Release Name: 0.3.0
Notes:
Rainbows! is a HTTP server for sleepy Rack applications. It is based on
Unicorn, but designed to handle applications that expect long
request/response times and/or slow clients. For Rack applications not
heavily bound by slow external network dependencies, consider Unicorn
instead as it simpler and easier to debug.
* http://rainbows.rubyforge.org/
* rainbows-talk@rubyforge.org
* git://git.bogomips.org/rainbows.git
Changes:
The major feature of this release is the new DeferredResponse
middleware for the Rev-based concurrency model. It should be
transparently compatible with non-Rev models, as well. As a
pleasant side effect, this change also allows large files to be
streamed to the client with Rev as the socket becomes writable
instead of slurping the entire file into an IO::Buffer first.
Bugfixes to graceful shutdowns support for all concurrency
models. The Rev-based model also gets a working heartbeat
mechanism (oops!) and fixed HTTP/1.1 pipelining support.
Eric Wong (38):
app_pool: note it being currently broken with Revactor
Revactor tests can sleep more easily
tests: sleep.ru handles "Expect: 100-continue"
Fix graceful shutdown handling of Thread* models harder
DRY setting of rack.multithread
test-lib: dbgcat adds headers with key name
use timeout correctly to join threads on SIGQUIT
Rev: simplification to error handling
tests: sleep.ru slurps rack.input stream
refactor graceful shutdowns again, harder
tests: introduce require_for_model function
tests: add unbuffered tee(1)-like helper
tests: rack.input trailer tests for all models
tests: fix issues with non-portable shell constructs
tests: fix random_blob dependency
tests: factor out a common parser error "library"
tests: DRY setting of the "model" environment var
tests: DRY Ruby requires based on model
test-lib: quiet down pipefail error message
tests: DRY require tests for Rev/Revactor
rev: handle fully-buffered, pipelined requests
rev: avoid stack overflow through pipelining
tests: common basic HTTP tests for all models
tests: rack.input hammer concurrency testing
tests: for log reopening for all concurrency models
http_response: filter out X-Rainbows-* headers
rev: fix heartbeat timeouts
revactor: switch to a 1 second heartbeat
rev: async response bodies with DevFdResponse middleware
tests: more reliable error checking
tests: DWIM FIFO creation
tests: predictable and simpler tempfile management
rev: AsyncResponse => DeferredResponse API cleanup
rev: update documentation for this model
TUNING: update documentation notes
TODO: update with new items
local.mk.sample: sync with BDFL's version
Rainbows! 0.3.0
Changes:
GIT-VERSION-GEN | 2 +-
TODO | 15 ++
TUNING | 14 ++
lib/rainbows.rb | 7 +
lib/rainbows/app_pool.rb | 7 +-
lib/rainbows/base.rb | 28 ++--
lib/rainbows/const.rb | 6 +-
lib/rainbows/dev_fd_response.rb | 69 ++++++++
lib/rainbows/http_response.rb | 1 +
lib/rainbows/http_server.rb | 2 +
lib/rainbows/rev.rb | 179 +++++++++++++++-----
lib/rainbows/revactor.rb | 20 +--
lib/rainbows/thread_pool.rb | 24 ++--
lib/rainbows/thread_spawn.rb | 8 +-
local.mk.sample | 10 +-
t/GNUmakefile | 8 +-
t/README | 2 +-
t/async-response-no-autochunk.ru | 24 +++
t/async-response.ru | 13 ++
t/bin/content-md5-put | 2 +-
t/bin/utee | 12 ++
t/env.ru | 3 +
t/large-file-response.ru | 13 ++
t/lib-async-response-no-autochunk.sh | 6 +
t/lib-async-response.sh | 45 +++++
t/lib-graceful.sh | 40 +++++
t/lib-input-trailer.sh | 63 +++++++
t/lib-large-file-response.sh | 45 +++++
t/lib-parser-error.sh | 29 +++
...actor-tee-input.sh => lib-rack-input-hammer.sh} | 12 +-
t/lib-reopen-logs.sh | 60 +++++++
t/lib-simple-http.sh | 92 ++++++++++
t/sleep.ru | 19 ++-
t/t0000-basic.sh | 37 +----
t/t1000-thread-pool-basic.sh | 42 +-----
t/t1002-thread-pool-graceful.sh | 37 +----
t/t1003-thread-pool-reopen-logs.sh | 2 +
t/t1004-thread-pool-async-response.sh | 1 +
t/t1005-thread-pool-large-file-response.sh | 1 +
t/t1006-thread-pool-async-response-no-autochunk.sh | 1 +
t/t1100-thread-pool-rack-input.sh | 2 +
t/t1101-thread-pool-input-trailer.sh | 2 +
t/t2000-thread-spawn-basic.sh | 38 +----
t/t2002-thread-spawn-graceful.sh | 37 +----
t/t2003-thread-spawn-reopen-logs.sh | 2 +
t/t2004-thread-spawn-async-response.sh | 1 +
