Skip to content
This repository has been archived by the owner on Jan 23, 2023. It is now read-only.
/ corefx Public archive

Enable SendReceive_VaryingLengthBuffers_Success on Windows run #27181

Merged
merged 4 commits into from
Feb 17, 2018

Conversation

caesar-chen
Copy link
Contributor

@caesar-chen caesar-chen commented Feb 15, 2018

Have run the test 500 times on local Windows 10 machine. No failure. The previous stack trace #22845 is too old, cannot reflect which code path results in this failure.

Enable the test on Windows to see if it fails in CI. If it fails, I can investigate further.

Related: #23765

@caesar-chen caesar-chen added this to the 2.1.0 milestone Feb 15, 2018
@caesar-chen caesar-chen self-assigned this Feb 15, 2018
@@ -0,0 +1,14 @@
SETLOCAL EnableDelayedExpansion
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Did you mean to add this file?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

No... It's removed in the new commit.


for /L %%x in (1,1,500) do (

mkdir "D:/TestDiag/Run %%x"
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I don't recommend we hard code drive letters in these scripts. Not everyone has a "D:" drive. I don't for example. I have my other drives set to letters like "S:" (enlistments) and "W:" (work stuff)

msbuild /t:rebuildandtest > "D:/TestDiag/Run %%x/output.log"

IF NOT ERRORLEVEL 1 (
rmdir "D:/TestDiag/Run %%x/" /s /q
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

same comment as above about drive letters.

@rmkerr
Copy link
Contributor

rmkerr commented Feb 15, 2018

It looks like the only failures in windows were on 8.1 and not 10. I think it might be worth trying to recreate the situation more closely by running the script in a VM on 8.1.

@davidsh
Copy link
Contributor

davidsh commented Feb 15, 2018

It looks like the only failures in windows were on 8.1 and not 10. I think it might be worth trying to recreate the situation more closely by running the script in a VM on 8.1.

Weren't those Win 8.1 failures a while ago and when we had the WinHTTP implementation for Windows?

Now, we use the managed sockets implementation for both Linux and Windows. So, maybe getting failures on Win 8.1 won't be possible (or the same type of failure) as before.

@rmkerr
Copy link
Contributor

rmkerr commented Feb 15, 2018

I hadn't considered the switch from WinHTTP to managed sockets. Since the code under test has changed, this looks good to me.

@caesar-chen
Copy link
Contributor Author

@dotnet-bot test Outerloop Windows x64 Debug Build
@dotnet-bot test Outerloop Windows x86 Debug Build
@dotnet-bot test Outerloop Windows x64 Release Build
@dotnet-bot test Outerloop Windows x86 Release Build

@caesar-chen
Copy link
Contributor Author

caesar-chen commented Feb 15, 2018

Now, we use the managed sockets implementation for both Linux and Windows.

Yes, I found the related PR here: #26429. If there is no failure reported after this PR get merged, we can consider enable the test and close #23765.

@caesar-chen
Copy link
Contributor Author

Unrelated failure: System.Net.WebSockets.Client.Tests.CancelTest/CloseAsync_Cancel_Success(server: ws://corefx-net.cloudapp.net/WebSocket/EchoWebSocket.ashx)

I can open a new issue and investigate it. (Doesn't look like this issue #22904)

@stephentoub
Copy link
Member

@dotnet-bot test Outerloop Windows x64 Debug Build
@dotnet-bot test Outerloop Windows x86 Debug Build
@dotnet-bot test Outerloop Windows x64 Release Build
@dotnet-bot test Outerloop Windows x86 Release Build
@dotnet-bot test OSX x64 Debug Build please

@stephentoub
Copy link
Member

@dotnet-bot test Windows x86 Release Build please

@caesar-chen
Copy link
Contributor Author

Disable System.Net.WebSockets.Client.Tests.CancelTest/CloseAsync_Cancel_Success (#27211)

@dotnet-bot test Outerloop Windows x64 Debug Build
@dotnet-bot test Outerloop Windows x86 Debug Build
@dotnet-bot test Outerloop Windows x64 Release Build
@dotnet-bot test Outerloop Windows x86 Release Build

@caesar-chen
Copy link
Contributor Author

caesar-chen commented Feb 16, 2018

Disable System.Net.Http.Functional.Tests.HttpClientHandlerTest/GetAsync_AllowAutoRedirectTrue_RetainsOriginalFragmentIfAppropriate (open #27217 because saw it failed in both of the recent two runs.)

@dotnet-bot test Outerloop Windows x64 Debug Build
@dotnet-bot test Outerloop Windows x86 Debug Build
@dotnet-bot test Outerloop Windows x64 Release Build
@dotnet-bot test Outerloop Windows x86 Release Build

@caesar-chen caesar-chen merged commit d3a69ff into dotnet:master Feb 17, 2018
@rmkerr
Copy link
Contributor

rmkerr commented Feb 17, 2018

Have you gotten a chance to look and see when those tests started failing?

A-And pushed a commit to A-And/corefx that referenced this pull request Feb 21, 2018
…t#27181)

* enable test on Windows run

* delete script

* disable test issue #27211

* disable test see #27217
picenka21 pushed a commit to picenka21/runtime that referenced this pull request Feb 18, 2022
…t/corefx#27181)

* enable test on Windows run

* delete script

* disable test issue dotnet/corefx#27211

* disable test see dotnet/corefx#27217


Commit migrated from dotnet/corefx@d3a69ff
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants