Quantcast
Channel: File Services and Storage forum
Viewing all 13565 articles
Browse latest View live
↧

experience slow file server performance

April 29, 2014, 12:06 am
$
0
0

Hi all,

currently we are expecting some strange problems with our fileserver. It is a Windows Server 2012 server with newest hardware. After one to two weeks uptime, Windows 7 clients cannot load their explorer window. It takes a huge time while the green bar in the address bar is trying to load the explorer overview and finally stops with a blank explorer window.

The shares can still be accessed through UNC commands and the fileserver can be reached with a ping command. This error doesn't occur when opening the explorer window on another server.

On the fileserver I can see a lot of open files called "\msftewds". 

I have found an article of microsoft describing this problem but there was no final solution for me:

http://support.microsoft.com/kb/822219/en-us

The only workaround for the moment is to restart the server (what is not that easy because it's productive). After a restart the "\msftewds" files are gone in the open files overview.

Appreciate any help. :)

Best regards

Dennis 

↧
Search

Robocopy not copying ACLs

April 30, 2014, 1:22 am
$
0
0

Take following directory:

D:\Directory\Sub\

D: has explicit permissions and are passed to sub directories.

D:\Directory\ has inherited and explicit permissions.

D:\Directory\Sub\ only has inherited permissions.

I'm using robocopy to make an exact copy D:\Directory\ on a new drive, including ACLs and ACEs, with the following syntax:

robocopy "D:\Directory" "G:\Directory" /e /sec /log:logfile

But not a ACL is copied over. I end up with plain "User" and "Administrators" ACLs on the new directory.

This is being done on a file server, and I'm using a domain administrator to perform the action in an elevated command prompt.

It might be worth mentioning that the storage is a SAN (IBM V7000). The D drive is a raw device, the new drive a plain VMDK.

↧
↧

Windows 2012 R2 RSS on Nic Teams

April 27, 2014, 9:41 pm
$
0
0

Hello,

I'm building an SMB3 SAN for Hyper-V machines (just like everyone else on the planet), and i have a question regarding RSS. I'm trying to implement a "converged" network, so that everything is software defined. 

The server has a total of 6 Nics, which are all Broadcom based, and are all RSS Capable.

My initial reaction was to throw them all into a single team, create a vSwitch, and then create 5 vNics, one for management, 4 for SAN. The SAN nic's are all on different subnets. 

When Teaming RSS nics, and throwing them into a switch, and then creating vNic's on top, am i loosing the "RSS"-ability ? Or will the RSS still function "under the hood" even through get-smbservernetworkinterface is saying the nic's are not rss capable ?

I'm having a hard time justifying SMB3 over iSCSI on the same hardware, when everyone appears to be saying it is going to perform exactly the same, but from my testing - its really not even close. The only advantage is the ease of setup (if you can get over the constrained delegation part!)

Jaysam Thanki

↧

Server 2008 R2 unable to access local DFS Namespaces \\localserver\share

April 30, 2014, 8:59 am
$
0
0

Our primary DFS server is unable to access DFS shares on itself.  \\localserver\share.  All clients (Windows 7) and all other servers can access the namespaces on\\localserver\share and the affected server can access namespaces on all other DFS servers\\server2\share even ones it replicates with.

This is an issue because we have software on this server that rely's on accessing the namespace via the UNC path\\localserver\share. 

It was working up until a couple of days ago and we believe it stopped after applying the latest round of Microsoft patches to the server although this has not been verified as the cause.

As well, yesterday the server BSOD'd in the afternoon.

Finally, we are unable to access \\localserver\C$ from any client or server.

**** Additionally

The server crashes randomly with Stop Error 0x3B, dfsc.sys.  This is on Server 2008 R2 Standard.

↧

File access

May 11, 2014, 9:57 pm
$
0
0

Dear Forum,

I have a root folder - shared permissions read,

sub folders shared permissions read - NTFS full

a folder called Management should be only access by management.

i disabled inheritance - full ntfs for management and admins - all other groups removed.

But still some users have and some have not full access to it.

checking with Effective Access tells me that my users do not have access, except management and admins.

But reality says something different.

How can I start my troubleshooting?

Many thanks for any help in advance

Karen

↧
↧

Problem with Multipath IO

March 4, 2014, 1:54 am
$
0
0

Hello 

I've configured my SCOM service on windows server 2008 r2 sp1, and it's database is placed on 2 disks on SAN storage (HP p9000). since MPIO is configured, sometimes SCOM was stop working. I looked Disk management console and I saw DB disks were gone! This problem solves when I restart the server. This was happening frequently and I've changed the MPIO configuration policies. that get better but it was happening in a bit longer periods. 

