Hi, all Kvrocks contributors, we want to donate Kvrocks to Apache Software Foundation, the first step for us is changing the license to Apache License Version 2.0. As we know, Kvrocks is an open source software under the BSD-3-Clause license, we truly thank every contributor and user, you guys make Kvrocks great and have an active community, before we could submit PR to change license, we must to get your approval.

@git-hulk @karelrooted @ShooterIT @Alfejik @caipengbo @ChrisZMF @shangxiaoxiong @ColinChamber @guoxiangCN @calvinxiao @patpatbear @WyattJia @smartlee @popunit @clyang @byronhe @QQxiaoyuyu @nioshield @singgel @liangjingyang @gaoyuanzong @rongzeng54 @unext-ops @fishery86 @xujianhai666 @LykxSassinator

Please reply "Approved" if you agree.

1. masterA 有数据 slaveA slaveof过来，数据可以同步，flushall masterA的数据， slaveA 没有同步清空掉。都执行过dbsize scan 2)masterB 没数据 slaveB 同步过来up后 ，masterB上执行写数据,数据没有同步到slaveB 3)slaveof后 slave节点上经常会报以下错误, 但这个目录是清空后再启动slave的

E0216 13:35:53.319345 26935 storage.cc:703] [storage] Failed to delete dir: IO error: file rmdir: ./kvrocksdatamaster/backup/meta: Directory not empty
E0216 13:35:53.415446 26935 replication.cc:584] [replication] Failed to restore backup while IO error: No such file or directoryWhile opening a file for sequentially reading: ./kvrocksdatamaster/backup/meta/5: No such file or directory
Segmentation fault (core dumped)

Reverts KvrocksLabs/kvrocks#442 l find that new multi-get api can get keys which have been deleted before. the case could be reproduced with my PR KvrocksLabs/kvrocks/pull/444 . It seems a bug of RocksDB and I would submit a relevant PR to RocksDB in the near future.

Search before asking

• [X] I had searched in the issues and found no similar issues.

Motivation

Modern CMake provides many amazing features to organize the build structure of a project in a more convenient way, and we can make full use of them to improve the development experience! @tisonkun had a good suggestion in https://github.com/apache/incubator-kvrocks/issues/562#issuecomment-1123771508 of creating a tracking issue to help keep the discussion more focused, so this issue is created for that purpose.

Solution

The list below tracks some of the improvements that have been proposed so far, and I'd appreciate it if anyone came up with new ideas for improvements, PRs for current improvements, or participated in this discussion!

• [x] #561
• [x] #562
• [x] #574
• [x] #569
• [x] #563
• [x] #580
• [x] #589
• [x] #590

Are you willing to submit a PR?

• [X] I'm willing to submit a PR!

Search before asking

• [X] I had searched in the issues and found no similar issues.

Version

redisGenericCommand use raiseError to process errors, but raiseError will perform a longjmp in the lua_error api call, so raiseError never returns.

longjmp does not compatible with RAII in C++ (C++ exceptions may use longjmp in implementation but the compiler will insert some dtor call autometically and carefully, i.e. stack unwinding), so every object in the stack of redisGenericCommand will never be destroyed, i.e. no destructor will be call, hence all resource allocated in ctor like std::vector, std::unique_ptr etc. cannot be released.

@git-hulk

Minimal reproduce step

./runtest --dont-clean on a kvrocks binary with Address Sanitizer enabled. (try to build with #599) You can only run tests in unit/scripting to save time : )

What did you expect to see?

No memory leaks

What did you see instead?

Memory leaks are reported by Address Sanitizer

Anything Else?

No response

Are you willing to submit a PR?

• [X] I'm willing to submit a PR!

Describe the bug A clear and concise description of what the bug is.

• OS: MAC OS 10.15.7
• Version 2.0.5 @6a4ed16a

get following error message when try to start up kvrocks ( first time run )

E1229 15:41:02.217978 276704704 main.cc:314] Failed to open: IO error: No such file or directory: While opening a file for sequentially reading: /tmp/kvrocks/db/CURRENT: No such file or directory

/tmp/kvrocks directory did exist but no sub-directory ./db under it.