t/t2005-thread-spawn-large-file-response.sh | 1 +
...006-thread-spawn-async-response-no-autochunk.sh | 1 +
t/t2100-thread-spawn-rack-input.sh | 2 +
t/t2101-thread-spawn-input-trailer.sh | 2 +
t/t3000-revactor-basic.sh | 40 +-----
t/t3001-revactor-pipeline.sh | 46 -----
t/t3002-revactor-graceful.sh | 38 +----
t/t3003-revactor-reopen-logs.sh | 54 +------
t/t3004-revactor-async-response.sh | 1 +
t/t3005-revactor-large-file-response.sh | 2 +
t/t3006-revactor-async-response-no-autochunk.sh | 1 +
t/t3100-revactor-rack-input.sh | 2 +
t/t3101-revactor-rack-input-trailer.sh | 2 +
t/t4000-rev-basic.sh | 51 +------
t/t4002-rev-graceful.sh | 52 +------
t/t4003-rev-parser-error.sh | 34 +----
t/t4003-rev-reopen-logs.sh | 2 +
t/t4004-rev-async-response.sh | 1 +
t/t4005-rev-large-file-response.sh | 2 +
t/t4006-rev-async-response-no-autochunk.sh | 1 +
t/t4100-rev-rack-input.sh | 44 +-----
t/t4101-rev-rack-input-trailer.sh | 51 +------
t/t9000-rack-app-pool.sh | 5 +-
t/test-lib.sh | 98 +++++++++--
70 files changed, 895 insertions(+), 721 deletions(-)
commit 25c0b000d50ed65e5118b17725334be916e93f47
Author: Eric Wong
Date: Mon Oct 19 11:13:34 2009 -0700
Rainbows! 0.3.0
The major feature of this release is the new DeferredResponse
middleware for the Rev-based concurrency model. It should be
transparently compatible with non-Rev models, as well. As a
pleasant side effect, this change also allows large files to be
streamed to the client with Rev as the socket becomes writable
instead of slurping the entire file into an IO::Buffer first.
Bugfixes to graceful shutdowns support for all concurrency
models. The Rev-based model also gets a working heartbeat
mechanism (oops!) and fixed HTTP/1.1 pipelining support.
Eric Wong (37):
app_pool: note it being currently broken with Revactor
Revactor tests can sleep more easily
tests: sleep.ru handles "Expect: 100-continue"
Fix graceful shutdown handling of Thread* models harder
DRY setting of rack.multithread
test-lib: dbgcat adds headers with key name
use timeout correctly to join threads on SIGQUIT
Rev: simplification to error handling
tests: sleep.ru slurps rack.input stream
refactor graceful shutdowns again, harder
tests: introduce require_for_model function
tests: add unbuffered tee(1)-like helper
tests: rack.input trailer tests for all models
tests: fix issues with non-portable shell constructs
tests: fix random_blob dependency
tests: factor out a common parser error "library"
tests: DRY setting of the "model" environment var
tests: DRY Ruby requires based on model
test-lib: quiet down pipefail error message
tests: DRY require tests for Rev/Revactor
rev: handle fully-buffered, pipelined requests
rev: avoid stack overflow through pipelining
tests: common basic HTTP tests for all models
tests: rack.input hammer concurrency testing
tests: for log reopening for all concurrency models
http_response: filter out X-Rainbows-* headers
rev: fix heartbeat timeouts
revactor: switch to a 1 second heartbeat
rev: async response bodies with DevFdResponse middleware
tests: more reliable error checking
tests: DWIM FIFO creation
tests: predictable and simpler tempfile management
rev: AsyncResponse => DeferredResponse API cleanup
rev: update documentation for this model
TUNING: update documentation notes
TODO: update with new items
local.mk.sample: sync with BDFL's version
commit da59c30e7265d2718110286f226e6b4caaaa766c
Author: Eric Wong
Date: Mon Oct 19 11:09:22 2009 -0700
local.mk.sample: sync with BDFL's version
commit 8453ad07c53907a54283a8eddc423840e0d9649c
Author: Eric Wong
Date: Sat Oct 17 16:02:24 2009 -0700
TODO: update with new items
commit eba5b17340c24cf70e09f7d8fe381691e32feb9e
Author: Eric Wong
Date: Mon Oct 19 10:20:34 2009 -0700
TUNING: update documentation notes
commit 9c5ce5b2f86591977727b7fdaa8fcc50aea089d6
Author: Eric Wong
Date: Mon Oct 19 10:20:33 2009 -0700
rev: update documentation for this model
commit 77e526918d16caffdee371f96a3fbc40f8e95dcb
Author: Eric Wong
Date: Sun Oct 18 23:53:02 2009 -0700
rev: AsyncResponse => DeferredResponse API cleanup
The "async" moniker wasn't appropriate since this API also
handles static files without slurping, so "deferred" is a more
appropriate term (even if I have trouble speling words with
double conssonants in them).
