STATUS_UNSUCCESSFUL in RsaCryptRoundtrip_OaepSHA1 #29683
Labels
area-System.Security
Known Build Error
Use this to report build issues in the .NET Helix tab
tracking-external-issue
The issue is caused by external problem (e.g. OS) - nothing we can do to fix it directly
Milestone
Comments
|
Correction, 0xC0000001 (STATUS_UNSUCCESSFUL) not an AV. |
danmoseley
changed the title
|
Oh, the joys of status codes. What's wrong? (UNSUCCESSFUL.) Where? Beats me. |
|
Without a stable (or "try this in a loop 30 times") repro or a TTT trace it's hard to bring this up with Windows. Since it's not really actionable right now, marking as Future. |
|
|
We've hit this issue in several NuGet CI test runs. |
|
Configuration: @bartonjs can you please take a look? A dump is available. |
|
@ViktorHofer I don't see a dump... the test simply failed. Looking at history on the message over the past 90 days, it looks like one pops up every week or so (sometimes sooner); but there's no real trend (other than it's private key operations (decrypt, sign)). I think I'm willing to do an experimental retry-once on this error and see if problems go away, now that there's longer to measure it and decide it didn't work or was otherwise a bad idea. |
|
Oh I must have mixed two builds up, you are right, there's no dump :(
You mean retrying that operation with the RetryHelper? |
No. Making the product code call the Win32 function again if it returns STATUS_UNSUCCESSFUL. |
|
Trying that out doesn't sound like a bad idea. |
|
Sounds like we'll be reverting the retry, then, and that there's very little we can do about it. |
maryamariyan
added
the
untriaged
This was referenced May 14, 2020
Closed
|
Again here: https://dev.azure.com/dnceng/public/_build/results?buildId=725147&view=ms.vss-test-web.build-test-results-tab&runId=22467048&resultId=166891&paneView=debug in #39101 |
|
failed again in job: runtime-libraries outerloop 20200713.2 Error message |
|
@bartonjs Given this failure is occurring now, should it still be in milestone "Future"? |
bartonjs
added
the
tracking-external-issue
|
@BruceForstall There's nothing we can do about it, it's a problem in the OS (possibly resource exhaustion?). |
|
failed again in job: runtime-libraries outerloop 20200819.2 failed test: System.Security.Cryptography.X509Certificates.Tests.RevocationTests.DynamicRevocationTests.RevokeRoot(pkiOptions: IssuerRevocationViaOcsp | AllEndEntityRevocation | IssuerAuthorityHasDesignatedOcspResponder | RootAuthorityHasDesignatedOcspResponder) net5.0-Windows_NT-Release-x64-CoreCLR_release-(Windows.Nano.1809.Amd64.Open)windows.10.amd64.serverrs5.open@mcr.microsoft.com/dotnet-buildtools/prereqs:nanoserver-1809-helix-amd64-08e8e40-20200107182504 Error message |
|
failed again in job: runtime-libraries outerloop 20200830.4 net5.0-Windows_NT-Release-x86-CoreCLR_release-Windows.10.Amd64.ServerRS5.Open Error message |
|
failed again in job: runtime-libraries-coreclr outerloop 20200909.3 failed test: System.Security.Cryptography.X509Certificates.Tests.RevocationTests.DynamicRevocationTests.RevokeRoot(pkiOptions: AllRevocation | IssuerAuthorityHasDesignatedOcspResponder | RootAuthorityHasDesignatedOcspResponder) net5.0-Windows_NT-Release-x86-CoreCLR_release-Windows.10.Amd64.ServerRS5.Open Error message |
|
No hit in last 10 days (runfo says 2 hits per month), removing blocking-clean-ci label. |
karelz
removed
the
blocking-clean-ci
|
Failed again in: runtime-libraries-coreclr outerloop 20220705.5 Failed test: Error message: |
|
Failed again in: runtime-libraries-coreclr outerloop 20220713.6 Failed test: Error message: |
|
Failed again in : runtime-libraries-coreclr outerloop 20220731.3 Failed test: Error message: |
|
Failed again in: runtime-libraries-coreclr outerloop 20220801.5 Failed test: Error message: |
|
another example |
vcsjones
added
the
Known Build Error
carlossanlop
added
Known Build Error
|
Another case: |
ghost
added
the
|
We should take another look at this since it does occasionally cause CI to fail. Adding to the 9.0.0 milestone. |
|
I chatted with @bartonjs about this and here are some notes:
The best thing we could do is seek an approach of using an xUnit hook where we could apply an attribute to tests (or test classes) indicating, "if this test fails with an exception that looks like this, retry the test once". Such an approach, if feasible, would likely stop these CI failures. |
|
Failed in: runtime-coreclr libraries-jitstressregs 20240303.1 Failed tests: Error message: Stack trace: |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
and others
Not actionable as is.
https://mc.dot.net/#/user/dotnet-bot/pr~2Fdotnet~2Fcorefx~2Frefs~2Fpull~2F37960~2Fmerge/test~2Ffunctional~2Fcli~2Finnerloop~2F/20190528.20/workItem/System.Security.Cryptography.Cng.Tests/analysis/xunit/System.Security.Cryptography.Rsa.Tests.EncryptDecrypt_Span~2FRsaCryptRoundtrip_OaepSHA1
UPD by @CarnaViire: I went ahead and added Build Information below, from the last occurrence on my PR -- in hopes that known-issues infra would be able to validate and attach the failures not attached yet.
Build Information
Build: https://dev.azure.com/dnceng-public/public/_build/results?buildId=565017&view=results
Build error leg or test failing: System.Security.Cryptography.Rsa.Tests.EncryptDecrypt_Span.RsaCryptRoundtrip_OaepSHA512
Pull request:
Error Message
Fill the error message using step by step known issues guidance.
{ "ErrorMessage": "CryptographicException : Unknown error (0x", "BuildRetry": false, "ExcludeConsoleLog": false }Report
Summary
Known issue validation
Build: 🔎 https://dev.azure.com/dnceng-public/public/_build/results?buildId=565017
Error message validated:
[CryptographicException : Unknown error (0x]Result validation: ✅ Known issue matched with the provided build.
Validation performed at: 2/19/2024 11:07:21 AM UTC
The text was updated successfully, but these errors were encountered: