Commit Graph

665 Commits

Author SHA1 Message Date
Pavel Krymets d22f689fd2 Add support to use System.Memory as a package instead of internilized source (#1821) 2017-05-16 16:54:03 -07:00
David Fowler cf16d601d6 Remove synchronous Write APIs (#1838) 2017-05-15 19:04:04 -07:00
Stephen Halter 41f1922502 Simplify LibuvConnection.OnRead() (#1828)
* Simplify LibuvConnection.OnRead()

- Fix a null reference that sometimes occurs given an EOF status
2017-05-15 11:02:16 -07:00
Kiran Challa 01b8d5fad1 Skip flaky temporarily to unblock the mirror 2017-05-13 20:41:10 -07:00
Stephen Halter e149852d62 Skip flaky tests on TeamCity agents (#1832) 2017-05-12 14:36:47 -07:00
Cesar Blum Silveira 37f15bdd85 Create Frame in FrameConnection. 2017-05-10 16:48:37 -07:00
David Fowler b9518e3684 Invert the dependency between connection adapters and Frame (#1822)
* Invert the dependency between connection adapters and Frame
- Removed PrepareRequest from IAdaptedConnection and instead added
a feature collection to the ConnectionAdapterContext. This allows features to be set
once by the adapter instead of per request. It's the Frame's job to copy features
from the connection level feature collection into the per request feature collection.
- Set the scheme to "https" based on the presence of ITlsConnectionFeature.
- Always set ITlsConnection feature if the HttpsAdaptedConnection doesn't throw during
the handshake
2017-05-10 15:29:43 -07:00
David Fowler c8b6a2be56 More FrameConnection refactoring (#1820)
* More FrameConnection refactoring
- This change reverts the change to complete the writer with an
exception on abort because of the number of first chance exceptions
that get thrown.
- This change also moves connection logging into FrameConnection instead
of being split between the ConnectionHandler and FrameConnection.
- Fixed issues with LibuvOutputConsumerTests that leak WriteReq since
cancelled writes no longer end the connection.
2017-05-09 17:40:25 -07:00
David Fowler c48113ad80 Refactoring and of FrameConnection and Frame (#1816)
* Refactoring and of FrameConnection and Frame
- Building on top of the last refactoring of FrameConnection, this change aims to clean up
the communication between the Frame and FrameConnection by removing some concepts and
being consistent about the communication between Frame and FrameConnection with or without
connection adapters. Changes include:
- Removing ConnectionLifetimeControl, ISocketOutput, StreamSocketOutput
- Moving more initialization of the frame to FrameConnection after the pipes
are setup
- OutputProducer communicates cancellation via the IPipeWriter instead of the output's IPipeReader.
- Frame always communicates via the pipes and that communications flows through the layers to the transport.
This means that each 1/2 of the adapted pipeline handles closing the right side of the transport at the
right time, propagating exceptions as necessary.
- This is how the flow looks now:
            ->                        ->
[transport]     [connection adapters]     [frame]
            <-                        <-
- Transports need to handle a ConnectionAbortedException on the output as a signal to stop
writing and end the connection. This will no longer try to drain the output but will just stop
writing and end the response immediately.
- Remove frame.Abort when cancellation on Write fails.
- Unify the connection shutdown logic
- Dispose 1/2 initialized connection adapters

#1815
2017-05-08 20:44:13 -07:00
Stephen Halter 6e2fdda162 Simplify connection lifetime control flow (#1776)
* Also make IAdaptedConnection disposable
2017-05-05 16:11:01 -07:00
Pavel Krymets 28b479c99a Merge branch 'rel/2.0.0-preview1' into dev 2017-05-05 14:35:41 -07:00
Pavel Krymets a9c165e666 netcoreapp2.0 (#1810) 2017-05-05 14:31:34 -07:00
David Fowler cd1568d7f4 Remove uv_shutdown because FIN is already sent during uv_close (#1811)
* Remove uv_shutdown because FIN is already sent during uv_close

