kankamuso's questions

Under a 2012 AD, as usual, users cannot set date/time on their computers as this is managed by organization. For specific purposes, we need to let one of the users set date to past but options are greyed out. Through GPO we have granted permission to change date/time and timezone to that user. Policies are updated and seem correct as rsop.msc for that user shows that directive is correctly enabled... But no luck,still greyed-out option to change time by hand.

Is there any directive I am missing to allow this?

EDIT: Managed to set time through classical control panel but it reverts back to NTP after a moment.

I am experiencing sudden Windows Server 2012R2 STOPS since a week ago after some Citrix drivers (this is a xcp-ng virtualized system) failed. Since, I installed manually newest drivers and the system was again running smooth. Nevertheless, from time to time we get a BSOD. I finally got minidump (it was not being generated initially, before drivers were corrected). Here you are the last minidump analysis. I am pretty sure this is a network driver issue but I am not sure on how to proceed.

Not only this dump points me into that direction but also the fact that I lost VPN correct functioning with clients stablishing connection but losing Internet navigation and being incapable to connect to other network computers or see them through pin.

Also, sfc /scannow hangs at 69-70%.

Suggestions on how to proceed safely to isolate the real problem and solve it?

Everything is up-to-date: OS, Virtualization host, drivers, etc...

Thanks in advance, Jose

I have a production 2012R2 virtualized server (previously Xenserver, migrated to xcp-ng, 8.0 currently) that has been running fine for 4 years. Recently, the system tried to update Xen drivers and it failed, making the server sloooow and unusable. I was advised to migrate Xen drivers to XCP-np drivers as my system was not accepting Xen Drivers anymore. This worked and the system was again up and running fast. Nevertheless, I have been experiencing sudden reboots (once per day or couple of days) but cannot find any clue on the system logs to what is causing it. The last event I can find when the shutdown happened is 7036 related to VSS but this seems to be a normal event in the service life.

EDIT: Does not seem to be related to VSS as another reboot is preceded by a different service message (that, by the way, look a normal life event of service started/service stopped).

How can I debug the problem down to its root to find out what is causing it?

I own a HP DL380 Gen10 with a HPE Ethernet 10/25Gb 2-port 640SFP28 (817753-B21) network card. The thing is that now I need 1gbps RJ45 connectors and those are not (strangely) present in the server. There seems to be no SFP+ adapters including RJ45 connectors (at least not listed at HP docs).

My question is regarding compatibility, in general:

• Can I use non-HP SFP+ with this server?
• Can I use SFP transceivers on a SFP28 connector? I understand I can use SFP on SFP+ ports and SFP+ connectors on SFP28 ports so I would suppose this would be also true for SFP-SFP28 but something tells me don't do it.

What are you experiences with this issue? SFP* are way cheaper than purchasing a new PCIe network card.

Thanks in advance, Jose

I have a WS 2012 R2 VM running under Google Cloud Platform. It is using a datacenter WS license provided by Google. What are the options to export that VM to an on-premises infrastructure and keep WS running? Do I need a datacenter license or can I re-activate (if exporting is possible under such scenario) using WS 2012 R2 Standard?

Thanks in advance

I am used a backup software (under Windows Server 2012 R2) that seems to use smb to write to the remote backup device. I have realized that, upon completion, credentials are "still there" so a ransomware attack would make its way to the backup copies (no worries as we have external copies). Nevertheless, I have tried with no success to remove such credentials.

If I do it through command line, it seems to work (if the files explorer is not open):

net use \\server\share /d
klist purge

Nevertheless, if I include both files on a batch file and invoke it from my programmed tasks, it does not seem to work, only when invoked from the command line. I have created the programmed task using the domain Administrator account (the same I use when invoking from the command line)

What can be the reason for this behaviour?

EDIT to add expected behavior

I expect to be unable to access the network share without being prompted for the user/password. This does happen whenever I run the script from the command line but not when the script is invoked from a scheduled task. Reminder: the task is run under the same user that is running the script from the command line.

My remote desktop server users are having a lot of problems because, from time to time, one shortcut on their desktop loses its path.