The DeferredResponse.write method now emulates the
HttpResponse.write method for consistency.
commit dbb2e039fac19d9b74075fd79cddade534cafeca
Author: Eric Wong
Date: Sun Oct 18 23:09:09 2009 -0700
tests: predictable and simpler tempfile management
Instead of using completely random names, we'll use
predictable ones since we already depend on them for
exit codes and such. This drops our ability to run
the same test for the same version of Ruby in the
same working tree, but that's an unlikely scenario.
While we're at it, avoid remove tempfiles if a test
failed. This should make debugging easier.
commit 9922b5e683e86c6df48931ee90da1c5e6ad419ef
Author: Eric Wong
Date: Sun Oct 18 22:11:39 2009 -0700
tests: DWIM FIFO creation
If we're going to name a variable "fifo", it'll be for
descriptive reasons...
commit c41a08821bf11eefebda3ea3e536eda8f4b4629b
Author: Eric Wong
Date: Sun Oct 18 22:05:13 2009 -0700
tests: more reliable error checking
We now check for SIGKILL, too
commit 7b01d94dd9287ac402d91451f1e93c9faaf913c4
Author: Eric Wong
Date: Sun Oct 18 15:59:29 2009 -0700
rev: async response bodies with DevFdResponse middleware
This new middleware should be a no-op for non-Rev concurrency
models (or by explicitly setting env['rainbows.autochunk'] to
false).
Setting env['rainbows.autochunk'] to true (the default when Rev
is used) allows (e)poll-able IO objects (sockets, pipes) to be
sent asynchronously after app.call(env) returns.
This also has a fortunate side effect of introducing a code path
which allows large, static files to be sent without slurping
them into a Rev IO::Buffer, too. This new change works even
without the DevFdResponse middleware, so you won't have to
reconfigure your app.
This lets us epoll on response bodies that come in from a pipe
or even a socket and send them either straight through or with
chunked encoding.
commit d0103759ae63b0ed1084f6a9d2b7ede538e8c871
Author: Eric Wong
Date: Sun Oct 18 20:08:17 2009 -0700
revactor: switch to a 1 second heartbeat
Like everybody else... Closing the listener sockets doesn't seem
to wakeup the actors reliably and since it's easier to use a 1
second heartbeat than correct signal/messaging for all the rest
of the other clients, we'll just do that instead of relying on
one-off signal handlers.
commit 87a9cf757bbef7bb54a187933c1bc61ed1f05583
Author: Eric Wong
Date: Sun Oct 18 20:04:57 2009 -0700
rev: fix heartbeat timeouts
Oops, looks like they were never implemented at all.
commit 1a1335fe5b1894848b48a233c9f6a80634329b40
Author: Eric Wong
Date: Sun Oct 18 09:35:59 2009 -0700
http_response: filter out X-Rainbows-* headers
We'll be using some custom headers to craft responses
commit 1c0dab453688e8bfc9f21656e9b89b88fb3cc1d3
Author: Eric Wong
Date: Sat Oct 17 22:42:56 2009 -0700
tests: for log reopening for all concurrency models
commit 49981da0edf3cb142079ee9341640bd5b798c880
Author: Eric Wong
Date: Sat Oct 17 22:42:56 2009 -0700
tests: rack.input hammer concurrency testing
commit 61a50566ea0799a9b19c3dcfdf9b1ae5ea083dd2
Author: Eric Wong
Date: Sat Oct 17 22:42:56 2009 -0700
tests: common basic HTTP tests for all models
We support pipelining, keepalive, and even HTTP/0.9!
commit 0719c5a7ac09078c810507244a86ba44623757c4
Author: Eric Wong
Date: Sat Oct 17 22:42:56 2009 -0700
rev: avoid stack overflow through pipelining
Handling HTTP pipelining through recursion is not good since
several hundred kilobytes worth of GET/HEAD requests can be a
LOT of GET/HEAD requests...
commit b4d479681c257d91c7784a6698264719632be374
Author: Eric Wong
Date: Sat Oct 17 22:42:56 2009 -0700
rev: handle fully-buffered, pipelined requests
This leaves us vulnerable to stack overflows through excessive
pipelining. The next patch will fix things hopefully.
commit 2dd23ad0a97d8818eba8298acd8c51405725b099
Author: Eric Wong
Date: Sat Oct 17 22:42:56 2009 -0700
tests: DRY require tests for Rev/Revactor
commit 926bf35b9ead1224ae3037fabb58625de9a66d98
Author: Eric Wong
Date: Sat Oct 17 22:42:56 2009 -0700
test-lib: quiet down pipefail error message
We'll spit out a proper warning later anyways...