I searched and found out I have to install Fibrechannel hotfix (fcinfo_amd64) and Windows6.1-KB2470949-v2-x64 patches. I did this and for last 3 months that was ok but last night it happens again. 

It is our monitoring system and send very important messages. please help me.  

thanks in advance


↧

DFS redesign ideas?

May 12, 2014, 2:16 am
$
0
0

Hi,

Basically I am looking for a few ideas on how to redesign our file servers.

We have multiple physical file servers and a few virtual servers and what is replicated and what is not is quite confusing. Total storage size is around 6TB made up of home directories, and shared resources - no particularly special file types etc. Using DFS with home directories however does mean that I need to essentially have only a single point of reference to be supported by Microsoft as per:

http://blogs.technet.com/b/askds/archive/2010/09/01/microsoft-s-support-statement-around-replicated-user-profile-data.aspx

What I am thinking about doing is consolidating everything onto 4 servers.

We have a large single site with a few remote sites. The remote sites have had their links upgraded to 1Gb and we have been removing our server infrastructure from these areas due to not having environmental/physical space/security in place.

On our main site we have two separate buildings which each contain a SAN (Not linked to each other).

Microsoft's guides show concepts of using a DFS Failover Cluster in a main site with replication to a single server at a remote site.

I could do this model but just in one site, but due to the fact I have two equally sized SAN's on the main site, the issue with this is that I would like to spread the load to both SAN's. Therefore if I have anything running on the single server as primary I am creating an SPOF.

What I am thinking of doing is create 2 x 2 node DFS Failover Clusters (One in each building connected to that building's SAN).

This means:

  • I can load balance the primary DFS shares at the cluster level (SAN's)
  • Rapid failover can occur if needed between individual nodes within a cluster
  • The single point of failure (Storage) in just using a single DFS cluster is eliminated

However I am not sure if this is supported or recommended?


↧

Work folders - certificates with .local domain

May 12, 2014, 3:14 am
$
0
0

Hello,

We'd like to deploy Work folders in our domain.local environment.
Technet states that the certificate name should be the public URL workfolders.domainname and that for every file server a SAN needs to be listed. I was wondering how we need to implement Work folders as you can't add local server names anymore to public certificates. For us, the public URL would be workfolders.company.eu and the server name is fileserver.company.local. Anyone already built a setup like this?

↧
Search

Strange Defrag error 0x80070057

November 19, 2013, 11:45 pm
$
0
0

Hello everyone,

I am observing a strange issue with defrag on Windows Server 2012 R2. There is a hardware RAID volume using deduplication in VDI mode and containing some virtual machines. Deduplication works okay, chkdsk shows no logical errors and RAID is consistent, but defrag reports error "0x80070057 The Parameter is Incorrect" after some minutes or even hours of runtime.

I've tried to search for such error but all I've found was about SSD that do not need defragmentation at all. I have also found an advice to set the CopyFileBufferedSynchronousIo to "1" but I am a bit unsure if it is safe in my case.

Any help is greatly appreciated.

Here's a sample output of defrag (sorry for Cyrillic localization):

C:\>defrag d: -u -v
Оптимизация диска (Майкрософт)
(c) Корпорация Майкрософт (Microsoft Corporation), 2013.

Вызов дефрагментация диска на Data (D:)...

        Анализ:  выполнено на 100%.

Отчет о предварительной оптимизации:

        Сведения о томе:
                Размер тома                 = 1,59 ТБ
                Размер кластера                = 4 КБ
                Используемое место                  = 961,62 ГБ
                Свободное место                  = 674,03 ГБ

        Фрагментация диска:
                Общий объем фрагментированного пространства      = 30%
                Среднее число фрагментов на один файл  = 21,56

                Перемещаемые файлы и папки   = 10757
                Неперемещаемые файлы и папки = 8

        Файлы:
                Фрагментированные файлы            = 1003
                Общее число фрагментов файлов        = 211688

        Папки:
                Общее число папок               = 22
                Фрагментированные папки          = 0
                Общее число фрагментов папок      = 0

        Свободное место:
                Число свободных мест            = 12470
                Средний размер свободного места     = 55,33 МБ
                Максимальный размер свободного места     = 3,72 ГБ

        Основная таблица файлов (MFT):
                Размер MFT                    = 36,50 МБ
                Число записей MFT            = 37375
                Использование MFT                   = 100%
                Общее число фрагментов MFT         = 2

        Примечание. В статистику фрагментации не включаются фрагменты файлов
змер которых превышает 64 МБ.

