IIS logging has been disable via the GUI:
I just want to enable it, but apparently this is impossible without using the GUI.
Commands I have tried:
appcmd set config /section:httpLogging /dontLog:False /selectiveLogging:LogAll
Set-ItemProperty "IIS:\Sites\MySite" -name logFile -value #{enabled=$true}
Set-WebConfigurationProperty -pspath 'MACHINE/WEBROOT/APPHOST' -filter "system.applicationHost/sites/site[#name='MySite']/logFile" -name "LogExtFileFlags" -value #{enabled=$true}
When I do a Get-Windowsfeature I can see IIS-HttpLogging is marked as enabled.
Nothing I have tried has enabled IIS logging, I have to use the GUI and click enable.
When I run this it returns true:
PS C:\Windows\System32\inetsrv> (Get-WebConfiguration -PSPath 'MACHINE/WEBROOT/APPHOST' -filter "/system.applicationHost/sites/MYSite" ).logfile.enabled
Edit:
So logging seems to work even though the GUI displays it as "disabled". Perhaps this is a bug in the GUI?
Edit2:
OK so if I try the appcmd command and target my specific site I get this error:
.\appcmd set config "MySite" /section:httpLogging /dontLog:False
PS C:\Windows\System32\inetsrv> .\appcmd set config "MySite" /section:httpLogging /dontLog:False
ERROR ( message:Can not set attribute "dontLog" to value "False".. Reason: This configuration section cannot be used at this path. This happens when the section is locked at a parent level. Locking is either by default (overrideModeDefault="Deny"), or set explicitly by a location tag with overrideMode="Deny" or the legacy allowOverride="false
Is says the section is locked at the parent level, but I was able to click enable for the site's logging in the GUI and did not get any errors.
As I wrote the clone of IIS Manager, I can tell you the "Enable/Disable" button links to an attribute called "dontLog", https://github.com/jexuswebserver/JexusManager/blob/master/JexusManager.Features.Logging/LoggingFeature.cs
Its reference can be found at
https://www.iis.net/configreference/system.webserver/httplogging
Related
We have developed a Blazor WebAssembly Application that has already gone into productive usage for a certain group of customers.
The Application works well in all Browsers with Standard Security settings. However, this morning I got a call from one of the customers, where the Application did not load at all in their Chrome Browser.
I saw the following Errors in the console:
Unknown error occurred while trying to verify integrity.
Failed to load resource: the server responded with a status of 403 (forbidden)
Failed to find a valid digest in the 'integrity' attribute for ressource '<somepath.dll>' with SHA-256 integrity <sha56>. the resource has been blocked
Now my question is, what could cause this? Is this a Browser Security setting, or another security setting e.g on server, in code etc.? How can I fix this?
Here's a picture of the errors mentioned above
The most likely reason why this is happening, is that some Antiviruses block the execution of downloaded .dll files. That's why it is working in some networks, but doesn't in some others.
What you can do, and what is also suggested as a Workaround by microsoft, is to rename all .dll to .bin - and also change the config json. it worked in my case.
