From: Nikita Pettik <korablev@tarantool.org> To: tarantool-patches@dev.tarantool.org Subject: [Tarantool-patches] [PATCH] Introduce CHANGELOG.md Date: Wed, 23 Dec 2020 01:43:57 +0300 [thread overview] Message-ID: <2bbbd5ec35903b3418446b82f57b29463d82982e.1608659706.git.korablev@tarantool.org> (raw) --- This one is for upcoming 2.7 release. All suggestions concerning changelog format are welcome. CHANGELOG.md | 69 ++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 69 insertions(+) create mode 100644 CHANGELOG.md diff --git a/CHANGELOG.md b/CHANGELOG.md new file mode 100644 index 000000000..7823cf7c1 --- /dev/null +++ b/CHANGELOG.md @@ -0,0 +1,69 @@ +This file is a manually maintained list of changes for each release. Feel free +to add your changes here when sending pull requests or patches to the mailing list. +Also send corrections if you spot any mistakes. + +# 2.7.1 beta + +## Overview + +Date: 2020-12-29 +Tag: 2.7.1-0-XX + +## Compatibility + +## Functionality added or changed + +### Core +* Now it is allowed to define index without extra braces if there is one part only: parts = {field1, type1, ...} (gh-2866). +* Varbinary type is now supported by bitset index (gh-5392). +* Provide more pedantic key-parts verification so that now index-related options can't be specified in their definition (gh-5473). +* Print log warning if schema version is older than last available schema version (gh-4574). +* UUID values created via uuid Lua module (`require('uuid')`) can be compared using normal comparison operators `<`, `>=`, and so on (gh-5511). +* Added `box.ctl.is_recovery_finished()` function which allows user to determine whether memtx recovery is finished or not. +* Changed `force_recovery` option behaviour: now it also ignores errors during snapshot recovery (gh-5422). + +### Vinyl + +### LuaJIT + +### SQL +* Added ALTER TABLE ADD COLUMN statement support for empty tables (gh-2349, gh-3075). + +### Replication + +### Build + +### Misc + +### Module API + +## Bugs fixed + +### Core +* `fiber.cond:wait()` ignored the timeout if the fiber is cancelled and returned without any signs of an error. Now it throws an error that the fiber is cancelled (gh-5013). +* Fixed memory corruption, which was most visible on Mac, but could affect any system (gh-5312). +* Fixed a dynamic module not being unloaded from memory in case of attempt to load a not existing function from it (gh-5475). +* Fixed a crash on `swim:quit()` call (gh-4570). +* Fixed performance degradation in snapshot recovery when no JSON path or multikey indices are involved (gh-4774). + +### Replication +* Fixed a false-positive "too long WAL write" message for synchronous transactions (gh-5139). +* `box.ctl.wait_rw()` could return even if the instance is not writable due to having foreign synchronous transactions. As a result, there was no a proper way to wait until the automatically elected leader becomes writable. Now `box.ctl.wait_rw()` works correctly (gh-5440). +* Fixed a couple of crashes on various tweaks of election mode (gh-5506). + +### SQL +* Forbid data changes in read-only mode (gh-5231). +* Fixed occasionally raised unrelated error ("Space '0' does not exist") during query execution (gh-5592). +* Fixed segfault caused by coinciding names of temporary files (used to store data during execution) having two instances running on the same machine (gh-5537). + +### LuaJIT +* Fixed address clashing occurring while __call metamethod dispatching (gh-4518, gh-4649). + +### Lua +* Fixed an error reporting at attempt to use a net.box connection that is not established yet (gh-4787). +* Fixed NULL dereference on error paths in merger, usually on a 'wrong' key_def (gh-5450). + +### Memtx +* Fixed bug with `on_schema_init` triggers which led to duplicates in primary key (gh-5304). + +### Module API -- 2.17.1
next reply other threads:[~2020-12-22 22:44 UTC|newest] Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top 2020-12-22 22:43 Nikita Pettik [this message] 2020-12-24 11:02 ` Sergey Bronnikov
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=2bbbd5ec35903b3418446b82f57b29463d82982e.1608659706.git.korablev@tarantool.org \ --to=korablev@tarantool.org \ --cc=tarantool-patches@dev.tarantool.org \ --subject='Re: [Tarantool-patches] [PATCH] Introduce CHANGELOG.md' \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: link
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox