I'm trying to build out a bunch of stateful microservices written in .NET Core 2.1. These are not ASP.NET applications, no http interface, but rather designed as console apps that are consuming messages off a service bus.
I'm having a deuce of a time trying to figure out how to reference my appsettings.json file into the service packages. There's a bunch of articles related to using the appsettings.json file in an ASP.NET service fabric app (basically spelling it out in a webhost builder) but that's not applicable for me, since I'm not using ASP.NET.
I have a single appsettings.json file that all the microservices in this application should be able to use.
I also found this article Service Fabric include additional files which talks about putting it into an sfproj target for MSBuild, but there seems to be no information whatsoever about what the file structure looks like on the target end.
What I tried to do is add targets to my sfproj file like so:
<Target Name="AfterPackage" AfterTargets="Package">
<Copy SourceFiles="appsettings.json" DestinationFolder="$(PackageLocation)\appsettings.json" />
</Target>
and then in my service, I'd try to instantiate an IConfiguration object based upon ..\appsettings.json but as one might guess by now, the service, upon attempting to publish, throws a FileNotFoundException looking for the appsettings.json file, so clearly the file is in a different location relative to the micro service. (Of course, the publish failure just says there was a problem and rolls back, you have to dig into the the SF Explorer to actually see that message)
What should this configuration look like? I can't find any documentation describing what the file system of a stateful microservice looks like, I'm not clear on where I should be putting this common file, or how the services should be accessing the file. Some help would be appreciated.
I hope I got everything right.
Here are the steps:
Put appsettings.json into shared folder in *.sfproj and include it in project using Visual Studio.
Unload *.sfproj and start editing it as .xml
Add the bellow code right after the last <Target> element.
<Target Name="AfterPackage" AfterTargets="Package">
<Copy SourceFiles="appsettings.json"
DestinationFolder="$(PackageLocation)\%(_DeployableServiceProjectReference.ServiceManifestName)\%(_DeployableServiceProjectReference.CodePackageName)" />
</Target>
The _DeployableServiceProjectReference is the internal <ItemGroup> initialized by the previous targets.
These should force the package command to copy appsettings.json file to service's code package directory for each service references in *.sfproj.
Hope this helps.
Related
I'm trying to find all the locations in which a connectionString can be defined for an iis site (to write a script to extract them all).
I know it can be part of a web.config. I would like to have a complete list of files it can be configured in.
Does it make sense for it to be configured in the site code?
Which other configuration files can define a site's connectionStrings?
And a bonus question - how do I know the order of the files in which the connectionString is searched in ?
Thanks,
EDIT:
Additional info - all IIS sites are pure dotnet sites.
Also, specifying the general location of files, rather then file names, is also helpful.
E.g. - connectionStings can be located in external configuration files, whose location is defined at a in an appSettings element in the "%runtime install path%\config\machine.config" file.
Another option is to just link to the relevant docs.
My issue is that I haven't found anything conclusive.
As far as I know, there are several ways to configure connectionstring in ASP.NET applications.
Define it in code. This is the method used by many beginners. Because at this time they focus on code learning and logical understanding. But some people are accustomed to using it if the database is fixed, it does not need to be modified.
SqlConnection connection = new SqlConnection("Data Source=.\\SQLEXPRESS;Initial Catalog=mytest;Integrated Security=True");
Define it in web.config or App.config. The benefit of it is easy to modified connectionstring after publishing application. Developers can change web.config, no need to change code and deploy application again.
<connectionStrings>
<add name="mytest" connectionString="Data Source=.\SQLEXPRESS;Initial Catalog=mytest;Integrated Security=True;" providerName="System.Data.SqlClient" />
</connectionStrings>
Using External Configuration Files. ConnectionString is stored in a independent file for example connections.config. The benefit of it is modifying an external configuration file does not cause an application restart.