Выполнение прохода 1:
        Дефрагментация диска:  выполнено на 100%.
        Объединение свободного пространства:  выполнено на 100%.

Выполнение прохода 2:
        Дефрагментация диска:  выполнено на 100%.
        Объединение свободного пространства:  выполнено на 100%.

Выполнение прохода 3:
        Дефрагментация диска:  66% выполнено...
Параметр задан неверно. (0x80070057)


↧
↧

Is It Possible to Add a Fileserver to a DFS Replication Group Without Connectivity to FSMO Roles Holder DC But Connectivity to Site DC???

May 9, 2014, 7:28 am
$
0
0
I apologize in advance for the rambling novella, but I tried to include as many details ahead of time as I could.

I guess like most issues, this one's been evolving for a while, it started out with us trying to add a new member 
to a replication group that's on a subnet without connectivity to the FSMO roles holder. I'll try to describe the 
layout as best as I can up front.

The AD only has one domain & both the forest & domain are at 2008R2 function level. We've got two sites defined in 
Sites & Services, Site A is an off-site datacenter with one associated subnet & Site B with 6 associated subnets, A-F. 
The two sites are connected by a WAN link from a cable provider. Subnets E & F at Site B have no connectivity to Site A 
across that WAN, only what's available through the front side of the datacenter through the public Internet. The network 
engineering group involved refuses to route that WAN traffic to those two subnets & we've got no recourse against that 
decision; so I'm trying to find a way to accomplish this without that if possible.

The FSMO roles holder is located at Site A. I know that I can define a Site C, add Subnets E & F to that site, & then 
configure an SMTP site link between Sites A & C, but that only handles AD replication, correct? That still wouldn't allow me, for example, 
to enumerate DFS namespaces from subnets E & F, or to add a fileserver on either of those subnets as a member to an existing
DFS replication group, right? Also, root scalability is enabled on all the namespace shares.

Is there a way to accomplish both of these things without transferring the FSMO roles from the original DC at Site A to, say, 
the bridgehead DC at Site B? 

When the infrastructure was originally setup by a former analyst, the topology was much more simple & everything was left
under the Default First Site & no sites/subnets were setup until fairly recently to resolve authentication issues on 
Subnets E & F... I bring this up just to say, the FSMO roles holder has held them throughout the build out & addition of 
all sorts of systems & I'm honestly not sure what, if anything, the transfer of those roles will break. 

I definitely don't claim to be an expert in any of this, I'll be the first to say that I'm a work-in-progress on this AD design stuff, 
I'm all for R'ing the FM, but frankly I'm dragging bottom at this point in finding the right FM. I've been digging around
on Google, forums, & TechNet for the past week or so as this has evolved, but no resolution yet. 

On VMs & machines on subnets E & F when I go to DFS Management -> Namespace -> Add Namespaces to Display..., none show up 
automatically & when I click Show Namespaces, after a few seconds I get "The namespaces on DOMAIN cannot be enumerated. The 
specified domain either does not exist or could not be contacted". If I run a dfsutil /pktinfo, nothing shows except \sysvol 
but I can access the domain-based DFS shares through Windows Explorer with the UNC path \\DOMAIN-FQDN\Share-Name then when 
I run a dfsutil /pktinfo it shows all the shares that I've accessed so far.

So either I'm doing something wrong, or, for some random large, multinational company, every sunbet & fileserver one wants 
to add to a DFS Namespace has to be able to contact the FSMO roles holder? Or, are those ADs broken down with a child domain 
for each Site & a FSMO roles holder for that child domain is located in each site?
↧

Share folder + Allow user to see only the folder he has access to

May 12, 2014, 7:55 am
$
0
0

Hi,

I have a shared folder called Backup, under this folder there's folders for each user in my organization which can only be accessed by the username associated with it.

When an user access the share \\server\Backup, the user can see all other users folder.

For example

\\server\Backup - when John access he can see his own folder, Bob and Bill's folder (he cannot access Bob and Bill's folder)

I want that when John try access \\server\Backup he can only see his own folder

Is that possible?

