This commit reintroduces 31251c9 and 63a5799. These commits introduced a
memory leak and were reverted. This commit brings those commits back
and fixes the memory leak by removing unnecessary retain method calls.
This reverts commit 31251c9 introduced in #30695.
We suspect this commit is causing the OOME's reported in #30811 and we will use this PR to test this assertion.
This is related to #29500 and #28898. This commit removes the abilitiy
to disable http pipelining. After this commit, any elasticsearch node
will support pipelined requests from a client. Additionally, it extracts
some of the http pipelining work to the server module. This extracted
work is used to implement pipelining for the nio plugin.
This commit is related to #28898. It adds an nio driven http server
transport. Currently it only supports basic http features. Cors,
pipeling, and read timeouts will need to be added in future PRs.
With this commit we determine the maximum number of buffers that Netty keeps
while accumulating one HTTP request based on the maximum content length (default
1500 bytes, overridable with the system property `es.net.mtu`). Previously, we
kept the default value of 1024 which is too small for bulk requests which leads
to unnecessary copies of byte buffers internally.
Relates #29448
This commit removes the http.enabled setting. While all real nodes (started with bin/elasticsearch) will always have an http binding, there are many tests that rely on the quickness of not actually needing to bind to 2 ports. For this case, the MockHttpTransport.TestPlugin provides a dummy http transport implementation which is used by default in ESIntegTestCase.
closes#12792
I am not sure why we have this leniency for HTTP max content length, it
has been there since the beginning
(5ac51ee93f) with no explanation of its
source. That said, our philosophy today is different than the philosophy
of the past where Elasticsearch would be quite lenient in its handling
of settings and today we aim for predictability for both users and
us. This commit removes leniency in the parsing of
http.max_content_length.
We use a latch when sending requests during tests so that we do not hang
forever waiting for replies on those requests. This commit increases the
timeout on that latch to 30 seconds because sometimes 10 seconds is just
not enough.
Today we have a few problems with how we handle bad requests:
- handling requests with bad encoding
- handling requests with invalid value for filter_path/pretty/human
- handling requests with a garbage Content-Type header
There are two problems:
- in every case, we give an empty response to the client
- in most cases, we leak the byte buffer backing the request!
These problems are caused by a broader problem: poor handling preparing
the request for handling, or the channel to write to when the response
is ready. This commit addresses these issues by taking a unified
approach to all of them that ensures that:
- we respond to the client with the exception that blew us up
- we do not leak the byte buffer backing the request
* Decouple XContentBuilder from BytesReference
This commit removes all mentions of `BytesReference` from `XContentBuilder`.
This is needed so that we can completely decouple the XContent code and move it
into its own dependency.
While this change appears large, it is due to two main changes, moving
`.bytes()` and `.string()` out of XContentBuilder itself into static methods
`BytesReference.bytes` and `Strings.toString` respectively. The rest of the
change is code reacting to these changes (the majority of it in tests).
Relates to #28504
* Remove log4j dependency from elasticsearch-core
This removes the log4j dependency from our elasticsearch-core project. It was
originally necessary only for our jar classpath checking. It is now replaced by
a `Consumer<String>` so that the es-core dependency doesn't have external
dependencies.
The parts of #28191 which were moved in conjunction (like `ESLoggerFactory` and
`Loggers`) have been moved back where appropriate, since they are not required
in the core jar.
This is tangentially related to #28504
* Add javadocs for `output` parameter
* Change @code to @link
We have code used in the networking layer to search for errors buried in
other exceptions. This code will be useful in other locations so with
this commit we move it to our exceptions helpers.
Relates #28691
The is a follow up to #28567 changing the method used to capture stack traces, as requested
during the review. Instead of creating a throwable, we explicitly capture the stack trace of the
current thread. This should Make Jason Happy Again ™️ .
This commit updates netty to 4.1.16.Final. This is the latest version that we can have work without
extra permissions. This updated version of netty fixes issues seen with Java 9 and some data
not being sent, which results in timeouts.
The method `initiateChannel` on `TcpTransport` is explicit in that
channels can be connect asynchronously. All production implementations
do connect asynchronously. Only the blocking `MockTcpTransport`
connects in a synchronous manner. This avoids testing some of the
blocking code in `TcpTransport` that waits on connections to complete.
Additionally, it requires a more extensive method signature than
required for other transports.
This commit modifies the `MockTcpTransport` to make these connections
asynchronously on a different thread. Additionally, it simplifies that
`initiateChannel` method signature.
This is related to #27933. It introduces a jar named elasticsearch-core
in the lib directory. This commit moves the JarHell class from server to
elasticsearch-core. Additionally, PathUtils and some of Loggers are
moved as JarHell depends on them.
This is related to #27260. This commit moves the NioTransport from
:test:framework to a new nio-transport plugin. Additionally, supporting
tcp decoding classes are moved to this plugin. Generic byte reading and
writing contexts are moved to the nio library.
Additionally, this commit adds a basic MockNioTransport to
:test:framework that is a TcpTransport implementation for testing that
is driven by nio.