The program is installed in C:\MyApp\app.exe but due to app design, it must be mounted through network share. Thus, the shared is created as:

T: -> \SERVER\Myapp

And from there I create the shortcut which has on its path: T:\app.exe

Everything goes well but, randomly in time, the path goes back from T:\app.exe to C:\MyApp\app.exe and stops working.

This is a 2012R2 box but I cannot find any clue as to what this is happening.

Any hints?

We currently have two (2012 and 2012 R2) DC but SYSVOL seems to be corrupted as we cannot apply GPOs due to permissions complains (from either server). But we don't have a valid system backup so GPOs and AD cannot be restored completely.

We have tried to restore permissions in both filesystem and GPOs but it does not help. We have gone through every and each "solution" we could find on the internet with no luck.

Can anyone provide the per directory permissions and groups as well as for the GPOs?

If we finally must restore the system from scratch, how can I keep users profiles and passwords for existing users?

Is there a way to install a new server, promote to DC and get a clean SYSVOL without loosing AD information?

What would happen with PCs where current domain users are login in if we do demote all current DCs, install a new one with the same domain name and create users with same credentials by hand? Would the still be able to log in or there is some hidden information that will detect that server has changed and will draw such accounts unusable even after rejoining the new (same name) domain?

What do you recommend in this scenario?

Thanks in advance

If E3-1285 v6 supports a maximum of 64 GB of RAM, does using a dual socket motherboard increase max memory? I guess not, but would like to know the reason. My logic says that if RAM is shared, both processors should be able to address all available RAM and thus, it will also be limited to 64GB. Any technical explanation for this?

As far as I know, I won't be able to use RDS under WS2016 Essentials edition (only the Gateway role). Is there any workaround to this using Multipoint or something so my users can work in the office with their own session on the server (just like a desktop)?

Thanks in advance

Perhaps this is quite a silly question. I installed a new VM on XenServer. It loads the full system from a iSCSI from a NAS. Now I am planning to migrate to Proxmox as XenServer falls short regarding USB passthrough. Nevertheless, I don't have a clue on how to:

a) Access the iSCSI disk and convert it to another format (qcow2 in this case, but the format is not of importance now). b) Get that image out from the system. c) Attach it again on the new machine to the iSCSI NAS.

I don't know how to do this because I cannot "read" the iSCSI partition from the NAS (QNAP). From XenServer, I know how to create a snapshot to local storage, so It would not be so difficult to migrate that image. But... how to go back to iSCSI later?.

Any suggestion is welcome.

Thanks in advance

I have a 2012 R2 server with some RDS users. One of the applications, that is locally installed, is asking them for administrative permissions everythime it runs. This annoys my users and the only workaround I have found is to add them to the Administrators group which is a wrong solution.

Users and server belong to a domain. I have tried adding read/execute permissions for "Users", "Domain users" and nothing works. Also tried on the "compatibility tab" to set "Run as administrator for all users" with no luck.

It seems UAC us working here, but I need to solve this as no user wants to enter credentials each and everythime they run the application.

This is a trusted application but unsigned... what ca we do?.

Thanks in advance

I have migrated a server from physical to virtual using XenServer 7. Nevertheless, we need now a USB dongle for security purposes but have not been able to find the way to properly pass it to the virtual machine.

Most resources point to using USB over Ethernet, but this is not an option as we have no more hardware and this must be kept on the CPD.

There must be a way to do this. Additionally, we have another device which is connected to the old server through RS232. We have purchased an USB adapter for it and also want to passthrough it... Let's cross fingers.

I have had no problems with this in the past using Proxmox, but we are very surprised that a "more commercial" product like XenServer does not provide this mechanism out of the box.. :S

Any manual on how to do this?

Thanks in advance

I need to migrate some databases from one server to another to decommission the first one. This server is sharing a network share and a database. Prior to migration and to ensure data integrity, I need to avoid users connections to DB as well as network resources. How can I kick all users connections (and keep them disconnected) but administrator ones?.

Both servers are now DC in case this is important.

I have two Hyper-V Server (core) host machines. Both have access to a NAS shared storage where VMs disks are stored. If I create a new VM on host 'A' and, after some time, I need to move it to host 'B', how do I manage it considering:

a) I cannot build a high availability cluster, so no quick migration is possible. b) a) is not possible because I caanot have a 3rd machine to act as a DC.