commit 49806778ce03114cb2e29b1adc53020de2299abe
Author: Eric Wong
Date: Sat Oct 17 22:42:56 2009 -0700
tests: DRY Ruby requires based on model
commit 3a5af445e2a3af83cd2e7e68b2d4c0d1dc8079b5
Author: Eric Wong
Date: Sat Oct 17 22:42:56 2009 -0700
tests: DRY setting of the "model" environment var
commit c50132e156da457186929e5c36d9e16f636f48df
Author: Eric Wong
Date: Sat Oct 17 22:42:56 2009 -0700
tests: factor out a common parser error "library"
commit cd91835b826b968df08ae1bcde794a9d54334182
Author: Eric Wong
Date: Sat Oct 17 22:42:56 2009 -0700
tests: fix random_blob dependency
commit caddc46054b720bd88757b0c15de1d305216fa77
Author: Eric Wong
Date: Sat Oct 17 22:42:56 2009 -0700
tests: fix issues with non-portable shell constructs
At least these tests all run with dash now, but ksh93
or bash is still recommended for pipefail
commit b32e9240ec68f2493038c5cfbc9ca6859cca6ad8
Author: Eric Wong
Date: Sat Oct 17 22:42:56 2009 -0700
tests: rack.input trailer tests for all models
Just in case we break something. Also add staggered
blob test to simulate slow client uploads.
commit 95417ca711a75612da86a25acd20134efdbc0e67
Author: Eric Wong
Date: Sat Oct 17 22:42:56 2009 -0700
tests: add unbuffered tee(1)-like helper
Buffering enabled in tee(1) was making tests more
difficult to debug.
commit b68b53c0323f691dbc412fa5ff581ff0b9a7ef1f
Author: Eric Wong
Date: Sat Oct 17 22:42:56 2009 -0700
tests: introduce require_for_model function
This makes it easier to write/share code for multi-model tests.
commit a764983fccd6cce64043d76e09a5e1718e7f8fd6
Author: Eric Wong
Date: Sat Oct 17 22:42:56 2009 -0700
refactor graceful shutdowns again, harder
We use the "G" global constant from the Rev model everywhere
to simplify things a little.
Test cases are more consistent now, too.
commit 71a1cf90ce16ce6951f1ce76fcea98b13bc54576
Author: Eric Wong
Date: Sat Oct 17 22:42:56 2009 -0700
tests: sleep.ru slurps rack.input stream
Since we do "Expect: 100-continue" handling, make sure
we read anything in the request body the client
will send us.
commit ce5bb179bcd0ac2921da636375cb3821b79c428c
Author: Eric Wong
Date: Sat Oct 17 22:42:55 2009 -0700
Rev: simplification to error handling
on_write_complete has no chance of being called
there so remove the unnecessary ensure statement
commit d838e5ef01471eb12aec42639246946f7b3da4a5
Author: Eric Wong
Date: Sat Oct 17 22:42:55 2009 -0700
use timeout correctly to join threads on SIGQUIT
Subtraction is a difficult concept for some folks (like
myself) to grasp and implement.
commit 394e1aa79d16ed9405e35e3cd3749a7e8c51522a
Author: Eric Wong
Date: Sat Oct 17 22:42:55 2009 -0700
test-lib: dbgcat adds headers with key name
Makes it easier to track down empty files this way
commit b79ce7e6594259669763251e3cbf4404a6381d7c
Author: Eric Wong
Date: Sat Oct 17 22:42:55 2009 -0700
DRY setting of rack.multithread
It's more fool-proof this way and prevents us from using
idiotic/non-obvious concurrency model names.
commit e0c33c583c5b4d2a4661edb6c41d6a60b44dae9d
Author: Eric Wong
Date: Sat Oct 17 22:42:55 2009 -0700
Fix graceful shutdown handling of Thread* models harder
I need better tests for graceful shutdown...
commit 031127b1d047d05be7a2ec8af7480d02e9bbdd69
Author: Eric Wong
Date: Sat Oct 17 22:42:55 2009 -0700
tests: sleep.ru handles "Expect: 100-continue"
commit 7a3ebef424f3e7db0a48ae35885c446d79b7d971
Author: Eric Wong
Date: Sat Oct 17 22:42:55 2009 -0700
Revactor tests can sleep more easily
Since we export "rainbows.model" to Rack now, we don't
have to worry about setting the "SLEEP_CLASS" env for
the application.
commit 08c52c81bc49023a645dd211c4c983a60ac64e13
Author: Eric Wong
Date: Thu Oct 15 13:42:07 2009 -0700
app_pool: note it being currently broken with Revactor
Revactor does not use threads and blocking on a stock Queue
class does not work. Eventually this should be made to work
with the Actor model, but until then, we'll at least document
it...
|