Thanks,

Vinicius Deschamps

↧

Fileservice (smb) unresponsive - other service (TCP) available - smbclient event 30809

November 29, 2013, 3:49 am
$
0
0

Problem: Connections to fileshares(smb) on windows 2012 r2 server are suddenly not possible. This server is also a AD DS server and also GPO`s can`t be retrieved by other clients (example w2008 r2 RDS servers).
Rebooting the server will solve it. No 3rd party software on the server.
Seems no relationship with IO intensity. Events are logged outside of the normal working hours and outside of the backup window.

EventId: 30809 - SMBclient is logged on the windows 2012 r2 server.

A request timed out because there was no response from the server.

Server name: \DOM001
Session ID:0x0
Tree ID:0x0
Message ID:0x0
Command: Negotiate
Guidance:
The server is responding over TCP but not over SMB. Ensure the Server service is running and responsive, and the disks do not have high per-IO latency, which makes the disks appear unresponsive to SMB. Also, ensure the server is responsive overall and not paused; for instance, make sure you can log on to it.


↧

The IO operation at logical block address # for Disk # was retried

February 20, 2013, 2:52 am
$
0
0

Hello everyone,

A warning appears in the system log:
===
Log Name:      System
Source:        disk
Date:          2/20/2013 1:00:28 PM
Event ID:      153
Task Category: None
Level:         Warning
Keywords:      Classic
User:          N/A
Computer:      STRANGE.aqa.com.ru
Description:
The IO operation at logical block address af7ff for Disk 7 was retried.
Event Xml:
<Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event">
  <System>
    <Provider Name="disk" />
    <EventID Qualifiers="32772">153</EventID>
    <Level>3</Level>
    <Task>0</Task>
    <Keywords>0x80000000000000</Keywords>
    <TimeCreated SystemTime="2013-02-20T09:00:28.199176700Z" />
    <EventRecordID>12669</EventRecordID>
    <Channel>System</Channel>
    <Computer>STRANGE.aqa.com.ru</Computer>
    <Security />
  </System>
  <EventData>
    <Data>\Device\Harddisk7\DR142</Data>
    <Data>af7ff</Data>
    <Data>7</Data>
    <Binary>0F01040003002C00000000009900048000000000000000000000000000000000000000000000000000020828</Binary>
  </EventData>
</Event>

===
This warning occurred in several seconds after the Windows Server Backup start. Our backup job finishes successfully. That server is in provisioning without a heavy workload, and we have not experienced any problem yet. But we do not want to face any problems due to this error in the production environment.
All disks of the server are managed by the LSI MegaRAID controller, which doesn’t report any errors in the disk system.

it is Windows Server 2012 with the latest updates.

↧
↧

Access control entry flags

May 12, 2014, 12:57 pm
$
0
0

Hi! Have the ACE flags changed from earlier Windows Server versions to Server 2012 R2?

I'm a bit confused, checking a Server 2012 fileserver we have, that was just upgraded from 2003 I think.

I was reading this article on the ACE flags: http://ss64.com/nt/icacls.html and http://support.microsoft.com/kb/220167/en-us
where it says

 (OI)(CI)(IO) Subfolders and files only.

(OI)(CI)     This folder, subfolders, and files.

I check a folder like this with ICACLS, and it says:

.\ConvertX\ BUILTIN\Administrators:(I)(F)
            CREATOR OWNER:(I)(OI)(CI)(IO)(F)
            NT AUTHORITY\SYSTEM:(I)(OI)(CI)(F)
           %DOMAIN%\VSWCOM-S-Applications:(I)(OI)(CI)(M)
            BUILTIN\Administrators:(I)(OI)(CI)(IO)(F)

The screenshot shows that VSWCOM-S-Applications is correct according to the above.

But the screenshot shows that for  BUILTIN\Administrators this applies: This folder, subfolders, and files.

I don't quite get ii

↧

Repair Degraded Storage Spaces Virtual Disk - All physical disks show healthy

May 12, 2014, 6:37 pm
$
0
0

I've seen this mentioned a few times but no definitive answers. I have a Storage Pool with 8 physical disks. Each physical disk shows as healthy and auto select. I have thick provisioned Virtual Disk created on this Storage Pool in a parity configuration. This virtual disk is configured for the maximum amount on the storage pool. 50% of the space is available on the volume of the virtual disk...I click repair it does nothing.