So, prior to mounting all this scenario, I wanted to know how to do this without the need to copy all data. I mean, is this possible?:

1. Switch off VM on host a)
2. On host b), create a new VM with the same shared storage
3. Swith newly created VM on host b)

This brings some clear problems to me as the loss of virtual NICs IDs, MACs and so on... so, is it possible to clone the VM parameters to host b) prior to migration and keep it there, switched off so at any time I can just swith off a) and restart on b)?.

Thanks in advance,

Jose.

I am installing a new host server based on Hyper-V server. Currently it has a network cable plugged-in and four more free network interfaces (including one ILO). When I start the server and reach the blue command line screen, I can just configure IP for the pluged interfaces. As this machine is intended to be configured off-site and moved to final location, I cannot plug all connectors right now, but need the different IPs configured prior to moving.

My question is how (if any) can I access the configuration options for unplugged network interfaces?.

Thanks in advance,

Jose.

We are planning to move from an on-premises solution to a hosted one for our office applications. Office 365 includes Access for some of its Enterprise plans (namely, E3 and E5 I think). As we are currently using Access, some doubts about migration, compatibility and needed products arise:

• Are there still restrictions (like it used to be with VBA) for the migration of Access to cloud solutions?.
• Which products do I need to purchase separately in order to be able to share an Access database among all the colleagues? (I have read about Access Services, but I don't know if this implies I also need Sharepoint, if sharepoint is included with Office 365 nowadays... I cannot come to a conclusion).

Hope someone cane shed some light on this as I am always afraid of Windows ecosystem limitations and extra, unexpected costs and incompatibilities...

Thanks in advance,

Jose.

I have just installed a brand new W2012R2 that includes sensible medical data folders. I was considering Bitlocker but as I need remote access to the server, it will not be a good alternative for me (although I could access the server using iLo from Hp) to enter boot password each time the server is rebooted. Anyway, I have read from EFS and as it can work on the folder level, I have two main questions regarding it:

A) What will happen with scheduled backup tasks that run when the user is not logged-in?. Does EFS work under such scenario?. I understand EFS only decrypts data when you are logged-in (in this case as Administrator). The backup app will do its own encryption so data should not be encrypted.

B) Again, such folders are mounted externally as network shares. Therefore, can they still be accessed from other computers when Admin is not logged-in? (Intuition tells me it does work as we input credentials to access such shares).

Thanks in advance,

Jose

I have a legacy machine from IBM (A Netfinity 7100) running a Windows 2000 SP4 server OS. We are trying to move it to a virtual environment (KVM under Linux) to decommission this old, costly hardware. We have tried everything we know so far with no luck. This machine is also configured with SCSI drives attached to a RAID card. What we have tried so far includes:

-Clonezilla: Fails to load as it seems to lack some drivers (I think it is the RAID card one).

-VMWare Converter: No way to install it. We don't have enough free hard drive space and, although it allows us to select another disk for installation, it wont install and will insist on doing it on the System drive.

-Citrix Converter: Again, no way to install it for the same reasons.

At this point we have no more ideas to proceed with. Has anybody got any experience on such systems?. I know P2V under W2000 can be more painful than other, newer Windows releases as I have performed this, but that it can be done.

My main concern is the RAID driver, but I suppose that if I find any Linux distro that works on this hardware, there will be any procedure (like dd and ssh) to achieve this.

By the way, I will also love if someone can provide me with some alternatives (software, procedures, linux ways to do this, etc..) to achieve this.

Thanks a lot in advance,

Jose

This is my first question on this forum. Hope I will be able to help others in the future.

I have two boxes with two NICs each. I know that using bonding I can have some degree of fault tolerance but I feel I need something else. What I want is different traffic (PING, LOGIN, etc..) following one path and the other traffic (DRBD, web, etc..) following the other one.

Just in the case that one of the links fail, I want part of the other traffic (in my case, just IP for a heartbeat mechanism) following the same path as DRBD and web do. Is this possible?, what do I need to monitor links status and acct accordingly?. I would know how to do this statically (iproute and so on), but need it dynamically...