Storage Ready Nodes - S2D - Slow Server 2008 R2 Guest VM Performance

$
0
0

Hello,

Hoping someone may be experiencing the same issues that we have at one of our customers sites.

We have implemented a 2 node Server 2019 S2D (Azure HCI Stack) solution, the hardware was purchased from Dell and is completely HCI compliant (the nodes we have on the bezels "Storage Ready Nodes", however, I believe its now called Azure HCI Stack).

This is a 2 node ROBO installation, 2 x 25GB adapters for storage networking, back to back (no switches). The VM network and management are in a SET team using 2 x 10GB adapters (going through 10GB switches).

We have implemented these as per the Dell HCI best practice guide;

https://www.dell.com/support/manuals/uk/en/ukbsdt1/microsoft-storage-ready-node-r740xd/robos2dreadynodedg/notes-cautions-and-warnings?guid=guid-5b8de7b7-879f-45a4-88e0-732155904029&lang=en-us

We are seeing no issues in stability at all, the solution is functioning and generally speaking we and the customer are very happy with it.

Of course there is a BUT...

We are finding that certain guest OS's (namely Server 2008 R2) are performing badly on throughput testing (using iPERF). The general performance of the 2008 R2 VMs is great. Booting/Shutting down and running tasks (SQL, Exchange) on them perform brilliantly. Its when clients (Windows 7) or other guest VMs (on the other node to it) try to copy data its very slow.

I have run lots of iPERF tests and I will list them below;

*** S2D Troubleshooting ***

Node Owner for S2D disk makes no difference to performance, either node can be owner and I get the same results.
Any VM on same node (Doesn't matter what Guest OS) - No Issues (Masking the problem) - Using local BUS.

VM to VM (Servers)

Server 2008 R2 writing to Server 2008 R2 different node - 166MBit/s
Server 2008 R2 writing to Server 2019 different node - 2.65GBit/s
Server 2019 writing to Server 2008 R2 different node - 638MBit/s
Server 2019 writing to Server 2019 different node - 3.23GBit/s

Client to Servers

Windows 7 writing to Server 2008 R2 (any node) - 232MBit/s
Windows 7 writing to Server 2012 R2 (any node) - 175MBit/s
Windows 7 writing to Server 2019 (any node) - 1GBit/s

Servers to Client

Server 2008 R2 (any node) writing to Windows 7 - 820MBit/s
Server 2012 R2 (any node) writing to Windows 7 - 800MBit/s
Server 2019 (any node) writing to Windows 7 - 832MBit/s

All Guest Integration Tools are up to date, I have tried previous versions also.

As you can see its an issues with the older OS's, we are planning on removing these (upgrading or building new) soon. I am wondering if we are the only ones seeing this particular issue? I also have a case open with Dell and they are looking into it.

If it makes any difference this was a migration from VMware to Hyper-V 2019 S2D, VMware Tools removed and the VM disks we converted and VM guests created from scratch. No legacy network cards on any VM.

Disabling VMQ makes no difference, same with any TCP offload, this issue seems tied to the guest OS and its driving me mad!

Any thoughts would be very welcome!

S2D Volume type recommendation for exchange 2016

$
0
0

I couldn't find any Microsoft document on recommendations for using S2D volume type for exchange 2016 deployment. As one can create simple,mirror and parity type in S2D world, Which one is recommended? Is it OK to use Simple to save on storage, when we have multiple database active/passive copies? I appreciate your help

Thanks

FSRM quotas enforced but not visible from Mac OS clients

$
0
0

Hi,

I use FSRM on Server 2008 to apply hard quotas on shared folders. None of our Macs correctly display the available space according to the quota but instead show the entire remaining free space for the volume containing the folder. If the quota limit is reached, it in enforced.

Should this be working correctly for Macs? If so, any ideas why it's not working here?

File server migration from failover cluster to stand alone server

$
0
0
Hello,
I have a question about the migration of fileserver resources within a cluster failover.
Currently I have a failover cluster consisting of 2 windows 2012 storage server nodes.
These nodes are physical.
On this cluster there are 3 file server resources that I would like to move into 3 separate virtual machines with windows server 2012 r2 operating system.
What is the best procedure to move the file server resource from the cluster to a stand alone server?

Thank you
Viewing all 13580 articles
Browse latest View live


<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>