<?xml version='1.0' encoding='utf-8'?>
<configuration>
<connectionStrings configSource="connections.config"/>
</configuration>
About list all connectionstrings, you can use ConnectionStringSettingsCollection. It can get a connection by name and provider name.
I found a pretty good source for the locations of the IIS dotnet Framework configuration files - https://learn.microsoft.com/en-us/previous-versions/windows/it-pro/windows-server-2008-R2-and-2008/cc754617(v=ws.10)?redirectedfrom=MSDN#inheritance
You could define additional connectionStrings
In dotnet code.
In env.
In external config files.
For non-IIS scopes - such as user scopes and role scopes.
These 4 options are not specific to IIS.
I don't know where options (3) and (4) are defined, and I'm not sure if this list is complete. But by combing this list and the one in the doc, I think we have 99% coverage of defined connectionStrings.
I'm trying to deploy my F# ASP.NET Core app to Azure App Service. Unfortunately, each time I try to deploy, a <UserSecretsId> element is added to my project file, which as explained in this article causes the build to fail with the following error:
A function labeled with the 'EntryPointAttribute' attribute must be the last declaration in the last file in the compilation sequence
The article explains why the error occurs and instructs to fix it by removing the element from the project file and instead adding the user secrets ID in an AssemblyInfo.fs. I have tried this and can then build manually, but each time I try to deploy, the deployment process still adds a <UserSecretsId> element with a new ID in my project file, causing the build to fail.
Is there any way I can publish an F# ASP.NET Core app to Azure App Service?
(Also reported on Microsoft/visualfsharp#5549)
Add this to your .fsproj file to suppress the attribute generation.
<PropertyGroup>
<GenerateUserSecretsAttribute>false</GenerateUserSecretsAttribute>
</PropertyGroup>
If you want to use user secrets, you will manually need to add the UserSecrets assembly attribute.
module Your.Namespace.AssemblyInfo
open Microsoft.Extensions.Configuration.UserSecrets
[<assembly: UserSecretsIdAttribute("Your UserSecretsId")>]
do()
Also, this workaround should be unnecessary in ASP.NET Core 2.2. See https://github.com/aspnet/Configuration/pull/872
As far as I know, regarding web-roles, it is possible to have multiple ServiceConfiguration.cscfg files, one for each publish profile (for example, Cloud1, Cloud2, Cloud3, Local and so on).
Though, is it possible to have different diagnostics.wadcfg (wadcfgx) files?
How to instruct Visual Studio to use a specific diagnostics file when deploying to Azure? Is there a specific entry in the publish profile XML file?
Out of the box, it is not possible because there can be only one diagnostics.wadcfg (wadcfgx) file associated with a role in a cloud project a cloud project.
We are using SDK 2.4 in our project and the way we have worked around this problem is by creating separate cloud projects for each environment - Dev, Staging, and Production. Since these are separate cloud projects, we can have different diagnostics configurations.
If you're using SDK 2.5, one thing that works in your favor is the diagnostics is now an extension (instead of plugin). What that allows you to do is change/enable/disable diagnostics on the fly. Though I have not tried it but one thing you could possibly do is enable diagnostics as a part of your deployment process. By default there won't be any diagnostics.wadcfgx file in your project. When you deploy your code via PowerShell, after successful deployment, you can enable this extension. Please see Step 4 here: http://azure.microsoft.com/en-in/documentation/articles/cloud-services-dotnet-diagnostics/#worker-role. (Since I have not tried it myself, I would be very much interested in knowing the outcome. Please do share if this approach worked for you).
I have a web application project in VS2012 which I'm publishing using a "Web Deploy Package". I want this package to include app-pool settings, specifically creating an IIS app-pool and assigning the newly created application to it.
I'm familiar with the option "Include application pool settings used by this Web project" available when the project is configured to use an IIS instance (not IIS Express), but IIS configuration is not part of the project file, and thus not source controlled. What happens when somebody builds a deployment package on a machine that hasn't had IIS meticulously configured? Not ideal.
