I have a script that deploys a WIM to a partition and makes it bootable and it uses an unattend.xml to configure the name and everything else. I include the BackInfo.exe with settings for it and when the OS boots up I have it run a logon script to initialize the OS for a bunch of stuff. I have tried in the script to run the BackInfo.exe, I have tried in the unattend.xml, I have tried adding a registry key into HKLM\SOFTWARE\Microsoft\Windows\CurrentVersion\Run, I have tried adding a scheduled task.
No matter what happens the background is always the default one and never set to the generated one from BackInfo. When I run the command manually it works perfectly. I for the life of me can't seem to get this working.
Any help would be very much appreciated!
So turns out I had placed the code into the FirstLogonCommands section which turns out wouldn't work, I changed it over to LogonCommands.
<LogonCommands>
<AsynchronousCommand wcm:action="add">
<CommandLine>%SystemDrive%\Applications\BackInfo\BackInfo.exe</CommandLine>
<Description>BgInfo</Description>
<Order>1</Order>
</AsynchronousCommand>
</LogonCommands>
Hopefully this can help someone else in the future and prevent them from wasting time trying to figure out other ways.
Related
Before explaining what my problem is, please know that I have looked up for solutions on similar topics but none of them seems to work nor even to corresponds to my problem.
What I am trying to do:
I have this python code on multiple files that I run with flask with the following command:
python -m flask run --host=0.0.0.0
So far, everything works, but I would like this code to automatically run everytime the computer boots. In the future this will be used on mini PCs without any graphical interface nor human intervention.
Since I need to do some configuration checks before running the web server, I've created a powershell script that ends with Flask running (using the previous command).
So far, everything works too. Now we're coming to the problem:
I'd like this script to run when I boot the machine. Specificity: Every things needs to work with Administrator privileges, on the local system without any interaction.
I've tried scheduled tasks but Flask won't run even if the rest of the script works (like creating folders or other things)
Ok, it's not a big deal I have other ways to do it, so I've created a Windows Service in C# to run the Script at startup on the local system.
The script works, I've checked the privileges too, everything's fine but arriving at the flask command line that is supposed to make it run, nothing works.
It's the same thing if I run flask using "pythonw" which is supposed to run python as a background process.
What the problem seems to be:
Well, as long as I run flask and I have either a command prompt or a powershell terminal, everything works greats. But if in a way or another I run the script as a background process, it won't work.
Normally it would take around 30 seconds for Flask to start-up. Here if I try to create a folder right after flask ended starting up (as a test) I can see the folder is created almost instantly, which means the process is immediately killed.
The problem doesn't seem to come from the service itself but really Windows that kills the process I don't know why
I'm running out of idea so if you guys have anything that I could try it would really help me.
I've been finding recently that my KDB launch script is crashing. I need to restart my computer and then it will run fine. I launch KDB from a CMD prompt and I can't figure out where to look to see what process is running in the background which is causing it to crash. Does anybody know what I should check?
Thanks.
Trying running a blank kdb instance with just:
set QHOME=C:\q
set PATH=%PATH%;c:\q\w32
and then
q -p 1234
(doesn't need to be in a batch script, you should be able to just run those commands in a cmd prompt). If kdb comes up then kdb isn't the problem.
Then try to manually load the startup script to see if the problem is there:
q)system"l C:\\q\\ServerFiles\\server.q"
If this works fine then the problem isn't there. The last place it can be is in the database/directory load, so load that:
q)system"l c:\\q\\files"
If none of these cause an error then something else is affecting your kdb instance, either something running on a timer (check .z.ts) or something is connecting externally
So I did finally discover what the issue was. I had a few instances of KDB that would run via Task Scheduler in the evening and then close. One of the instances wouldn't end and so it was continually running. This would cause KDB to crash since there was already an instance running. The restart would stop the process so it could be run again. I deleted the task scheduler event and recreated it and now the event runs KDB and closes the way it is supposed to. Thanks for all of the help in trying to figure this out!
At the company I work for, we use Bit9 as part of our security stack. We are in the process of upgrade the version to 8.0 (and eventually 8.2) on all of our devices. Between the automatic upgrades and a different script I wrote, I was able to upgrade about 1000. But there are still about 700 left where the CLI password from Bit9 is not working, and the devices are not checking in to allow auto upgrade.