FetchContent has been adopted by many projects as a dependency fetching method introduced in newer versions of cmake, for example the official documentation for gtest already recommends using FetchContent to import itself into other projects (replacing the previous recommended ExternalProject).

The advantage of FetchContent is that it is executed during the cmake configuration phase, which allows the configuration information of dependent projects to be introduced into the main project, unlike ExternalProject, which is executed during the build phase.

This is quite useful especially for dependency projects built with cmake, because the configured cmake target of a dependency project can be used directly in the main project via add_subdirectory (e.g. gtest::gtest), instead of having to manually configure the dependency's header directories, static or shared libraries, compilation options, install directories, etc. and keeping maintaining these redundant information in the main project.

This also makes git submodules unnecessary: by FetchContent, dependencies can be downloaded and configured at the cmake configuration stage, and URLs and commit hash/tag can be flexibly adjusted based on cmake options.

There are NOT a lot of dependencies in kvrocks, and I think using FetchContent to manage dependencies as opposed to ExternalProject will improve the development experience. Of course, introducing a dependency manager such as vcpkg or conan may be a good option, but it introduces additional complexity in development and build management, so there is a trade-off.

I'm happy to provide PRs if that sounds like a good idea.

@tisonkun

Describe the bug Kvrocks could not swich databases, any database key-value changes effect to all-other databases;

• OS: Ununtu 20.04 /docker
• Version 2.06/2.05

To Reproduce Steps to reproduce the behavior:

1. select 1 -> set t 1 -> get t -> "1"
2. select 2 -> get t -> "1"