#1808
2017-05-05 12:52:10 -07:00
Stephen Halter 557cf29e4a arespr/knownmethods-optimizations cleanup 2017-05-04 12:27:22 -07:00
Stephen Halter f464760bf8 Merge branch 'arespr/knownmethods-optimizations' into dev 2017-05-04 11:44:54 -07:00
David Fowler b21f84543a Use WebHostBuilder as an implementation detail of TestServer (#1804)
- This will make tests run more similarly so we can swap out the
transport more easily.
2017-05-03 19:35:09 -07:00
Stephen Halter 0afb874555 Fix flaky tests that bind to discovered IP addresses (#1795) 2017-05-01 12:20:53 -07:00
Stephen Halter 6a3d6d4273 Fix flaky tests that bind to discovered IP addresses (#1795) 2017-05-01 11:41:10 -07:00
Nate McMaster 4dc7946cd8 Implement new request trace identifier format
The format:
The trace identifier begins with connection ID and ends with a number that increments with each request per connection.

Example:
Connection ID = xyz
Request 1 = "xyz:00000001"
Request 2 = "xyz:00000002"
...
Request 15 = "xyz:0000000F"
Request 16 = "xyz:00000010"
2017-05-01 09:50:47 -07:00
Stephen Halter 7122b6c4aa Fix flaky heartbeat test (#1794) 2017-04-29 00:42:27 -07:00
David Fowler 749e282102 Prepare for OnReader/WriterCallbacks changes (#1791)
- This change does a few things:

1. It adds the events we will replace with
pipe events to IConnectionContext and IConnectionInformation to get out of
band notifications about pipe completions.

2. It also implements those callbacks
and exposing slight changes we'll need to make once we have them. The idea is
that we can delete/replace these methods once we have the new pipe API and things
will keep working.
2017-04-29 00:41:48 -07:00
Cesar Blum Silveira 3b2d0a52f3 Merge branch 'rel/2.0.0-preview1' into dev 2017-04-28 10:00:17 -07:00
Cesar Blum Silveira 9185deecda Avoid port conflicts in AddressRegistrationTests (#1721). 2017-04-28 09:56:29 -07:00
David Fowler 5b62024fc8 * Properly handle FINs and resets in the SocketConnection (#1782)
- FIN from the client shouldn't throw
- Forced close from the server should throw
- Properly wrap connection reset exceptions and other exceptions
in IO exceptions
- This gives kestrel control over when the output closes
- Fixed one test that assumed libuv
- Dispose the connection to yield the reader

Fixes #1774
2017-04-27 17:55:35 -07:00
Stephen Halter 79ea2bb9b3 Quick fix for SslStream ODEs in HttpsAdaptedConnection.PrepareRequest (#1786) 2017-04-27 16:56:07 -07:00
David Fowler 0755495ce5 Remove Logging dependency from the http parser (#1780)
* Remove Logging dependency from the http parser
2017-04-27 08:46:08 -07:00
Nate McMaster c2f15fcac3 Move user-facing strings into resource files 2017-04-26 17:22:55 -07:00
Nate McMaster 9464003bda Add wait in ListenerPrimaryTests to avoid race condition with List.Add 2017-04-26 11:03:28 -07:00
Nate McMaster eb1301f28d Skip HostName binding test when network is unreachable 2017-04-25 14:33:06 -07:00
Pranav K 5b976a9fa3 Use Bundled NETStandard.Library \ NETCoreApp versions instead of explicitly specifying one 2017-04-25 12:51:48 -07:00
Stephen Halter a4def946a6 Fix flaky 400 response tests (#1758) 2017-04-24 23:48:47 -07:00
Pavel Krymets e3e78bc461 Fix LibuvOutputConsumerTests (#1757)
* Fix LibuvOutputConsumerTests
2017-04-24 17:37:30 -07:00
David Fowler 9e80fb65bd Removed the wrote after disconnect log (#1747)
- It's not very useful
2017-04-24 17:34:56 -07:00
Stephen Halter f26c31c116 Fix scope id test failures on mac and linux (#1754) 2017-04-24 17:15:41 -07:00
Nate McMaster 39047638cc Add debug logging in address binding when IPv6Any fails 2017-04-24 17:14:43 -07:00
Nate McMaster 089ff49643 Put hostname address registration into separate test 2017-04-24 17:14:43 -07:00
David Fowler a98581670e Schedule connection adapter reads on configured thread pool (#1741)
* Schedule connection adapter reads on configured thread pool
- This should speed up connection adapters as we don't block
new reads. It *might* also help some of the test flakiness
2017-04-24 12:19:22 -07:00
Pavel Krymets df9e48b1f0 Add cancellation support for write async (#1736) 2017-04-24 11:58:08 -07:00
David Fowler a749939be4 Clean up libuv connection (#1726)
* Clean up libuv connection
- Cancel all pending flushes on the input writer before
disposing the stream handle.
- Complete the pipe before disposing the socket
- Added logging for connection pause/resume.
- Added test
2017-04-21 13:13:05 -07:00
David Fowler dcea15dba7 Skipped newly failing AddressRegistration tests failing tests on OSX and Linux (#1732)
* Unblock CI
2017-04-21 01:54:44 -07:00
Stephen Halter a14eabce9a RequestTests and ResponseTests cleanup (#1723) 2017-04-20 16:33:05 -07:00
Nate McMaster ee9feedc27 Improve implementation of IHttpUpgradeFeature
After upgrade has been accepted by the server:
 - Reads to HttpRequest.Body always return 0
 - Writes to HttpResponse.Body always throw
 - The only valid way to communicate is to use the stream returned by IHttpUpgradeFeature.UpgradeAsync()