Bit9 has come back and suggested the following:
Boot the endpoint into Safe Mode w/ Networking
Run a script that executes the following Administrative commands from a CMD prompt (please note the proper spacing between start= disabled):
sc config parity start= disabled
sc config paritydriver start= disabled
Boot into Normal Mode
I've written a script that is supposed to do all of this, except I cannot for the life of me get the script to run once the device starts in safe mode. I've tried everything that I can think of:
HKLM:\SOFTWARE\Microsoft\Windows\CurrentVersion\Run
HKLM:\SOFTWARE\Microsoft\Windows\CurrentVersion\RunOnce
Placing a batch file in the Startup folder
Scheduled task
Nothing I do is working.
And one other caveat, the user that is logging in (safe mode & normal mode) is not an administrator. It is a regular user. I am kicking off the initial script with BigFix, which does run as an administrator.
Thanks in advance.
I was finally able to get it to work. The service idea was the starting point, though instead of creating a "fake" service, I actually wrote a service in C# that calls the PS script. I was then able to edit the registry, where I made that service able to start in safe mode. That seemed to do the trick. Now the only issue is that I can't seem to disable safe mode programmatically, but I will ask that as a separate question.
I have powershell script which is present on chef server to run on remote windows server, how can i run this powershell script from chef server on remote windows server.
Chef doesn't do anything like this. First, Chef Server can never remotely access servers directly, all it does is stores data. Second, Chef doesn't really do "run a thing in a place right now". We offer workstation tools like knife ssh and knife winrm as simplistic wrappers but they aren't made for anything complex. The Chef-y way to do this would be to make a recipe and run your script using the the powershell_script resource.
Does it mean chef is also running on Windows server ?
If yes, why not to use psexec from Windows Ps tools ?
https://learn.microsoft.com/en-us/sysinternals/downloads/psexec
Here is my understanding of what you are trying to achieve. If I'm wrong then please correct me in a comment and I will update my answer.
You have a powershell script that you need to run on a specific server or set of servers.
It would be convenient to have a central management solution for running this script instead of logging into each server and running it manually.
Ergo you either need to run this script in many places when a condition isn't filled, such as a file is missing, or you need to run this script often, or you need this script to be run with a certain timing in regards to other processes you have going on.
Without knowing precisely what you're trying to achieve with your script the best solution I know of is to write a cookbook and do one of the following
If your script is complex place it in your cookbook/files folder (assuming the script will be identical on all computers it runs on) or in your cookbook/templates folder (if you will need to inject information into it at write time). You can then write the .ps file to the local computer during a Chef converge with the following code snippet. After you write it to disk you will also have to call it with one of the commands in the next bullet.
Monomorphic file:
cookbook_file '<destination>' do
source '<filename.ps>'
<other options>
end
Options can be found at https://docs.chef.io/resource_cookbook_file.html
Polymorphic file:
template '<destination>' do
source '<template.ps.erb>'
variables {<hash of variables and values>}
<other options>
end
Options can be found at https://docs.chef.io/resource_template.html
If your script is a simple one-liner you can instead use powershell_script, powershell_out! or execute. powershell_out! has all the same options and features as the shell_out! command and the added advantage that your converge will pause until it receives an exit status for the command, if that is desirable. The documentation on using it is a bit more spotty though so spend time experimenting with it and googling.
https://docs.chef.io/resource_powershell_script.html
https://docs.chef.io/resource_execute.html
Which ever option you end up going with you will probably want to guard your resource with conditions on when it should not run, such as when a file already exists, a registry key is set or what ever else your script changes that you can use. If you truly want the script to execute every single converge then you can skip this step, but that is a code smell and I urge you to reconsider your plans.
https://docs.chef.io/resource_common.html#guards
It's important to note that this is not an exhaustive list of how to run a powershell script on your nodes, just a collection of common patterns I've seen.
Hope this helped.
I am trying to user the package manager console it will not get past Initializing PowerShell Host....
I try to open PowerShell directly and I get
I have had this before and eventually had to restore computer to prior date. I tried that and it changed nothing. Does anyone have any ideas?
UPDATE
Information under event viewer says
Windows PowerShell has started an IPC listening thread on process: 22300 in AppDomain: DefaultAppDomain.
I Restored computer again and this worked. I tried this once, it didn't work, but it did on third time of trying.
Clearing cache did the trick for me, no idea why though.