Commit Graph

245 Commits

Author SHA1 Message Date
bgallagher
c149f4bcc0 Remove support from deregister a Channel from a EventLoop manually 2013-08-29 18:11:16 +02:00
Norman Maurer
9e7529b2f5 [#1743] Fix IllegalStateException by remove usage of PendingWrite in ChunkedWriteHandler. This needs more thoughts before re-introduce it 2013-08-16 08:11:19 +02:00
Norman Maurer
48eb73f9b1 [#1733] Correctly pass the promise to the ctx.write(...) in AbstractTrafficShapingHandler 2013-08-13 08:21:26 +02:00
Norman Maurer
f8f19d1539 [#1732] Fix bug which lead ChannelTrafficShapingHandler stop to work after first Channel was disconnected 2013-08-13 08:09:59 +02:00
Norman Maurer
8af5ea8d4f [#1658] Allow to use AbstractTrafficShapingHandler also with ByteBufHolder based messages 2013-08-10 20:25:35 +02:00
Norman Maurer
c1d787bff7 [#1722] ChunkedWriteHandler.doFlush(...) should not call fireExceptionCaught(..) 2013-08-09 21:32:17 +02:00
Norman Maurer
72395bf400 [#1719] Make sure PendingWrite is only recycled once in ChunkedWriteHandler 2013-08-09 18:44:58 +02:00
Norman Maurer
56646920c0 Formatting 2013-07-29 20:59:45 +02:00
Norman Maurer
9da81acec1 [#1657] Log content of ByteBufHolder 2013-07-29 20:57:33 +02:00
Norman Maurer
00f1533fa9 [#1671] Fix bug in ChunkedWriteHandler which produce a NPE on empty chunks 2013-07-29 16:23:43 +02:00
Norman Maurer
0cb3541b53 [#1669] Correctly notify the ChannelPromise of delayed writes 2013-07-29 08:13:29 +02:00
Norman Maurer
5e703e7419 [#1658] Let trafficshaping work with ByteBuf and ByteBufHolder out of the box 2013-07-29 07:31:09 +02:00
Frédéric Brégier
61b1214b24 Fix for first issue from #1652 on computation of time to wait in AbstractTrafficShapingHandler for Netty 4
Fix for first issue from #1652 on computation of time to wait in AbstractTrafficShapingHandler for Netty 4, using the same formula than in Netty 3 (wrong place for parenthese).
Was:

    (bytes * 1000 / limit - interval / 10) * 10;

Becomes:

    (bytes * 1000 / limit - interval) / 10 * 10;
2013-07-26 15:24:31 +02:00
Norman Maurer
2b3ac3d446 Factor out the PendingWrite class and put it in internal package. Make use of it in SslHandler and ChunkedWriteHandler to reduce GC-pressure 2013-07-25 12:36:24 +02:00
Norman Maurer
fa4e15e198 Make PendingWrites recyclable to reduce GC pressure 2013-07-20 18:15:35 +02:00
Trustin Lee
4f6ba4fe3d Use tryProgress just in case of user error 2013-07-19 13:30:33 +09:00
Trustin Lee
762adfcb69 Update HttpStaticFileServer example / Fix bugs related with progress notification
- Fix a bug in DefaultProgressivePromise.tryProgress() where the notification is dropped
 - Fix a bug in AbstractChannel.calculateMessageSize() where FileRegion is not counted
 - HttpStaticFileServer example now uses zero copy file transfer if possible.
2013-07-19 13:21:32 +09:00
Trustin Lee
f96a8e5951 Implement ProgressivePromise notification in NIO byte channels and ChunkedWriteHandler
- Refine the contract of GenericProgressiveFutureListener.
- Negative 'total' now means 'unknown', which is useful for ChunkedWriteHandler.
2013-07-19 12:53:23 +09:00
Trustin Lee
6791984146 Remove unnecessary code from SslHandler
- Remove CloseNotifyListener which was used only to reduce the noisy logging.
- Instead, simply do a string match.
- Fixes #1608
2013-07-19 09:24:19 +09:00
Norman Maurer
def3dbe035 Add missing closing paren 2013-07-16 07:39:33 +02:00
Norman Maurer
a215ba6ef6 Some javadocs 2013-07-12 15:45:09 +02:00
Norman Maurer
023201f684 [#1570] ChunkedWriteHandler.discard(..) must not call fireExceptionCaught(...) 2013-07-12 10:58:38 +02:00
Norman Maurer
1a7d1f7023 [#1568] Fix problem where an exception was fired after the channel was closed.
* Calling fireExceptionCaught(...) in this case was not correct as we failed writes which are outbound operations.
2013-07-12 10:42:39 +02:00
Jeff Pinner
be963d4caf rewrite SslHandler's flush0 message to match docs 2013-07-12 07:18:47 +02:00
Jeff Pinner
c77ab7d092 Fix a NoSuchElementException and out-of-order event problem caused by SslHandler
The fix prevents from reentering channelRead incorrectly. It also
prevents from getting the inbound requests out of order.
2013-07-12 09:51:28 +09:00
Norman Maurer
d8479a04df Allocate buffer from ByteBufAllocator in ChunkedFile 2013-07-12 00:02:54 +02:00
Norman Maurer
64686deaff Tiny optimization 2013-07-11 15:47:29 +02:00
Norman Maurer
4d94f1d1e1 [#1534] Finally fix SslHandler to also correctly handle partial data 2013-07-11 06:40:04 +02:00
Norman Maurer
b57d9f307f Allow per-write promises and disallow promises on flush()
- write() now accepts a ChannelPromise and returns ChannelFuture as most
  users expected.  It makes the user's life much easier because it is
  now much easier to get notified when a specific message has been
  written.
- flush() does not create a ChannelPromise nor returns ChannelFuture.
  It is now similar to what read() looks like.
2013-07-11 00:49:48 +09:00
Norman Maurer
da5c6add14 Rename CodecOutput to RecyclableArrayList and move it to internal package.
* Also reuse it in SslHandler
2013-07-10 07:50:26 +02:00
Trustin Lee
26e9d70457 Remove channelReadSuspended / Rename messageReceived(Last) to channelRead(Complete)
- Remove channelReadSuspended because it's actually same with messageReceivedLast
- Rename messageReceived to channelRead
- Rename messageReceivedLast to channelReadComplete