This commit attempts to continue unifying the logic between different
transport implementations. As transports call a `TcpTransport` callback
when a new channel is accepted, there is no need to internally track
channels accepted. Instead there is a set of accepted channels in
`TcpTransport`. This set is used for metrics and shutting down channels.
This is related to #27563. This commit modifies the
InboundChannelBuffer to support releasable byte pages. These byte
pages are provided by the PageCacheRecycler. The PageCacheRecycler
must be passed to the Transport with this change.
We currently do not have any server-side read timeouts implemented in
elasticsearch. This commit adds a read timeout setting that defaults to
30 seconds. If after 30 seconds a read has not occurred, the channel
will be closed. A timeout of value of 0 will disable the timeout.
This is related to #27422. Right now when we send a write to the netty
transport, we attach a listener to the future. When you submit a write
on the netty event loop and the event loop is shutdown, the onFailure
method is called. Unfortunately, netty then tries to notify the listener
which cannot be done without dispatching to the event loop. In this
case, the dispatch fails and netty logs and error and does not tell us.
This commit checks that netty is still not shutdown after sending a
message. If netty is shutdown, we complete the listener.
Currently we use ActionListener<TcpChannel> for connect, close, and send
message listeners in TcpTransport. However, all of the listeners have to
capture a reference to a channel in the case of the exception api being
called. This commit changes these listeners to be type <Void> as passing
the channel to onResponse is not necessary. Additionally, this change
makes it easier to integrate with low level transports (which use
different implementations of TcpChannel).
This commit is a follow up to the work completed in #27132. Essentially
it transitions two more methods (sendMessage and getLocalAddress) from
Transport to TcpChannel. With this change, there is no longer a need for
TcpTransport to be aware of the specific type of channel a transport
returns. So that class is no longer parameterized by channel type.
Right now our different transport implementations must duplicate
functionality in order to stay compliant with the requirements of
TcpTransport. They must all implement common logic to open channels,
close channels, keep track of channels for eventual shutdown, etc.
Additionally, there is a weird and complicated relationship between
Transport and TransportService. We eventually want to start merging
some of the functionality between these classes.
This commit starts moving towards a world where TransportService retains
all the application logic and channel state. Transport implementations
in this world will only be tasked with returning a channel when one is
requested, calling transport service when a channel is accepted from
a server, and starting / stopping itself.
Specifically this commit changes how channels are opened and closed. All
Transport implementations now return a channel type that must comply with
the new TcpChannel interface. This interface has the methods necessary
for TcpTransport to completely manage the lifecycle of a channel. This
includes setting the channel up, waiting for connection, adding close
listeners, and eventually closing.
This is a followup to #26521. This commit expands the alias added for
the elasticsearch client codebase to all codebases. The original full
jar name property is left intact. This only adds an alias without the
version, which should help ease the pain in updating any versions (ES
itself or dependencies).
If creating the REST request throws an exception (for example, because
of invalid headers), we leak the request due to failure to release the
buffer (which would otherwise happen after replying on the
channel). This commit addresses this leak by handling the failure case.
Relates #27222
This commit removes the `ByteBufStreamInput` `readBytesReference` and
`readBytesRef` methods. These methods are zero-copy which means that
they retain a reference to the underlying netty buffer. The problem is
that our `TcpTransport` is not designed to handle zero-copy. The netty
implementation sets the read index past the current message once it has
been deserialized, handled, and mostly likely dispatched to another
thread. This means that netty is free to release this buffer. So it is
unsafe to retain a reference to it without calling `retain`. And we
cannot call `retain` because we are not currently designed to handle
reference counting past the transport level.
This should not currently impact us as we wrap the `ByteBufStreamInput`
in `NamedWriteableAwareStreamInput` in the `TcpTransport`. This stream
essentially delegates to the underling stream. However, in the case of
`readBytesReference` and `readBytesRef` it leaves thw implementations
to the standard `StreamInput` methods. These methods call the read byte
array method which delegates to `ByteBufStreamInput`. The read byte
array method on `ByteBufStreamInput` copies so it is safe. The only
impact of this commit should be removing methods that could be dangerous
if they were eventually called due to some refactoring.
Right now we are attempting to set SO_LINGER to 0 on server channels
when we are stopping the tcp transport. This is not a supported socket
option and throws an exception. This also prevents the channels from
being closed.
This commit 1. doesn't set SO_LINGER for server channges, 2. checks
that it is a supported option in nio, and 3. changes the log message
to warn for server channel close exceptions.
While opening a connection to a node, a channel can subsequently
close. If this happens, a future callback whose purpose is to close all
other channels and disconnect from the node will fire. However, this
future will not be ready to close all the channels because the
connection will not be exposed to the future callback yet. Since this
callback is run once, we will never try to disconnect from this node
again and we will be left with a closed channel. This commit adds a
check that all channels are open before exposing the channel and throws
a general connection exception. In this case, the usual connection retry
logic will take over.
