VCPKG curl error/warning

[Galatis-Giuliano]

Recently I've started to get an error in my build pipeline that reads as follows:

warning(0,0): Error : curl failed to return the expected number of exit codes; this can happen if something terminates curl before it has finished. curl exited with 0 which is normally the result code for the last operation, but may be the result of a crash. The command line was curl --create-dirs --retry 3 -L -w "5ec47b8e-6776-4d70-b9b3-ac2a57bc0a1c%{http_code} %{exitcode} %{errormsg} " "https://binarycache.blob.core.windows.net/storageName/0c23ef01e9fa5b132ff81d7bf24e18a31bd0479898ae8bddc77864d8bcb9eb19.zip?*** SECRET " -o "C:\vcpkg\buildtrees\vcpkg-cmake-config\x64-windows.zip" "https://binarycache.blob.core.windows.net/storageName/8273ad8f25c2429941d8f681016e22686adff21f5da64b2da411d8b96d51c06e.zip? SECRET " -o "C:\vcpkg\buildtrees\vcpkg-cmake\x64-windows.zip" "https://binarycache.blob.core.windows.net/storageName/48cd5568e265c3fbfa17672e2e567f5e84422792f39972ec741c8cfc0195ee5d.zip? SECRET ***" -o

It keeps going like this.

I'm using blob storage as the location for my binary cache. My vcpkg.json looks like this:

{
"dependencies": [
"grpc",
"boost-uuid",
"sttp",
"private-package"
],
"overrides": [
{
"name": "boost-uuid",
"version": "1.85.0"
}
]
}

I did some testing and without grpc the error doesn't happen but there may be something I'm missing.

What's strange is that in my managed pool it is an error while in the Microsoft Hosted Agent it is only a warning. The two also seems to provide different image versions at the moment.

Managed Pool Agent image version: 20250120.2.0
Microsoft Hosted Agent image version: 20250127.1.0

I'm using windows-latest for these. I've only recently started to see this error/warning appear. Any help or advice is appreciated.

[vidyasagarnimmagaddi]

Hi @Galatis-Giuliano , Sorry for the delay, Could you please try with recent roll out image ie, 20250291.1 . If still issue, share your error logs or the repro steps. thanks

[Galatis-Giuliano]

Thank you for the reply. I have since changed my container which fixed the issue at the time. I will update here if the issue comes back.

[vidyasagarnimmagaddi]

Thank you for the reply. I have since changed my container which fixed the issue at the time. I will update here if the issue comes back.

Hi @Galatis-Giuliano , Thanks for confirming. Closing this discussion.