Expected behavior database-1 key-value should not exist in database-2

127.0.0.1:6666> info
# Server
version:2.0.6
git_sha1:7c23f2fe
os:Linux 5.11.0-36-generic x86_64
gcc_version:7.5.0
arch_bits:64
process_id:1
tcp_port:6666
uptime_in_seconds:1277
uptime_in_days:0
# Clients
maxclients:10000
connected_clients:2
monitor_clients:0
# Memory
used_memory_rss:26529792
used_memory_human:25.30M
used_memory_lua:36864
used_memory_lua_human:36.00K
# Persistence
loading:0
bgsave_in_progress:0
last_bgsave_time:-1
last_bgsave_status:ok
last_bgsave_time_sec:-1
# Stats
total_connections_received:514
total_commands_processed:3090
instantaneous_ops_per_sec:1
total_net_input_bytes:76505
total_net_output_bytes:3490146
instantaneous_input_kbps:0
instantaneous_output_kbps:5
sync_full:0
sync_partial_ok:0
sync_partial_err:0
pubsub_channels:0
pubsub_patterns:0
# Replication
role:master
connected_slaves:0
master_repl_offset:1175757
# CPU
used_cpu_sys:0
used_cpu_user:1
# Commandstats
cmdstat_auth:calls=514,usec=1788,usec_per_call=3
cmdstat_client:calls=511,usec=747,usec_per_call=1
cmdstat_command:calls=3,usec=1770,usec_per_call=590
cmdstat_config:calls=510,usec=82148,usec_per_call=161
cmdstat_get:calls=4,usec=148,usec_per_call=37
cmdstat_info:calls=515,usec=373014,usec_per_call=724
cmdstat_keys:calls=5,usec=1709,usec_per_call=341
cmdstat_select:calls=6,usec=16,usec_per_call=2
cmdstat_set:calls=2,usec=268,usec_per_call=134
cmdstat_slowlog:calls=1020,usec=3864,usec_per_call=3
# Keyspace
# Last scan db time: Thu Jan  1 00:00:00 1970
db0:keys=0,expires=0,avg_ttl=0,expired=0
sequence:1175757
used_db_size:51464
max_db_size:0
used_percent: 0%
disk_capacity:490651459584
used_disk_size:111368089600
used_disk_percent: 22%
# RocksDB
estimate_keys[default]:0
block_cache_usage[default]:4256
block_cache_pinned_usage[default]:480
index_and_filter_cache_usage[default]:0
level0_file_limit_slowdown[default]:0
level0_file_limit_stop[default]:0
pending_compaction_bytes_slowdown[default]:0
pending_compaction_bytes_stop[default]:0
memtable_count_limit_slowdown[default]:0
memtable_count_limit_stop[default]:0
estimate_keys[metadata]:4
block_cache_usage[metadata]:4256
block_cache_pinned_usage[metadata]:480
index_and_filter_cache_usage[metadata]:224
level0_file_limit_slowdown[metadata]:0
level0_file_limit_stop[metadata]:0
pending_compaction_bytes_slowdown[metadata]:0
pending_compaction_bytes_stop[metadata]:0
memtable_count_limit_slowdown[metadata]:0
memtable_count_limit_stop[metadata]:0
estimate_keys[zset_score]:0
block_cache_usage[zset_score]:4256
block_cache_pinned_usage[zset_score]:480
index_and_filter_cache_usage[zset_score]:0
level0_file_limit_slowdown[zset_score]:0
level0_file_limit_stop[zset_score]:0
pending_compaction_bytes_slowdown[zset_score]:0
pending_compaction_bytes_stop[zset_score]:0
memtable_count_limit_slowdown[zset_score]:0
memtable_count_limit_stop[zset_score]:0
estimate_keys[pubsub]:0
block_cache_usage[pubsub]:432
block_cache_pinned_usage[pubsub]:0
index_and_filter_cache_usage[pubsub]:0
level0_file_limit_slowdown[pubsub]:0
level0_file_limit_stop[pubsub]:0
pending_compaction_bytes_slowdown[pubsub]:0
pending_compaction_bytes_stop[pubsub]:0
memtable_count_limit_slowdown[pubsub]:0
memtable_count_limit_stop[pubsub]:0
estimate_keys[propagate]:0
block_cache_usage[propagate]:0
block_cache_pinned_usage[propagate]:0
index_and_filter_cache_usage[propagate]:0
level0_file_limit_slowdown[propagate]:0
level0_file_limit_stop[propagate]:0
pending_compaction_bytes_slowdown[propagate]:0
pending_compaction_bytes_stop[propagate]:0
memtable_count_limit_slowdown[propagate]:0
memtable_count_limit_stop[propagate]:0
all_mem_tables:7343776
cur_mem_tables:7343776
snapshots:0
num_immutable_tables:0
num_running_flushes:0
memtable_flush_pending:0
compaction_pending:1
num_running_compactions:0
num_live_versions:5
num_superversion:7
num_background_errors:0
flush_count:0
compaction_count:2
put_per_sec:0
get_per_sec:0
seek_per_sec:0
next_per_sec:0
prev_per_sec:0
is_bgsaving:no
is_compacting:no

127.0.0.1:6666> set ttt 1
OK
127.0.0.1:6666> get ttt
"1"
127.0.0.1:6666> select 2
OK
127.0.0.1:6666[2]> get ttt
"1"

Describe the bug After SLAVEOF, 2 instances can contain different values for a given key.

To Reproduce Start 2 kvrocks instances and execute the following commands in the respective instances.

• SERVER2: SLAVEOF SERVER1
• SERVER2: SLAVEOF NO ONE
• SERVER2: SET test 222
• SERVER1: SET test 111
• SERVER2: SLAVEOF SERVER1
• SERVER2: GET test
  • Returns "222". However, it should return "111".

If a new SET command is used on SERVER1, SERVER2 immediately picks up the latest value. This issue doesn't occur if, on the "test" key, SERVER1 executed multiple SET commands while SERVER2 was not a slave.

Expected behavior After SLAVEOF command, SERVER2 should always contain identical data to SERVER1.

Describe

If your kvrocks config file has multiple rename-command, kvrocks would generate duplicate configuration items after executing config rewrite, whatever automatically or manually, and leads kvrocks to fail to restart.