I use the following PowerShell function for that:
Function Hide-BlazorDLL {
Param(
[string]$Path = (Get-Location).Path
)
<#
According to the following Links:
https://github.com/dotnet/aspnetcore/issues/19552
https://github.com/dotnet/aspnetcore/issues/5477#issuecomment-599148931
https://github.com/dotnet/aspnetcore/issues/21489
https://gist.github.com/Swimburger/774ca2b63bad4a16eb2fa23b47297e71
#>
# Test if path is correct and accessible
$WorkingDir = Join-Path $Path "_framework"
if (!(Test-Path $WorkingDir)) { Throw "Wrong path $Path. current location must be wwwroot folder of published application." }
# Get All Items
$AllItems = Get-ChildItem $WorkingDir -Recurse
$DLLs = $AllItems | Where-Object { $_.Name -like '*.dll*' }
$BINs = $AllItems | Where-Object { $_.Name -like '*.bin*' }
# End script if no .dll are found
if ($DLLs) {
# Delete all current .bin files
if ($BINs) {
Remove-item $BINs.FullName -Force
}
# Change .dll to .bin on files and config
$DLLs | Rename-item -NewName { $_.Name -replace ".dll\b",".bin" }
((Get-Content "$WorkingDir\blazor.boot.json" -Raw) -replace '.dll"','.bin"') | Set-Content "$WorkingDir\blazor.boot.json"
# Delete Compressed Blazor files
if (Test-Path "$WorkingDir\blazor.boot.json.gz") {
Remove-Item "$WorkingDir\blazor.boot.json.gz"
}
if (Test-Path "$WorkingDir\blazor.boot.json.br") {
Remove-Item "$WorkingDir\blazor.boot.json.br"
}
# Do the same for ServiceWorker, if it exists
$ServiceWorker = Get-Item "$Path\service-worker-assets.js" -ErrorAction SilentlyContinue
if ($ServiceWorker) {
((Get-Content $ServiceWorker.FullName -Raw) -replace '.dll"','.bin"') | Set-Content $ServiceWorker.FullName
Remove-Item ($ServiceWorker.FullName + ".gz")
Remove-Item ($ServiceWorker.FullName + ".br")
}
}
else {
Write-Host "There are no .dll Files to rename to .bin"
}
}
Basically you need to navigate to the wwwroot folder of your published application and run the function there. e.g:
PS D:\inetpub\wwwroot\<appname>\wwwroot> Hide-BlazorDLL
Solution for me was to delete the obj and the bin folder in both the client and the server project folder
This error for some reason kept happening for me when I tested my application in an anonymous browser window (Google Chrome).
Try using a normal browser window if you're getting integrity errors.
Also, if you're using Cloudflare CDN don't forget to "Purge Everything" in the cache.
We have experienced this issue using Cloudflare auto minify feature. That feature removes any comments from html, js and other files - which some of the blazor .js files seems to contain.
This means that the hash of the file contents no longer matches the hash found in blazor.boot.json -> an integrity issue will be thrown and stop the app from loading.
Disabling the auto minify feature fixed the issue.
Tech Stack:
.NET 6.0.11
I had a similar issue. In the local machine, it is working fine. But when it is deployed through GitHub Actions, I get integrity checks error. I got this issue for Blazor WebAssembly ASP.NET Core Hosted (WebAssemblyPrerendered) project. Here is the fix I followed.
Added the .gitattributes file to the solution root folder.
Added the below code at the end of the file.
# blazor dlls - treat all .dll files as binary
*.dll binary
I have a VM template that deploys a DSC extension. It's been working fine but my configuration is growing so I've refactored it to use a composite resource and republished it to blob storage with Publish-AzureRmVMDscConfiguration.
I verified that the .ps1.zip file in blob storage contains my custom module and that the module is listed under dscmetadata.json. However, when I deploy, the DSC extension fails. The logs under C:\WindowsAzure\Logs\Plugins\Microsoft.Powershell.DSC\2.17.0.0 reveal the reason:
C:\Packages\Plugins\Microsoft.Powershell.DSC\2.17.0.0\bin..\DSCWork\AppServerDev.ps1.0\AppserverDev.ps1
PSDesiredStateConfiguration\node : The module 'xCustomResource' could
not be loaded. For more information, run 'Import-Module
xCustomResource'.
