This commit fixes a bug where the cluster module fails to propagate
EADDRINUSE errors.
When a worker starts a (net, http) server, it requests the listen socket
from its master who then creates and binds the socket.
Now, OS X and Windows don't always signal EADDRINUSE from bind() but
instead defer the error until a later syscall. libuv mimics this
behaviour to provide consistent behaviour across platforms but that
means the worker could end up with a socket that is not actually bound
to the requested addresss.
That's why the worker now checks if the socket is bound, raising
EADDRINUSE if that's not the case.
Fixes#2721.
Register the 'close' event listener with .once(), not .on().
It doesn't matter in the grand scheme of things because the listener
doesn't keep references to any heavy-weight objects but using .once()
for a oneshot listener is something of a best practice.
This appears to fix#4673. That bug is very hard to reproduce, so it's
hard to tell for certain, but this approach is more correct anyway.
Hat-tip: @dougwilson
Prior to v0.10, Node ignored ECONNRESET errors in many situations.
There *are* valid cases in which ECONNRESET should be ignored as a
normal part of the TCP dance, but in many others, it's a very relevant
signal that must be heeded with care.
Exacerbating this problem, if the OutgoingMessage does not have a
req.connection._handle, it assumes that it is in the process of
connecting, and thus buffers writes up in an array.
The problem happens when you reuse a socket between two requests, and it
is destroyed abruptly in between them. The writes will be buffered,
because the socket has no handle, but it's not ever going to GET a
handle, because it's not connecting, it's destroyed.
The proper fix is to treat ECONNRESET correctly. However, this is a
behavior/semantics change, and cannot land in a stable branch.
Fix#4775
In zlibBuffer(), don't wait for the garbage collector to reclaim the zlib memory
but release it manually. Reduces memory consumption by a factor of 10 or more
with some workloads.
Test case:
function f() {
require('zlib').deflate('xxx', g);
}
function g() {
setTimeout(f, 5);
}
f();
Observe RSS memory usage with and without this commit. After 10,000 iterations,
RSS stabilizes at ~35 MB with this commit. Without, RSS is over 300 MB and keeps
growing.
Cause: whenever the JS object heap hits the high-water mark, the V8 GC sweeps
it clean, then tries to grow it in order to avoid more sweeps in the near
future. Rule of thumb: the bigger the JS heap, the lazier the GC can be.
A side effect of a bigger heap is that objects now live longer. This is harmless
in general but it affects zlib context objects because those are tied to large
buffers that live outside the JS heap, on the order of 16K per context object.
Ergo, don't wait for the GC to reclaim the memory - it may take a long time.
Fixes#4172.
Fix issue where SlowBuffers couldn't be passed as target to Buffer
copy().
Also included checks to see if Argument parameters are defined before
assigning their values. This offered ~3x's performance gain.
Backport of 16bbecc from master branch. Closes#4633.
Reject negative offsets in SlowBuffer::MakeFastBuffer(), it allows
the creation of buffers that point to arbitrary addresses.
Reported by Trevor Norris.
Fix the following exception:
http.js:974
this._httpMessage.emit('close');
^
TypeError: Cannot call method 'emit' of null
at Socket.onServerResponseClose (http.js:974:21)
at Socket.EventEmitter.emit (events.js:124:20)
at net.js:421:10
at process._tickCallback (node.js:386:13)
at process._makeCallback (node.js:304:15)
Fixes#4586.
Raise a TypeError when the argument to send() or sendto() is anything
but a Buffer.
Fixes the following assertion:
$ node -e 'require("dgram").createSocket("udp4").send("BAM")'
node: ../../src/udp_wrap.cc:220: static v8::Handle<v8::Value>
node::UDPWrap::DoSend(const v8::Arguments&, int): Assertion
`Buffer::HasInstance(args[0])' failed.
Aborted (core dumped)
Fixes#4496.
Apply the same optimization to res.end(buf) that is applied to res.end(str).
Speeds up `node benchmark/http_simple_auto -k -c 1 -n 25000 buffer/1`
(non-chunked response body) by about 750x. That's not a typo.
Chunked responses:
$ cat tmp/http-chunked-client.js
// Run `node benchmark/http_simple` in another terminal.
var http = require('http'), url = require('url');
var options = url.parse('http://127.0.0.1:8000/buffer/1/1');
options.agent = new http.Agent({ maxSockets: 1 });
for (var i = 0; i < 25000; ++i) http.get(options);
Before:
$ time out/Release/node tmp/http-chunked-client.js
real 16m40.411s
user 0m9.184s
sys 0m0.604s
After:
$ time out/Release/node tmp/http-chunked-client.js
real 0m5.386s
user 0m2.768s
sys 0m0.728s
That's still a 185x speed-up.
Fixes#4415.
Make parser errors bubble up to the ClientRequest instead of the underlying
net.Socket object.
This is a back-port of commit c78678b from the master branch.
Fixes#3776.
Fix#4331
Using double negate forces values into 32bit space. Because of this
Math.ceil needs to be used. Since NaN comparisons are always false, use
that to our advantage to return 0 if it is.
Also added two tests to verify the changes.
* Added isIP method to make use of inet_pton to cares_wrap.cc
* Modified net.isIP() to make use of new C++ isIP method.
* Added new tests to test-net-isip.js.
This is a back-port of commit fb6377e from the master branch.
Always add domain, _events, and _maxListeners properties, set to the
default values at first.
Leads to a very very slight perf improvement when using setMaxListeners,
or dealing with a lot of EE objects that don't have any listeners.
Before there was this weird module-scoped "context" variable which seemingly
shared the "context" of subsequent REPL instances, unless ".clear" was invoked
inside the REPL. To be proper, we need to ensure that each REPL gets its own
"context" object. I literally don't know why this "sharing" behavior was in place
before, but it was just plain wrong.
Just as the 'WWW-Authenticate' HTTP header the 'Proxy-Authenticate' header might
be received several times as well. Currently only one value is preserved. This
change allows to receive multiple values concatenated by space and comma.
This addresses #4034. There are two problems happening:
1. The domain is not exited automatically when calling dispose() on it.
Then, since the domain is disposed, attempting to exit it again will do
nothing.
2. The active domain is stored on process.domain. Since thrown errors
call `process.emit('uncaughtException', er)`, and the process is an
event emitter with a `.domain` member, it re-enters the domain a second
time before calling the error handler, pushing it onto the stack again.
Thus, if the handler calls `domain.dispose()`, then the domain is now on
the stack twice, and cannot be exited properly. Since the domain is
disposed, any subsequent IO will be no-op'ed, since we've declared that
this context is done and best forgotten.
The solution here is twofold:
1. In EventEmitter.emit, do not enter the domain if `this===process`.
2. Automatically exit the domain when calling `domain.dispose()`.
Make sure the deletion event gets reported in the following scenario:
1. Watch a file.
2. The initial stat() goes okay.
3. Something deletes the watched file.
4. The second stat() fails with ENOENT.
The second stat() translates into the first 'change' event but a logic error
stopped it from getting emitted.
Fixes#4027.
A HTTP/1.0 client does not support 'Transfer-Encoding: chunked' unless it
explicitly requests it by sending a 'TE: chunked' header.
Before this commit, node.js always disabled chunked encoding for HTTP/1.0
clients. Now it will scan for the TE header and turn on chunked encoding if
requested and applicable.
Fixes#940.
Ben Noordhuis
isaacs
Bert Belder
Trevor Norris
Dan Milon
Fedor Indutny
Brian White
Joshua Erickson
Nathan Rajlich
thewilli
Frédéric Germain
koichik
Shigeki Ohtsu