We renamed messageReceivedLast to channelReadComplete because it
reflects what it really is for.  Also, we renamed messageReceived to
channelRead for consistency in method names.
2013-07-09 23:58:51 +09:00
Norman Maurer
354de0a8f0 [#1534] Fix handling of partial frames in SslHandler
* Let SslHandler not extend ByteToMessageDecoder
2013-07-09 23:58:50 +09:00
Trustin Lee
cbd8817905 Remove MessageList from public API and change ChannelInbound/OutboundHandler accordingly
I must admit MesageList was pain in the ass.  Instead of forcing a
handler always loop over the list of messages, this commit splits
messageReceived(ctx, list) into two event handlers:

- messageReceived(ctx, msg)
- mmessageReceivedLast(ctx)

When Netty reads one or more messages, messageReceived(ctx, msg) event
is triggered for each message.  Once the current read operation is
finished, messageReceivedLast() is triggered to tell the handler that
the last messageReceived() was the last message in the current batch.

Similarly, for outbound, write(ctx, list) has been split into two:

- write(ctx, msg)
- flush(ctx, promise)

Instead of writing a list of message with a promise, a user is now
supposed to call write(msg) multiple times and then call flush() to
actually flush the buffered messages.

Please note that write() doesn't have a promise with it.  You must call
flush() to get notified on completion. (or you can use writeAndFlush())

Other changes:

- Because MessageList is completely hidden, codec framework uses
  List<Object> instead of MessageList as an output parameter.
2013-07-09 23:51:48 +09:00
Norman Maurer
1d196c5b59 [#1534] Fix handling of partial frames in SslHandler
* Let SslHandler not extend ByteToMessageDecoder
2013-07-09 15:53:18 +02:00
Norman Maurer
9c1b31d20a [#1535] Remove Channel.id() and so fix the possible leakage of the previous used Channel id map
The user can still use Channel.hashCode() for logging. It's just not 100% unique but should be good enough for most cases
2013-07-08 14:07:18 +02:00
Trustin Lee
da5a5af520 Add Test case for truncated SSL packets
- Related issue: #1534
- Also made sure ByteToMessageDecoder does not wrap EncoderException.
2013-07-07 11:52:34 +09:00
Trustin Lee
5010fe0a61 Fix a bug in SslHandler where a truncated packet isn't handled correctly
- Fixes #1534
2013-07-07 11:22:02 +09:00
Norman Maurer
824bb3419a [#1511] Fix NPE in AbstractTrafficShapingHandler which could happen if the READ_SUSPEND attr was not set yet 2013-07-02 19:30:05 +02:00
Norman Maurer
e06fcdbc6a [#1475] Correctly handle writes with empty MessageList in SslHandler 2013-06-25 11:07:14 +09:00
Norman Maurer
ad73dce7a1 [#1464] Make sure the ChannelPromise for writes is notified while using SslHandler 2013-06-25 11:07:14 +09:00
Trustin Lee
39d71440e1 Fix checkstyle 2013-06-25 11:07:13 +09:00
Trustin Lee
b6fdac7df3 Remove unnecessary logging handler - ByteLoggingHandler and MessageLoggingHandler / Extensible log message formatiing 2013-06-25 11:07:13 +09:00
Trustin Lee
01d9f10af6 Remove the volatile modifiers where they are unnecessary 2013-06-13 14:43:39 +09:00
Trustin Lee
7eb0f6105d Fix memory leaks 2013-06-13 13:32:47 +09:00
Norman Maurer
07858a3bd4 [#1411] Correctly check for null in AbstractTrafficShapperHandler.read() and so prevent NPE 2013-06-10 09:25:09 +02:00
Trustin Lee
14158070bf Revamp the core API to reduce memory footprint and consumption
The API changes made so far turned out to increase the memory footprint
and consumption while our intention was actually decreasing them.