Relates #26932
With this commit we simplify our network layer by only allowing to define a
fixed receive predictor size instead of a minimum and maximum value. This also
means that the following (previously undocumented) settings are removed:
* http.netty.receive_predictor_min
* http.netty.receive_predictor_max
Using an adaptive sizing policy in the receive predictor is a very low-level
optimization. The implications on allocation behavior are extremely hard to grasp
(see our previous work in #23185) and adaptive sizing does not provide a lot of
benefits (see benchmarks in #26165 for more details).
* Add additional low-level logging handler
We have the trace handler which is useful for recording sent messages
but there are times where it would be useful to have more low-level
logging about the events occurring on a channel. This commit adds a
logging handler that can be enabled by setting a certain log level
(org.elasticsearch.transport.netty4.ESLoggingHandler) to trace that
provides trace logging on low-level channel events and includes some
information about the request/response read/write events on the channel
as well.
* Remove imports
* License header
* Remove redundant
* Add test
* More assertions
We should unwrap the cause looking for any suppressed errors or root
causes that are errors when checking if we should maybe die. This commit
causes that to be the case.
Relates #26884
This commit changes the log level on a write and flush failure to warn
as this is not necessarily an Elasticsearch problem but more likely
indicative of an infrastructure problem.
This commit fixes a #26855. Right now we set SO_LINGER to 0 if we are
stopping the transport. This can throw a ChannelClosedException if the
raw channel is already closed. We have a number of scenarios where it is
possible this could be called with a channel that is already closed.
This commit fixes the issue be checking that the channel is not closed
before attempting to set the socket option.
This commit reorders a maybe die check and a logging statement for the
following reasons:
- we should die as quickly as possible if the cause is fatal
- we do not want the JVM to be so broken that when we try to log
another exception is thrown (maybe another out of memory exception)
and then the maybe die is never invoked
- maybe die will log the cause anyway if the cause is fatal so we only
need to log if the cause is not fatal
At current, we do not feel there is enough of a reason to shade the low
level rest client. It caused problems with commons logging and IDE's
during the brief time it was used. We did not know exactly how many
users will need this, and decided that leaving shading out until we
gather more information is best. Users can still shade the jar
themselves. For information and feeback, see issue #26366.
Closes#26328
This reverts commit 3a20922046.
This reverts commit 2c271f0f22.
This reverts commit 9d10dbea39.
This reverts commit e816ef89a2.
There is a group of five settings relating to raw tcp configurations
(no_delay, buffer sizes, etc) that we have for the http transport. These
currently live in the netty module. As they are unrelated to netty
specifically, this commit moves these settings to the
`HttpTransportSettings` class in core.
Right now it is possible for the `HttpPipeliningHandler` to queue
pipelined responses. On channel close, we do not clear and release these
responses. This commit releases the responses and completes the promise.
With this commit we remove the following three previously unused
(and undocumented) Netty 4 related settings:
* transport.netty.max_cumulation_buffer_capacity,
* transport.netty.max_composite_buffer_components and
* http.netty.max_cumulation_buffer_capacity
from Elasticsearch.
This commit fixes an issue with the Netty 4 multi-port test that a
transport client can connect. The problem here is that in case the
bottom of the random port range was already bound to (for example, by
another JVM) then then transport client could not connect to the data
node. This is because the transport client was in fact using the bottom
of the port range only. Instead, we simply try all the ports that the
data node might be bound to.
Closes#24441
This commit removes all external dependencies from the rest client jar
and shades them in an 'org.elasticsearch.client' package within the jar
using shadowJar gradle plugin. All projects that depended on the
existing jar have been converted to using the 'org.elasticsearch.client'
package prefixes to interact with the rest client.
Closes#25208
There is a bug when a call to `BytesReferenceStreamInput` skip is made
on a `BytesReference` that has an initial offset. The offset for the
current slice is added to the current index and then subtracted from the
length. This introduces the possibility of a negative number of bytes to
skip. This happens inside a loop, which leads to an infinte loop.
This commit correctly subtracts the current slice index from the
slice.length. Additionally, the `BytesArrayTests` are modified to test
instances that include an offset.
Currently when we close a channel in Netty4Utils.closeChannels we
block until the closing is complete. This introduces the possibility
that a network selector thread will block while waiting until a
separate network selector thread closes a channel.
For instance: T1 closes channel 1 (which is assigned to a T1 selector).
Channel 1's close listener executes the closing of the node. That
means that T1 now tries to close channel 2. However, channel 2 is
assigned to a selector that is running on T2. T1 now must wait until T2
closes that channel at some point in the future.
This commit addresses this by adding a boolean to closeChannels
indicating if we should block on close. We only set this boolean to true
if we are closing down the server channels at shutdown. This call is
never made from a network thread. When we call the closeChannels method
with that boolean set to false, we do not block on close.
Tim Brooks
Colin Goodheart-Smithe
Daniel Mitterdorfer
Ryan Ernst
Jason Tedor
Lee Hinman
Boaz Leskes
Jay Modi
Igor Motov
Armin Braun
Yannick Welsch
Simon Willnauer
Michael Basnight