At C:\Packages\Plugins\Microsoft.Powershell.DSC\2.17.0.0\DSCWork\AppServerDev.ps1.0\AppserverDev.ps1:9 char:3
node "localhost"
~~~~
CategoryInfo : ObjectNotFound: (xCustomResource\xCustomResource:String)
[PSDesiredStateConfiguration\node], ParentContainsErrorRecordException
FullyQualifiedErrorId : CouldNotAutoLoadModule,PSDesiredStateConfiguration\node
Here are the relevant bits of my configuration file:
configuration AppServerDev
{
param($environment)
Import-DscResource -ModuleName 'PSDesiredStateConfiguration'
Import-DscResource -ModuleName 'SaaSModule'
node "localhost"
{
LocalConfigurationManager
{
RebootNodeIfNeeded = $true
ConfigurationMode = "ApplyAndAutoCorrect"
ConfigurationModeFrequencyMins = 1440
}
xDCTPlatformVM VM {
OctopusParametersFile = $environment
ChocolateyPackages = #(
'googlechrome',
'notepadplusplus',
'7zip',
'microsoftwse',
'octopusdeploy.tentacle',
'sqlserver-cmdlineutils'
)
}
}
}
Running Get-Module -ListAvailable reveals that DankModule is found and I can, from powershell, run Import-Module DankModule and it works as expected. I assume the confusion is coming from it trying to import xCustomResource rather than DankModule but my .ps1 file under C:\Packages\Plugins\Microsoft.Powershell.DSC\2.17.0.0\DSCWork says
Import-Module "DankModule"
and not
Import-Module "xCustomType"
Why is it trying to import xCustomType rather than DankModule? How do I make it find DankModule which is available and contains xCustomType?
UPDATE: Get-Module -ListAvailable shows DankModule is installed but Get-DSCResource -Module DankModule doesn't return anything.
I'm not sure but I think this is a rookie mistake on my part because I have no experience building powershell modules. I was using this page as well as this one to construct my module and I'd gotten a valid module file by using New-ModuleManifest, I had the correct file structure mentioned in both of them e.g.
C:\Program Files\WindowsPowerShell\Modules\
DankModule
DankModule.psd1
DankModule.psm1
DSCResources
xCustomResource
xCustomResource.psd1
RootModule = ‘xCustomResource.schema.psm1'
xCustomResource.schema.psm1
Configuration, no Node block
but I didn't realize I had some tweaking left. I needed to correct the RootModule property in DankModule.psd1 to look like this:
RootModule = 'C:\Program Files\WindowsPowerShell\Modules\DankModule\DankModule.psm1'
And also needed to add to my blank DankModule.psm1 a line referencing my xCustomResource.schema.psm1 equivalent:
. .\DSCResources\xCustomResource\xCustomResource.schema.psm1
That is the only line in my module-level .psm1 file and everything is now working. I also, at one point, removed all the \0s from both my .psd1 files but now I'm not sure if that was strictly necessary.
I have a vdir with a physicalPath that points to a file share. Setting the "userName" and "password" attributes to a user with access to that share in the vdir config section in applicationhost.config works fine, but the password is then in clear text.
I want this password encrypted in the applicationhost.config file.
The only option I've found is to use "appcmd.exe" to do this, but I am unable to get it to work in the obvious way:
appcmd.exe set vdir /vdir.name:"Default Web Site/img" /password:"ThePassword"
I get this error:
ERROR ( hresult:80090016, message:Failed to commit configuration changes.
Keyset does not exist )
I have tried with an elevated cmd prompt, does nothing. Setting the other attributes, like userName, works fine.
Try it like this:
appcmd.exe set vdir "Default Web Site/img" /username:user /password:password
I would like to build Nuget package for my add-on which will be used by end user to install as startup task and after that they will upload their applications on window azure platform.
Let's take one simple web application and one cloud project now using Nuget end user will added add-on package it will add 2 files(exe & config) in web application project & add startup task to ServiceDefinition.csdef of cloud project as per shown in figure
How can i created this type of Nuget Pacakge?
Thanks in advance.
Update:
I have tried according with NICK's answer however i am getting problem with 2 webrole in cloud project according to shown in below fig.
and error i am getting as per below
Also i have one question that If i am installing that Nuget package with command line then how i can consider all webrole projects to add exe and config file in solution??
Nuget packages work based on a convention:
http://docs.nuget.org/docs/creating-packages/creating-and-publishing-a-package#From_a_convention_based_working_directory
As far as the exe and config you can do the following:
In your package directory make the following directories
mkdir lib (for the exe)
mkdir content (for the config)
All you have to do for the exe is drop the file in the lib directory and modify you .nuspec file just under the metadata node. There should be a "files" node (if not you can add one). Add something like this inside the files node:
<file src="content\my.exe" target="content\my.exe" />
The config is a little different. Just add a file named myname.config.transform to the content directory and add an entry into the .nuspec file:
A couple of things to note:
If the config file does not exist in your app it will add one for you.
