[German]It looks like there are currently issues with Microsoft's package manager Winget. Last week winget-pkgs was no longer able to validate packages or manifests. The problem seems to occur since the change to the 1.4 format and is discussed on Github.
The Windows Package Manager, lso called winget, is a free and open source package manager developed for Windows 10 (from version 1709) and Windows 11. It consists of a command-line service and a set of services for installing applications. Software vendors can use it as a distribution channel for their software packages.
Microsoft says: The winget command-line tool allows users to search, install, update, remove, and configure applications on Windows 10 and Windows 11 computers. This tool is the client interface for the Windows Package Manager service. The Windows Package Manager winget command-line tool is included on Windows 11 and modern versions of Windows 10 as part of App Installer. However, App Installer can also be retrieved from the Microsoft Store. If the tool is already installed, make sure it is updated with the latest version.
There are issues
German blog reader Karel v. G. alerted me via email (thanks for that) that there must be problems with winget-pkgs and wrote:
Hello Mr. Born!
Maybe this is something for your blog?
In the repository for Microsoft's package manager Winget there seems to be a problem with validating new packages/manifests at the moment, since a bot failed and switched to the new format 1.4. Validation errors and ready packages are piling up for moderator approval.
Karel posted a link to this Github discussion where the issue was recently addressed. One user writes:
Is there currently a problem with manifest-validation?
I wonder whether there is currently a problem with the validation-pipeline since about a week (aproximately since manifest 1.4 was officially introduced)?
I notice an increased rate of "Internal-Error-Dynamic-Scan"-batches, msftbot seems to be out of duty, validation-pipeline is reported unhealthy very often, Feedback-batches are not removed automatically after a reply and there seems to be an unusual high amount of open unapproved PRs.
Don't get me wrong, I don't want to criticize anything. I am just curious because of these observations (which can be completely wrong, of course)!
Since the official launch of Manifest 1.4, there seems to be a problem with the validation pipeline. There is an increased rate of "Internal-Error-Dynamic-Scan" batches and msftbot seems to be out of service. The validation pipeline is reported as "unhealthy" very often, feedback batches are not automatically removed after a response, and there seems to be an unusually high number of open, unapproved PRs. There is then confirmation in the thread:
Yes, the validation pipeline is experiencing issues after the validation for
.zipwas enabled and msftbot is also down as of typing this message. To keep up-to-date with such updates, you can follow the WinGet's Gitter Channel
Also within this GitHub entry the problem with winget and the bot is addressed. A developer confirms that they have encountered problems with the validation infrastructure and are working on it with the provider. In addition, there is a simultaneous problem with the bot that helps assign labels and merge PRs. Currently, Microsofters are manually triggering builds to increase throughput and applying labels, writes the developer, who goes by the name denelon.
Addendum: Reader Andreas E. just pointed out to me on Facebook that winget has to replace the "Store for Business" by the end of March. There could still be some problems.
Addendum 2: The colleagues at Bleeping Computer picked up on this and write that an expired SSL certificate is the cause.
Cookies helps to fund this blog: Cookie settings