Tested yanking a physical disk. Virtual disk becomes unavailable. No redundancy currently. How the heck do I fix this???

↧
Search

possible NFS version conflict when implementing LINUX BOOT over Windows deployment server

May 7, 2014, 7:05 pm
$
0
0
I have configured windows deployment service machine running on windows server 2008 r2 which I was using to install any version of windows.
I tweaked this server to be able to install the linux and major part is done. I can boot client linux installation and beginning of installation goes fine (successfully downloads vmlinuz kernel and initrd). However during later stage I had to supply NFS information to client setup. 

In order to do so I had to create NFS share in Windows Server 2008 r2 machine from server manager -> file services -> share and storage management. 
However it still complained about the share and from the packet sniffing program, I suspected the version conflict.
During the nfs request, I can see the nfs request coming from client is NFS V4 whereas the reply decoded by packet sniffer is min/max supported version is V2-V3 respectively.

Here is the screenshot of the packet sniffer:
https://www.dropbox.com/s/kmhya3vblwqds ... .V2.V3.PNG


Assuming my presumption is correct, is there a way to circumvent? For me the best way is to enable V4 NFS support on WINDOWS 2008 however at this time I have no clue on this one. 

Does someone have any suggestion for me? Thanks!
↧

Storage Spaces Tiering Service running forever, killing user performance

April 28, 2014, 5:41 pm
$
0
0

I just installed a new Storage Server 2012 R2 OEM box, with a single mirrored storage space using 4x Samsung 840 Pro SSDs and 12x 3TB Toshiba 7200RPM HDDs. I used the wizard to configure this as a tiered space with a single large virtual drive, and it appears to have configured with 2 columns and a 1GB writeback cache on the SSDs, as well as a proper tiering scheme between the SSDs and HDDs.

The scheduled task starts tiering maintenance at 10pm (I moved it back from 1am), and tiering runs all night with 100% disk activity at very low throughput (0.5 - 3.0 MBps), until I stop the Storage Spaces Tiering service in the AM when users arrive, as they report bad storage response times while it runs - very little data is added to the disk during the day (say, 5 - 10GB), and usage patterns don't change much, so I'm curious as to why this is happening.

I've let the service run a few times over weekends, and it sometimes "finishes" a run, but the next night it will go right back to churning at low throughput and continue into the morning until a manual service stop.

Any ideas?

↧
↧

DFS Namespace - Standalone / Exclude Targets Outside of the Clients Site

May 9, 2014, 5:26 am
$
0
0

Hey,

I've set up a DFS namespace in standalone mode (Server 2008 R2) due to the DFL. If the namespace is hosted on Server1 in Site1, what should I expect to happen if the 'Exclude Targets Outside of the Clients Site' option is checked and a user from Site2 tried to access \\server1\namespace\folderA?

FolderA has 2 targets, Server1 and Server2.

Thanks for any help.

↧

Windows 2008 connected via UNC Session to a isilon

April 28, 2014, 8:07 am
$
0
0

Hello,

following "Problem".

Isilon is a NAS System which provied NAS Space via SMB or NFS. In our solution the Windows Server 2008 connect the Fileshare via UNC Path witch works good.

The Isilon also provides the possiblity to distribute all incoming connections via a Round Robin DNS Service directly on the the Isilon across all front end NICs.

The problem now is when we connect to the ISILON with UNS the Server makes a DNS Lookup for the right IP-Adresse. But when the Server starts another session via UNC to the same "target" but to a other file while the first session is also opened the Server don´t  make a new DNS query. The TTL of the first DNS lookup was 0 so there is nothing in the DNS Cache. So what is the Problem? Where is the UNC Information stored on the Server. I need the possiblity that the Server starts a dns query for all UNC Sessions.

Can anybody help?


↧

Long file folder path restore

May 11, 2011, 8:12 am
$
0
0

Hello,

Within the company I work for there is a Windows Server 2008 Standard file server that has a few folders with long file and folder names. I have brought to the departments attention that the file paths must be considered. Meantime they have many folders that have reached the 256 character limit. Unfortunately they were allowed to save those folders and files anyway, and were backed up that was using Replay 4, which uses VSS. Now these same folders cannot be restored because when saving to the restore destination, the path is too long. Of course the backup software vendor points to the Windows limitation. My question is, is there a way around this problem, a reg tweek, or anything I can use to restore these folders as they were backed up?

Thank You

↧
Viewing all 13565 articles
Browse latest View live
Search
<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>