PR Pipeline - Part 1#4308
Conversation
|
Is there any chance we could also fix both the PR and the main pipeline to respect the
|
|
@edwardneal - you mean the PlatformSpecific attributes aren't being respected? |
# Conflicts: # eng/pipelines/common/steps/restore-dotnet-tools.yml
Leaving the unused build step for now just in case we need it later
…bles to OneBranch variables file
Move a couple things around WIP for manual test configuration generation
…TestExecuteReaderAsyncWithLargeQuery
…s_SqlAuth as flay
Moving the SA password retrieval deeper into the manual test job
cda511d to
3afd216
Compare
| @@ -4,11 +4,9 @@ | |||
| # See the LICENSE file in the project root for more information. # | |||
There was a problem hiding this comment.
These variables were always meant to be common to all pipelines.
| - name: BUILD_OUTPUT | ||
| value: $(REPO_ROOT)/artifacts | ||
|
|
||
| # Directory where downloaded pipeline artifacts (NuGet packages from earlier |
There was a problem hiding this comment.
These were moved to the onebranch-variables file
| ################################################################################# | ||
| # Licensed to the .NET Foundation under one or more agreements. # | ||
| # The .NET Foundation licenses this file to you under the MIT license. # | ||
| # See the LICENSE file in the project root for more information. # | ||
| ################################################################################# | ||
|
|
||
| name: $(DayOfYear)$(Rev:rr) | ||
|
|
||
| variables: | ||
| - template: /eng/pipelines/common/variables/common-variables.yml@self | ||
| - template: /eng/pipelines/pr/variables/pr-variables.yml@self | ||
|
|
There was a problem hiding this comment.
Yes, this will be covered in another PR to operationalize the pipeline.
| pool: | ||
| vmImage: 'ubuntu-latest' | ||
|
|
There was a problem hiding this comment.
This isn't actually a problem - the pack target will fail if the windows binaries aren't built, and I can confirm that the job is passing as well as the output contains a full build.
| "WorkloadIdentityFederationServiceConnectionId": "", | ||
| "KerberosDomainPassword": "", | ||
| "KerberosDomainuser": "", | ||
| "IsManagedInstance": false |
There was a problem hiding this comment.
This is fixed in #4297, but I'll also address it here.
| int numberOfRows = 10; | ||
|
|
||
| // Insert a bunch of rows in to the table. | ||
| // Insert a bunch of rows in to the table. |
There was a problem hiding this comment.
All the following changes were auto "trim whitespace on save" behavior...
| IsSqlVectorSupported && | ||
| CheckVectorFloat16Supported(); | ||
|
|
||
| public static bool IsDebugBuild |
There was a problem hiding this comment.
This enables us to skip tests if build configuration is DEBUG (or not). This is helpful because skipping will show that the test was skipped in test results, while #if DEBUG will omit it from complication and pretend the test never existed.
| public static void ConnectToSQLWithInstanceNameTest() | ||
| { | ||
| SqlConnectionStringBuilder builder = new(DataTestUtility.TCPConnectionString); | ||
| DataTestUtility.ParseDataSource(builder.DataSource, out string hostname, out _, out _); |
There was a problem hiding this comment.
This change was to address an issue where the tcp connection string parsing was brittle and couldn't handle hostnames with instance names or ports properly.
| { | ||
| // Misc constants | ||
| private const int CALLBACK_TIMEOUT = 5000; // milliseconds | ||
| private const int CALLBACK_TIMEOUT = 20000; // milliseconds |
There was a problem hiding this comment.
This was an initial attempt to improve this test on linux, but it didn't make a difference. Test has just been marked flaky.
| public string TCPConnectionStringHGSVBS = null; | ||
| public string TCPConnectionStringNoneVBS = null; | ||
| public string TCPConnectionStringAASSGX = null; | ||
| public bool EnclaveEnabled = false; |
There was a problem hiding this comment.
These changes are also being made in #4297
|
That's correct @benrr101 - |
|
This one has gotten too big and messy with copilot spam after a small change. Recreating the PR. |
Description
As the goal of this sprint, we want to create a new PR pipeline that simplifies all the complexities of the old one, reduces the test coverage to tests that can run simply and quickly, and completes quickly. This PR introduces the basics of that with a few more changes to come.
This puts the total run time at ~40min in its current state. Compared to the current pipeline which has full runtimes of anywhere from 1.5hrs to 3hrs, this is a massive improvement.
Noteworthy notes have been added inline.
Remaining work
Testing