What's Changed

• reactor: eager close of resource fds in the pending destruction list by @ymmt2005 in #98
• Prevent slave data delete when unable to connect to master by @nojima in #100

Full Changelog: v1.1.12...v1.1.13

What's Changed

• Fix busy loop when FDs are exhausted by @nojima in #91
• Early close resource file descriptors by @ymmt2005 in #95
• Bump version to 1.1.12 by @ymmt2005 in #96

Full Changelog: v1.1.11...v1.1.12

What's Changed

• New config option "initial_repl_sleep_delay_usec" by @arosh in #88

Full Changelog: v1.1.10...v1.1.11

See ChangeLog for the full details.

See ChangeLog for the full details.

See ChangeLog for the full details.

See ChangeLog for the full details.

Changes:

• New config option "bind_ip" as suggested by @justnx ( #63, #64 ).
• Fixed a minor replication bug #61.

This release fixes an important bug that blocked the master for unexpectedly long duration
when one of its slaves silently dies. Two new config options were introduced with the fix.

• repl_buffer_size adjusts the replication buffer size in the master process.
• slave_timeout adjusts how long the master waits a non-responsive slave before cutting the connection.

Other two minor bugs were fixed too. See ChangeLog for the full details.

Changes:

• add keys memcached extension.