This included removing some internal version constraints. It would be
nice if package A could say "use B@2.0.0" (when both have changed), but
when they're both in the release, we need to make a release that has a
B@2.0.0-rc in it, which doesn't match that constraint. Fortunately,
constraints aren't necessary within a release anyway.
(This is the third attempt to enable this. I believe as of
permessage-deflate 0.1.3, it works properly.)
By default, we attempt to use this for every websocket message on both
client and server.
On the server, we provide the SERVER_WEBSOCKET_COMPRESSION environment
variable to control compression. If $SERVER_WEBSOCKET_COMPRESSION is
set, then it must be valid JSON. If it represents a falsey value, then
we do not use permessage-deflate at all; otherwise, the JSON value is
used as an argument to deflate's configure method; see
https://github.com/faye/permessage-deflate-node/blob/master/README.md
We do not provide a way to use it only on some messages. The underlying
spec allows this but permessage-deflate does not; see
https://github.com/faye/permessage-deflate-node/issues/2
We do not provide a mechanism to control compression parameters on the
client side. The assumption is that the common reason to care about
compression parameters is to control server per-connection memory
usage. (The noContextTakeover configuration parameter should save some
memory and still allow for some compression, for example.)
Addresses #3007 (which will not be fixed until this change is deployed
on the package server as well).
This reverts commit 5d3cfa2d76, reversing
changes made to 2b466a9015.
Ah well, we tried to enable websocket compression again and ran into
more bugs.
First, newer versions of websocket-driver seem to sometimes send
duplicate close messages:
https://github.com/faye/faye-websocket-node/issues/41
This occurs whether or not deflate is actually used.
Second, in some circumstances permessage-deflate seems to completely
corrupt messages. This was reasonably easily observable by running
test-packages with Chrome, and seeing that sometimes (but not always) a
large number of bad JSON messages got printed to the client
console. (Another symptom was that the total number of tests would be
larger than it should be, leading to messages like "Passed 1109 of
1153", presumably because the test name got corrupted in some status
messages.) https://github.com/faye/permessage-deflate-node/issues/4
See #3007.
Specifically, it should only be an error if we actually try to make a
server, not if we're just loading this incidentally into our tool as
part of an isopacket. (Loading isopackets should never fail unless
there is a bug in our code; that's why it prints an ugly stack trace.)
The important part here is the upgrades of websocket-driver and
websocket-extensions, which fix the bugs that made compression
unusable. The other upgrades shouldn't hurt, though.
Also, note that 0.3.13 was a bad release which reverted the change we
needed from 0.3.12.
A merge of the windows-cordova branch did not increment the version numbers
of some packages before adding pre-release versions. This causes constraint
solver errors, when, for example, trying to build a release from devel. This
change should fix the most obvious candidates.
ie, when processing each oplog entry, don't even look at the triggers
from unrelated collection.
Also, just make "every trigger and notification needs a string
collection" into a hard rule.
This is apparently much faster for at least one real user; see #3697.
Summary: should be self-contained and documented
Test Plan: unit tests
Reviewers: ekatek
Differential Revision: https://phabricator.meteor.io/D29
Also remove SIGHUP handler. This was designed for a legacy system, and
SIGHUP would be a better fit for what autoupdate currently uses SIGUSR2
for. This includes removing some code in the ddp server to close
sockets on SIGHUP.
The server-side Meteor.call abstraction is "RPC", not "function call",
so you shouldn't be able to mutate internal state using it. This is a
similar change to that done in e91713d9 with function parameters.
Fixes#3201.
In Session.close, `self.socket.close` could trigger this event handler:
socket.on('close', function () {
if (socket._meteorSession) {
Fiber(function () {
socket._meteorSession.close();
}).run();
}
});
which could trigger a reentrant call to Session.close. The self.inQueue
guard was not sufficient to stop multiple execution, because it was too
low.
Symptoms included:
- The "sessions" server fact would be decremented twice and become
inaccurate (and even negative!)
- Connection.onClose callbacks could be called twice
Fixes#3331.
This reverts commit 67bea9c102.
See https://github.com/faye/permessage-deflate-node/issues/1
This can be consistently replicated by running test-packages ddp (note
that the tests pass but then the server crashes). "livedata server -
connection in publish function" specifically is enough
By default, we attempt to use this for every websocket message on both
client and server.
On the server, we provide the SERVER_WEBSOCKET_COMPRESSION environment
variable to control compression. If $SERVER_WEBSOCKET_COMPRESSION is
set, then it must be valid JSON. If it represents a falsey value, then
we do not use permessage-deflate at all; otherwise, the JSON value is
used as an argument to deflate's configure method; see
https://github.com/faye/permessage-deflate-node/blob/master/README.md
We do not provide a way to use it only on some messages. The underlying
spec allows this but permessage-deflate does not; see
https://github.com/faye/permessage-deflate-node/issues/2
We do not provide a mechanism to control compression parameters on the
client side. The assumption is that the common reason to care about
compression parameters is to control server per-connection memory
usage. (The noContextTakeover configuration parameter should save some
memory and still allow for some compression, for example.)
Addresses #3007 (which will not be fixed until this change is deployed
on the package server as well).