It is introduced in #272

one example

rename-command debug                               290033_dont_run_debug_rudely
rename-command bgsave                              290033_dont_run_bgsave_rudely
rename-command shutdown                            290033_dont_run_shutdown_rudely
rename-command keys                                290033_dont_run_keys_rudely
rename-command debug                               290033_dont_run_debug_rudely

there are two debug rename-commands, so there will be the following error logs.

* Failed to load config, err: at line: #L19, err: No such command in rename-command

How to fix

• you can find the duplicate configuration items, and remove duplicate items, only keep one remame-command for one command.
• of course, you can cherry-pick this commit https://github.com/KvrocksLabs/kvrocks/commit/6c8534a17f056f06faa09bfeada3f92df686975a (#429 )
• use 2.0.6 version (coming soon)

Close #621.

Benchmark between make and ninja build for kvrocks can be found in #625. TL;DR: The benchmark shows that ninja is slightly slower than make build in clean build, but has huge advantages in incremental build, which is very friendly to kvrocks developers who need to do frequent incremental builds.

So in this PR, kvrocks still use make by default, but the problem that blocks ninja backend to successfully build in cmake is fixed and now build.sh has a --ninja option to enable ninja.

Search before asking

• [X] I had searched in the issues and found no similar issues.

Version

unstable

Minimal reproduce step

./build.sh build -DENABLE_ASAN=ON
cd tests/tcl
./runtest --dont-clean
grep Sanitizer tests/tmp -r

This bug is not 100% reproducible and may require multiple test executions before it appears.

What did you expect to see?

No ASan report.

What did you see instead?

=================================================================
==1872112==ERROR: AddressSanitizer: heap-use-after-free on address 0x608000014070 at pc 0x55d1a126595a bp 0x7fa436fee820 sp 0x7fa436fee810
READ of size 8 at 0x608000014070 thread T98
    #0 0x55d1a1265959 in Redis::CommandBPop::TryPopFromList() /home/twice/incubator-kvrocks/src/redis_cmd.cc:1568
    #1 0x55d1a1266851 in Redis::CommandBPop::WriteCB(bufferevent*, void*) /home/twice/incubator-kvrocks/src/redis_cmd.cc:1590
    #2 0x55d1a190f519 in bufferevent_run_deferred_callbacks_unlocked /home/twice/incubator-kvrocks/sanbuild/_deps/libevent-src/bufferevent.c:215
    #3 0x55d1a19169ac in event_process_active_single_queue /home/twice/incubator-kvrocks/sanbuild/_deps/libevent-src/event.c:1726
    #4 0x55d1a1917406 in event_process_active /home/twice/incubator-kvrocks/sanbuild/_deps/libevent-src/event.c:1789
    #5 0x55d1a1917406 in event_base_loop /home/twice/incubator-kvrocks/sanbuild/_deps/libevent-src/event.c:2012
    #6 0x55d1a14ffe9f in Worker::Run(std::thread::id) /home/twice/incubator-kvrocks/src/worker.cc:234
    #7 0x55d1a150017d in operator() /home/twice/incubator-kvrocks/src/worker.cc:435
    #8 0x55d1a150017d in __invoke_impl<void, WorkerThread::Start()::<lambda()> > /usr/include/c++/11.2.0/bits/invoke.h:61
    #9 0x55d1a150017d in __invoke<WorkerThread::Start()::<lambda()> > /usr/include/c++/11.2.0/bits/invoke.h:96
    #10 0x55d1a150017d in _M_invoke<0> /usr/include/c++/11.2.0/bits/std_thread.h:253
    #11 0x55d1a150017d in operator() /usr/include/c++/11.2.0/bits/std_thread.h:260
    #12 0x55d1a150017d in _M_run /usr/include/c++/11.2.0/bits/std_thread.h:211
    #13 0x55d1a19fa4a3 in execute_native_thread_routine (/home/twice/incubator-kvrocks/tests/tcl/redis-server+0xacd4a3)
    #14 0x7fa440e6b5c1 in start_thread (/usr/lib/libc.so.6+0x8d5c1)
    #15 0x7fa440ef0583 in __clone (/usr/lib/libc.so.6+0x112583)