How else then, can I go about getting AppPool settings into my web deploy package? I understand that the appPoolConfig provider is IIS7+ only, I'm fine with that limitation. I've banged my head against this issue in the past and never found a solution. 18 months later, we've got a new VisualStudio version, and a new web-publishing-pipeline, are there new options to address this? Or maybe something I missed when I first tackled this problem?
Edit
OK, I'm seeing the following as options:
Configure my project to sync settings from an IIS instance. As mentioned, I'm not a fan of this given that it puts settings outside of the project, meaning the environment has to be meticulously configured to build + publish. Plus it drags along other IIS settings I don't want included.
Inject something into the web-publishing-pipeline (WPP) to modify the archive.xml. I've toyed with this in the past and had limited success. One problem is the pipeline isn't exactly co-operative with working directly on the archive.xml file, another problem is some of the more cryptic attributes involved, like MSDeploy.MSDeployProviderOptions which appears to have some Base64 encoded binary? No idea what to put in there.
Find an existing "provider" that can do what I want. I might be out of luck here, the appPoolConfig provider only seems to want to read / write IIS, not, say, an XML file of settings. Does anybody know otherwise?
Write my own "provider" to produce manifest output entries. I'm not sure, is it possible to write a custom provider that writes to a manifest using the name of an existing provider? As in, MyCustomPoolProvider writes appPoolConfig sections into a manifest? This sounds like a potentially painful exercise that may or may not work. Would I still need to figure out the encoding of whatever is going into MSDeploy.MSDeployProviderOptions?
I get the feeling that the fundamental obstacle with Web Deploy for what I'm trying to accomplish, is how strictly it leans on "providers". The pre-existing providers are largely designed for IIS synchronisation, not primary development and publication. It so happens that some of these providers can be relatively easily hooked into via MSBuild, but the majority insist on pulling data from IIS, and that's that.
You are correct in your understanding of the appPoolConfig provider, in that it can only sync between App Pools and can't be provided with the configuration directly. What you could potentially do is keep a copy of the appPool in question in package form (ie. msdeploy -verb:sync -source:appPoolConfig=PoolName -dest:package=apppool.zip) and attempt to hijack the pipeline so that the MSDeploy call adds the application content into the package, leaving the existing content there.
Alternatively, you could always keep the packages separate and deploy them with different calls to MSDeploy.
FYI, MSDeploy.MSDeployProviderOptions is simply an encoded version of the parameters supplied to the provider when it was packaged. For example, -source:dirPath=c:\,ignoreErrors=0x10293847 -dest:package=package.zip would package the ignoreErrors value.
I have been assigned a project with a lot of poorly written code that is based around SharePoint.
It consists of about 15 subprojects, some of them being windows services, some web services, some web applications running inside of SharePoint, some being webparts and even console applications. They all run on the same server and call each other.
There are already many issues in production but they are hard to trace down.
The original developer must have been a fan of either Salinger or Pokémon series judging by his tireless effort to catch all exceptions. Unfortunately, none of them get reported or logged, ever.
My current task is to introduce logging into the whole project so I could find now-invisible exceptions, follow tangled recurring calls and have some stack traces at least. I decided to go with NLog, seeing it's active and cool, as opposed to log4net which is perfectly fine but somewhat not as fancy to my taste.
Because the components are tightly coupled, I want to centralize logging in one file so related errors don't get scattered across the hard drive. Therefore, I am looking to have two or three different log files with five or more projects writing to each of them more or less simultaneously.
What is the best way to configure NLog to centralize logging? Should I have a config file for each project, or should related projects share them? Where should I put config file to log from SharePoint webparts? Am I going to face any permission issues?
I'm using SharePoint 2007.
The easiest way to centralize is probably to simply log to a database, one benefit being that multiple applications and write to the database easier than to the same log file. For each application, configure NLog to log to the Database target, using the same Database target configuration parameters for each. Your NLog.config file might look something like this:
<?xml version="1.0" encoding="utf-8" ?>
<!--
This file needs to be put in the application directory. Make sure to set
'Copy to Output Directory' option in Visual Studio.