You only have to add the nodes you want transformed if there is a file that already exists
The transform file will do a complete match on your node so if the following existed in
your config file:
<add key="test" value="myval"/>
And in your transform you had:
<add key="test" value="myval2"/>
The resulting file would look like:
<add key="test" value="myval"/>
<add key="test" value="myval2"/>
As far as adding the startup task, that's been a little more tricky for me (there might be a much better way). I use powershell in the install.ps1 (just like the files above but you create a "tools" directory for it):
param($installPath, $toolsPath, $package, $project)
#Modify the service config - adding a new Startup task
$svcConfigFile = $DTE.Solution.Projects|Select-Object -Expand ProjectItems|Where-Object{$_.Name -eq 'ServiceDefinition.csdef'}
$ServiceDefinitionConfig = $svcConfigFile.Properties.Item("FullPath").Value
[xml] $xml = gc $ServiceDefinitionConfig
#Create startup and task nodes
# So that you dont get the blank ns in your node
$startupNode = $xml.CreateElement('Startup','http://schemas.microsoft.com/ServiceHosting/2008/10/ServiceDefinition')
$taskNode = $xml.CreateElement('Task','http://schemas.microsoft.com/ServiceHosting/2008/10/ServiceDefinition')
$taskNode.SetAttribute('commandLine','my.exe')
$taskNode.SetAttribute('executionContext','elevated')
$taskNode.SetAttribute('taskType','simple')
$startupNode.AppendChild($taskNode)
#Check to see if the startup node exists
$modified = $xml.ServiceDefinition.WebRole.StartUp
if($modified -eq $null){
$modified = $xml.ServiceDefinition.WebRole
$modified.PrependChild($startupNode)
}
else{
$nodeExists = $false
foreach ($i in $xml.ServiceDefinition.WebRole.Startup.Task){
if ($i.commandLine -eq 'my.exe'){
$nodeExists = $true
}
}
if($taskNode -eq $null -and !$nodeExists){
$modified.AppendChild($taskNode)
}
}
$xml.Save($ServiceDefinitionConfig);
I hope this helps out.
--Nick
I have written a script that inserts some test data into a document library. I intend to use it as a post-deployment step in Visual Studio 2010, so that the library is not empty after a retract & deploy.
The relevant portions of the script are:
Install.ps1:
$scriptDirectory = Split-Path -Path $script:MyInvocation.MyCommand.Path -Parent
. "$scriptDirectory\Include.ps1"
$webUrl = "http://localhost/the_site_name"
$web = Get-SPWeb($webUrl)
...
Include.ps1:
function global:Get-SPSite($url)
{
return new-Object Microsoft.SharePoint.SPSite($url)
}
function global:Get-SPWeb($url,$site)
{
if($site -ne $null -and $url -ne $null){"Url OR Site can be given"; return}
#if SPSite is not given, we have to get it...
if($site -eq $null){
$site = Get-SPSite($url);
...
}
It works fine when run as follows from the command line, even immediately after a Visual Studio re-deploy:
powershell \source\ProjectFiles\TestData\Install.ps1
However, it does not work when I use the exact same command as a post-deployment command line in the SharePoint project's properties in Visual Studio:
Run Post-Deployment Command:
New-Object : Exception calling ".ctor" with "1" argument(s): "The Web applicati
on at http://localhost/the_site_name could not be found. Verify that you have t
yped the URL correctly. If the URL should be serving existing content, the syst
em administrator may need to add a new request URL mapping to the intended appl
ication."
At C:\source\ProjectFiles\TestData\Include.ps1:15 char:18
+ return new-Object <<<< Microsoft.SharePoint.SPSite($url)
+ CategoryInfo : InvalidOperation: (:) [New-Object], MethodInvoca
tionException
+ FullyQualifiedErrorId : ConstructorInvokedThrowException,Microsoft.Power
Shell.Commands.NewObjectCommand
Interestingly, I can reproduce the error on the command line if I run:
c:\windows\Syswow64\WindowsPowerShell\v1.0\powershell \source\ProjectFiles\TestData\Install.ps1
However, the post-deployment command fails even if I explicitly run \windows\System32\WindowsPowerShell\v1.0\powershell and \windows\Syswow64\WindowsPowerShell\v1.0\powershell.