0x608000014070 is located 80 bytes inside of 96-byte region [0x608000014020,0x608000014080)
freed by thread T98 here:
    #0 0x7fa4411a48d9 in operator delete(void*, unsigned long) /usr/src/debug/gcc/libsanitizer/asan/asan_new_delete.cpp:172
    #1 0x55d1a1424770 in std::default_delete<Redis::Commander>::operator()(Redis::Commander*) const /usr/include/c++/11.2.0/bits/unique_ptr.h:85
    #2 0x55d1a1424770 in std::__uniq_ptr_impl<Redis::Commander, std::default_delete<Redis::Commander> >::reset(Redis::Commander*) /usr/include/c++/11.2.0/bits/unique_ptr.h:182
    #3 0x55d1a1424770 in std::__uniq_ptr_impl<Redis::Commander, std::default_delete<Redis::Commander> >::operator=(std::__uniq_ptr_impl<Redis::Commander, std::default_delete<Redis::Commander> >&&) /usr/include/c++/11.2.0/bits/unique_ptr.h:167
    #4 0x55d1a1424770 in std::__uniq_ptr_data<Redis::Commander, std::default_delete<Redis::Commander>, true, true>::operator=(std::__uniq_ptr_data<Redis::Commander, std::default_delete<Redis::Commander>, true, true>&&) /usr/include/c++/11.2.0/bits/unique_ptr.h:212
    #5 0x55d1a1424770 in std::unique_ptr<Redis::Commander, std::default_delete<Redis::Commander> >::operator=(std::unique_ptr<Redis::Commander, std::default_delete<Redis::Commander> >&&) /usr/include/c++/11.2.0/bits/unique_ptr.h:371
    #6 0x55d1a1424770 in Server::LookupAndCreateCommand(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::unique_ptr<Redis::Commander, std::default_delete<Redis::Commander> >*) /home/twice/incubator-kvrocks/src/server.cc:1340
    #7 0x55d1a12c65ad in Redis::Connection::ExecuteCommands(std::vector<std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > >, std::allocator<std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > > > > const&) /home/twice/incubator-kvrocks/src/redis_connection.cc:330
    #8 0x55d1a12cef3f in Redis::Connection::OnRead(bufferevent*, void*) /home/twice/incubator-kvrocks/src/redis_connection.cc:89
    #9 0x55d1a190f4ad in bufferevent_run_deferred_callbacks_unlocked /home/twice/incubator-kvrocks/sanbuild/_deps/libevent-src/bufferevent.c:208