-->
<nlog xmlns="http://www.nlog-project.org/schemas/NLog.xsd"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
autoReload="true"
internalLogLevel="Debug"
internalLogFile="nlog_log.log">
<targets async="true">
<target name="sqlexpress" xsi:type="Database">
<connectionString>
Data Source=.\SQLEXPRESS;Initial Catalog=LoggingDB;Integrated Security=True;
</connectionString>
<commandText>
insert into LogTable(DateTime,Logger,LogLevel,Message,ProcessId,ManagedThreadId) values (#DateTime,#Logger,#LogLevel,#Message,#ProcessId,#ManagedThreadId);
</commandText>
<parameter name="#DateTime" layout="${date:format=yyyy\-MM\-dd HH\:mm\:ss.fff}"/>
<parameter name="#Logger" layout="${logger}"/>
<parameter name="#LogLevel" layout="${level}"/>
<parameter name="#Message" layout="${message}"/>
</target>
</target>
</targets>
<rules>
<logger name="*" minlevel="Trace" writeTo="sqlexpress" />
</rules>
</nlog>
You could certainly log to files in addition to (or instead of) logging to a database.
I am not familiar with doing this from SharePoint, so I can't comment on any configuration or permission issues that you might run into there.
Here is a link I found where there is a discussion of getting NLog to work in a SharePoint environment:
http://nlog-forum.1685105.n2.nabble.com/Is-anyone-using-NLog-with-SharePoint-td2171451.html
That link appears to put you at the top of the NLog forum instead of the specific post. Search for this text in the forum "Is anyone using NLog with SharePoint" and you should find the right post.
Good luck!
You could also just leverage the existing logging infrastructure in SharePoint and write to the ULS logs. This way your log information can be viewed in a complete context using the ULS log viewer. For SharePoint 2007 see this blog how to write to the ULS log:
SharePoint Trace Logs and the Unified Logging Service (ULS)
With SharePoint 2010 it has become even easier with improvements to the SPDiagnosticsService class where you can use the new WriteTrace method.
Personally I log Exceptions to the Event Logger. And I use NLog for logging details, debug information or tracing.
Since NLog can be easily switched on and off I only activate it when I'm debugging or when I need to inspect an exception in production. I never was a big fan of the default tracing functionality in .NET.
I prefer simple plain text log files. Although logging to a database works great if you don't have too many "log lines" implemented in your code.
I feel like we are working on the same project! Multiple projects consisting of web projects, core dll projects, console apps, services, etc. Unfortunately I'm not working in sharepoint like you are, but I can describe how I am trying to centralize our logging.
We have 1 core .Net framework project.This is where I placed our wrapper class of the log. This project also holds the nlog dlls and the nlog config file. In this core project file you can add this which automatically moves the config when you build projects with a dependency on this core project.
<None Include="Logging\NLog.config">
<link>NLog.config</link>
<CopyToOutputDirectory>Always</CopyToOutputDirectory>
</None>
We found that some web projects that don't compile a dll don't automatically pull in the config file, so we will leave that up to the build process. This helps centralize our logging so you only have to manage a single config across them all.
In addition remember when you create a logger per class, the log name should have the namespace in it, so you can make specific targets that filter based off namespace if you want different settings for particular projects.
As for centralizing where the logs end up, we chose to use a file target and specify the full path. This is because on our servers the applications run off a C:\ but we have larger D:\ which can store the logs. In our production servers, we also have multiple servers, so we are using splunk to aggregate all of our logs.
If splunk is out of the question and you are on a distributed system, a database sounds like a good idea as suggested above. If you don't want to stand up an sql instance, there are target wrappers for mongo db as well.
Hopefully helpful, I'm curious if anyone has suggestions or opinions on how I'm doing it as well!