Update: Solution found
I seem to be having a similar problem to the one discussed here:
http://social.technet.microsoft.com/Forums/en-US/sharepoint2010programming/thread/faa25866-330b-4e60-8eee-bd72dc9fa5be
I cannot access a 64-bit SharePoint API using 32-bit clients. Because Visual Studio is 32-bit, the post-deployment action will run in a 32-bit process and will fail. There is, however, a 64-bit MSBuild. If we let it run the PowerShell script, all is fine.
Wrap the script in an MSBuild file such as this:
<?xml version="1.0" encoding="utf-8"?>
<Project DefaultTargets="Install" xmlns="http://schemas.microsoft.com/developer/msbuild/2003">
<Target Name="Install">
<Exec Command="powershell .\Install" />
</Target>
</Project>
Then, set the post-deployment command line to:
%WinDir%\Microsoft.NET\Framework64\v4.0.30319\MSBuild $(SolutionDir)\ProjectFiles\TestData\Install.msbuild
Use
%WINDIR%\SysNative\WindowsPowerShell\v1.0\powershell.exe
It’s important that you use the virtual path of %WINDIR%\SysNative and not the actual
path of C:\Windows\System32. The reason for this is that Visual Studio 2010 is a 32-bit
application that needs to call the 64-bit version of powershell.exe to successfully load the
Microsoft.SharePoint.Powershell snap-in.
(c)"Inside Microsoft SharePoint 2010", Microsoft Press, Mar 2011
I had same situation, I needed the Post Deployment powershell script to create dummy data for lists on my local instance. I tried several other ways even using the MSBuild with the .msbuild file as suggested above, but i could not all the variables and had to hard code the file with path and url, this is not what i wanted.
I finally figured out a way to explicitly calling the 64-Bit powershell.exe
I know the 64-bit file has to be there on hard dirve. I know that WinSXS folder has all the files. So quick search for powershell.exe in C:\Windows\winsxs folder i got two files so i grabbed the path for one in amd64 folder.
This is what i have as command in post deployment option
C:\Windows\winsxs\amd64_microsoft-windows-powershell-exe_31bf3856ad364e35_6.1.7600.16385_none_c50af05b1be3aa2b\powershell.exe -command "&{$(ProjectDir)PowerShell\dataload.ps1 -xmlPath "$(ProjectDir)PowerShell\dataload.xml" -webUrl "$(SharePointSiteUrl)"}"
I hope this will help someone in future.
Visual Studio is a 32-bit application, so in 64-bit Windows it runs in a simulated 32-bit environment.
Strangely, the 32-bit environment is called "WoW64" (when 32-bit Windows did this for 16-bit apps, it was called "WoW16". The "WoW" part means "Windows on Windows".
It's similarly strange that "System32" didn't become "System64" with 64-bit Windows. The "32" is from the 16-bit -> 32-bit transition, to differentiate from "System". Whatever, that's legacy/compatibility for you.
In WoW64, everything looks like a 32-bit Windows.
For example, c:\windows\system32 just points to c:\windows\syswow64. 32-bit applications can't (easily) reach anything 64-bit.
It is possible to use PowerShell Remoting to get a 64-bit PowerShell session from a 32-bit environment.
PS>gci env:PROCESSOR_ARCH*
Name Value
---- -----
PROCESSOR_ARCHITECTURE x86
PROCESSOR_ARCHITEW6432 AMD64
PS>Invoke-Command -ConfigurationName Microsoft.PowerShell -ComputerName LOCALHOST { gci env:PROCESSOR_ARCH* }
Name Value PSComputerName
---- ----- --------------
PROCESSOR_ARCHITECTURE AMD64 localhost
I have success doing this as a post deployment command:
%comspec% /c powershell -File "c:\foo\bar.ps1"