Also, Kestrel returns HTTP 400 if requests attempt to send a request body along with Connection: Upgrade
2017-04-20 12:42:58 -07:00
Stephen Halter bebba2a113 Make UnrootedConnectionsGetRemovedFromHeartbeat test less flaky (#1727) 2017-04-20 10:58:48 -07:00
David Fowler d0743806d8 Disable UnrootedConnectionsGetRemovedFromHeartbeat because it's flaky 2017-04-20 10:21:20 -07:00
David Fowler d1494224f4 Make the tests that use the Heartbeat more deterministic. (#1724)
- Added Start to Heartbeat and made OnHeartbeat internal
- Explicitly call start in tests
2017-04-20 01:41:37 -07:00
Stephen Halter a053ca4758 Move shutdown logic from transport to core (#1707)
* Move shutdown logic from transport to core
* Use weak references to track FrameConnections
2017-04-19 18:08:36 -07:00
David Fowler d40dbb81ea Fixed broken tests that leak UvWriteReq 2017-04-19 17:50:43 -07:00
Nate McMaster 7a3a731686 Refactor address binding and handle EAFNOSUPPORT
- Simplify KestrelServer by refactoring address binding into a separate class
 - Use strategy pattern to implement address binding for different sceanrios
 - Add fallback from binding 0.0.0.0 if binding to [::] fails (can happen if UvException with EAFNOSUPPORT is thrown)
2017-04-19 15:58:12 -07:00
David Fowler 42d82a507d Make all UvRequest objects normal GC handles (#1698)
* Make all UvRequest objects normal GC handles
- This avoids the cost of using GCHandle.Alloc per operation.
- It *does* mean that we need to explicitly dispose UvRequest objects
after using them (which we did before anyways). This change does
add a few try catch statements to make sure we always dispose the UvRequest
if there are synchronous exceptions.
- This is ~1.5% of the overhead in the benchmarks today
- Keep track of all allocated UvRequest objects with a WeakReference in DEBUG
and assert none are kept around after cleaning up.
- Fixed a leak where we don't clean up UvWriteReq objects when writing
to the named pipe.
2017-04-19 15:16:11 -07:00
Stephen Halter e4ba1d01ce Ensure that KeepAliveTimeoutTests do not run with sync context (#1718)
- With Task.Delay(1).ConfigureAwait(false) there's a race where
  Task.IsCompleted could already be true by the time it's evaluated.
2017-04-18 17:20:59 -07:00