Memory consumption issue:

When there are many connections which does not exchange data frequently,
the old Netty 4 API spent a lot more memory than 3 because it always
allocates per-handler buffer for each connection unless otherwise
explicitly stated by a user.  In a usual real world load, a client
doesn't always send requests without pausing, so the idea of having a
buffer whose life cycle if bound to the life cycle of a connection
didn't work as expected.

Memory footprint issue:

The old Netty 4 API decreased overall memory footprint by a great deal
in many cases.  It was mainly because the old Netty 4 API did not
allocate a new buffer and event object for each read.  Instead, it
created a new buffer for each handler in a pipeline.  This works pretty
well as long as the number of handlers in a pipeline is only a few.
However, for a highly modular application with many handlers which
handles connections which lasts for relatively short period, it actually
makes the memory footprint issue much worse.

Changes:

All in all, this is about retaining all the good changes we made in 4 so
far such as better thread model and going back to the way how we dealt
with message events in 3.

To fix the memory consumption/footprint issue mentioned above, we made a
hard decision to break the backward compatibility again with the
following changes:

- Remove MessageBuf
- Merge Buf into ByteBuf
- Merge ChannelInboundByte/MessageHandler and ChannelStateHandler into ChannelInboundHandler
  - Similar changes were made to the adapter classes
- Merge ChannelOutboundByte/MessageHandler and ChannelOperationHandler into ChannelOutboundHandler
  - Similar changes were made to the adapter classes
- Introduce MessageList which is similar to `MessageEvent` in Netty 3
- Replace inboundBufferUpdated(ctx) with messageReceived(ctx, MessageList)
- Replace flush(ctx, promise) with write(ctx, MessageList, promise)
- Remove ByteToByteEncoder/Decoder/Codec
  - Replaced by MessageToByteEncoder<ByteBuf>, ByteToMessageDecoder<ByteBuf>, and ByteMessageCodec<ByteBuf>
- Merge EmbeddedByteChannel and EmbeddedMessageChannel into EmbeddedChannel
- Add SimpleChannelInboundHandler which is sometimes more useful than
  ChannelInboundHandlerAdapter
- Bring back Channel.isWritable() from Netty 3
- Add ChannelInboundHandler.channelWritabilityChanges() event
- Add RecvByteBufAllocator configuration property
  - Similar to ReceiveBufferSizePredictor in Netty 3
  - Some existing configuration properties such as
    DatagramChannelConfig.receivePacketSize is gone now.
- Remove suspend/resumeIntermediaryDeallocation() in ByteBuf

This change would have been impossible without @normanmaurer's help. He
fixed, ported, and improved many parts of the changes.
2013-06-10 16:10:39 +09:00
Norman Maurer
d9c700e9fe [#1392] Fix bug where SslHandler returned wrong value for getCloseNotifyTimeoutMillis() 2013-05-24 23:18:14 +02:00
Trustin Lee
50433f7b9a Make sure successful SslHandshakeCompletionEvent is a singleton
- Related: #1373
- This commit is an amendment to 2a7bea2ad3
2013-05-18 05:50:58 +09:00
Norman Maurer
2a7bea2ad3 [#1373] Make SslHandshakeCompletionEvent constructor public 2013-05-17 19:39:52 +02:00
Trustin Lee
dc13b68632 Make sure ChannelHandler.handlerRemoved() is always invoked
- Fixes #1366: No elegant way to free non-in/outbound buffers held by a handler
- handlerRemoved() is now also invoked when a channel is deregistered, as well as when a handler is removed from a pipeline.
- A little bit of clean-up for readability
- Fix a bug in forwardBufferContentAndRemove() where the handler buffers are not freed (mainly because we were relying on channel.isRegistered() to determine if the handler has been removed from inside the handler.
- ChunkedWriteHandler.handlerRemoved() is unnecessary anymore because ChannelPipeline now always forwards the content of the buffer.
2013-05-16 19:32:39 +09:00