Skip to content


Subversion checkout URL

You can clone with
Download ZIP
Commits on Oct 8, 2013
  1. @mirakui

    Merge pull request #1 from winebarrel/master

    mirakui authored
    add reconnect db by USR1 signal
Commits on Oct 4, 2013
  1. @winebarrel

    fix hadler arg num

    winebarrel authored
  2. @winebarrel

    fix typo

    winebarrel authored
  3. @winebarrel

    add after_usr1 hook

    winebarrel authored
  4. @winebarrel

    remove empty line

    winebarrel authored
Commits on Oct 3, 2013
  1. @winebarrel
  2. @winebarrel

    add clean_reconnect_txt

    winebarrel authored
  3. @winebarrel

    remove: rails_reconnect.rb

    winebarrel authored
  4. @winebarrel
  5. @winebarrel
  6. @winebarrel

    add mysql2 to isolate libs

    winebarrel authored
  7. @winebarrel
  8. @winebarrel
  9. @winebarrel
  10. @winebarrel
  11. @winebarrel
  12. @winebarrel

    delete: require "rails"

    winebarrel authored
  13. @winebarrel

    fix method end

    winebarrel authored
  14. @winebarrel

    add recconect_rails

    winebarrel authored
Commits on Apr 17, 2012
  1. @mirakui
Commits on Feb 1, 2012
  1. add "timeout_signal" configuration;

    issei-naruta authored
    A signal that kill workers on timeout
Commits on Jan 28, 2012
  1. unicorn 4.2.0

    Eric Wong authored
    The GPLv3 is now an option to the Unicorn license.  The existing GPLv2
    and Ruby-only terms will always remain options, but the GPLv3 is
    Daemonization is correctly detected on all terminals for development
    use (Brian P O'Rourke).
    Unicorn::OobGC respects applications that disable GC entirely
    during application dispatch (Yuichi Tateno).
    Many test fixes for OpenBSD, which may help other *BSDs, too.
    (Jeremy Evans).
    There is now _optional_ SSL support (via the "kgio-monkey"
    RubyGem).  On fast, secure LANs, SSL is only intended for
    detecting data corruption that weak TCP checksums cannot detect.
    Our SSL support is remains unaudited by security experts.
    There are also some minor bugfixes and documentation
    Ruby 2.0.0dev also has a copy-on-write friendly GC which can save memory
    when combined with "preload_app true", so if you're in the mood, start
    testing Unicorn with the latest Ruby!
  2. doc: update doc for Ruby 2.0.0dev CoW-friendliness

    Eric Wong authored
    Ruby 2.0.0dev is the future and includes a CoW-friendly GC,
    so we shall encourage folks to give Ruby 2.0.0dev a spin.
Commits on Jan 27, 2012
  1. script/isolate_for_tests: disable sqlite3-ruby for Ruby 2.0.0dev

    Eric Wong authored
    We don't need it because we don't test old Rails with bleeding edge Ruby.
  2. disable old Rails tests for Ruby 2.0.0

    Eric Wong authored
    I doubt anybody would attempt to run ancient, unsupported versions
    of Rails on the latest (unreleased, even) versions of Ruby...
  3. script/isolate_for_tests: update to kgio 2.7.2

    Eric Wong authored
    Again, we test with the latest version.
Commits on Jan 24, 2012
  1. update tests for Rack 1.4.1

    Eric Wong authored
    Trying to ensure things always work with the latest version.
Commits on Jan 8, 2012
  1. Rakefile: swap URL for

    Eric Wong authored
Commits on Dec 28, 2011
  1. update tests for rack 1.4.0

    Eric Wong authored
    It's the latest and greatest version, so ensure everything
    works with it.
Commits on Dec 17, 2011
  1. http: test case for "Connection: TE"

    Eric Wong authored
    We need to be sure we don't barf on this header.
Commits on Dec 13, 2011
  1. cleanup exception handling on SIGUSR1

    Eric Wong authored
    No need to duplicate logic here
  2. quiet possible IOError from SIGUSR1 (reopen logs)

    Eric Wong authored
    It's possible for a SIGUSR1 signal to be received in the
    worker immediately before calling  In that case,
    do not clutter logging with IOError and just process the
    reopen log request.
Commits on Dec 5, 2011
  1. socket_helper: fix grammerr fail

    Eric Wong authored
    Oops :x
  2. socket_helper: set SO_KEEPALIVE on TCP sockets

    Eric Wong authored
    Even LANs can break or be unreliable sometimes and socket
    disconnect messages get lost, which means we fall back to
    the global (kill -9) timeout in Unicorn.
    While the default global timeout is much shorter (60s) than
    typical TCP timeouts, some HTTP application dispatches take much
    I/O or computational time (streaming many gigabytes), so the
    global timeout becomes ineffective.
    Under Linux, sysadmins are encouraged to lower the default
    net.ipv4.tcp_keepalive_* knobs in sysctl.  There should be
    similar knobs in other operating systems (the default keepalive
    intervals are usually ridiculously high, too high for anything).
    When the listen socket has SO_KEEPALIVE set, the flag should be
    inherited by accept()-ed sockets.
  3. socket_helper: remove out-of-date comment for TCP_NODELAY

    Eric Wong authored
    We favor low latency and consistency with the Unix socket
    behavior even with TCP.
Something went wrong with that request. Please try again.