previously allocated by thread T98 here:
    #0 0x7fa4411a3811 in operator new(unsigned long) /usr/src/debug/gcc/libsanitizer/asan/asan_new_delete.cpp:99
    #1 0x55d1a11c3c56 in Redis::redisCommandTable::{lambda()#81}::operator()() const /home/twice/incubator-kvrocks/src/redis_cmd.cc:4807
    #2 0x55d1a11c3c56 in std::unique_ptr<Redis::Commander, std::default_delete<Redis::Commander> > std::__invoke_impl<std::unique_ptr<Redis::Commander, std::default_delete<Redis::Commander> >, Redis::redisCommandTable::{lambda()#81}&>(std::__invoke_other, Redis::redisCommandTable::{lambda()#81}&) /usr/include/c++/11.2.0/bits/invoke.h:61
    #3 0x55d1a11c3c56 in std::enable_if<is_invocable_r_v<std::unique_ptr<Redis::Commander, std::default_delete<Redis::Commander> >, Redis::redisCommandTable::{lambda()#81}&>, std::enable_if>::type std::__invoke_r<std::unique_ptr<Redis::Commander, std::default_delete<Redis::Commander> >, Redis::redisCommandTable::{lambda()#81}&>(std::unique_ptr<Redis::Commander, std::default_delete<Redis::Commander> >&&, (Redis::redisCommandTable::{lambda()#81}&)...) /usr/include/c++/11.2.0/bits/invoke.h:116
    #4 0x55d1a11c3c56 in std::_Function_handler<std::unique_ptr<Redis::Commander, std::default_delete<Redis::Commander> > (), Redis::redisCommandTable::{lambda()#81}>::_M_invoke(std::_Any_data const&) /usr/include/c++/11.2.0/bits/std_function.h:292
    #5 0x55d1a142439e in std::function<std::unique_ptr<Redis::Commander, std::default_delete<Redis::Commander> > ()>::operator()() const /usr/include/c++/11.2.0/bits/std_function.h:560
    #6 0x55d1a142439e in Server::LookupAndCreateCommand(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::unique_ptr<Redis::Commander, std::default_delete<Redis::Commander> >*) /home/twice/incubator-kvrocks/src/server.cc:1340
    #7 0x55d1a12c65ad in Redis::Connection::ExecuteCommands(std::vector<std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > >, std::allocator<std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > > > > const&) /home/twice/incubator-kvrocks/src/redis_connection.cc:330
    #8 0x55d1a12cef3f in Redis::Connection::OnRead(bufferevent*, void*) /home/twice/incubator-kvrocks/src/redis_connection.cc:89
    #9 0x55d1a190f4ad in bufferevent_run_deferred_callbacks_unlocked /home/twice/incubator-kvrocks/sanbuild/_deps/libevent-src/bufferevent.c:208

Thread T98 created by T0 here:
    #0 0x7fa441143eb7 in __interceptor_pthread_create /usr/src/debug/gcc/libsanitizer/asan/asan_interceptors.cpp:216
    #1 0x55d1a19fa78a in std::thread::_M_start_thread(std::unique_ptr<std::thread::_State, std::default_delete<std::thread::_State> >, void (*)()) (/home/twice/incubator-kvrocks/tests/tcl/redis-server+0xacd78a)
    #2 0x55d1a1443c6b in Server::Start() /home/twice/incubator-kvrocks/src/server.cc:125
    #3 0x55d1a1086cf5 in main /home/twice/incubator-kvrocks/src/main.cc:354
    #4 0x7fa440e0b30f in __libc_start_call_main (/usr/lib/libc.so.6+0x2d30f)

SUMMARY: AddressSanitizer: heap-use-after-free /home/twice/incubator-kvrocks/src/redis_cmd.cc:1568 in Redis::CommandBPop::TryPopFromList()
Shadow bytes around the buggy address:
  0x0c107fffa7b0: fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa
  0x0c107fffa7c0: fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa
  0x0c107fffa7d0: fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa
  0x0c107fffa7e0: fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa
  0x0c107fffa7f0: fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa
=>0x0c107fffa800: fa fa fa fa fd fd fd fd fd fd fd fd fd fd[fd]fd
  0x0c107fffa810: fa fa fa fa fd fd fd fd fd fd fd fd fd fd fd fd
  0x0c107fffa820: fa fa fa fa 00 00 00 00 00 00 00 00 00 00 00 fa
  0x0c107fffa830: fa fa fa fa 00 00 00 00 00 00 00 00 00 00 00 00
  0x0c107fffa840: fa fa fa fa 00 00 00 00 00 00 00 00 00 00 00 00
  0x0c107fffa850: fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa
Shadow byte legend (one shadow byte represents 8 application bytes):
  Addressable:           00
  Partially addressable: 01 02 03 04 05 06 07 
  Heap left redzone:       fa
  Freed heap region:       fd
  Stack left redzone:      f1
  Stack mid redzone:       f2
  Stack right redzone:     f3
  Stack after return:      f5
  Stack use after scope:   f8
  Global redzone:          f9
  Global init order:       f6
  Poisoned by user:        f7
  Container overflow:      fc
  Array cookie:            ac
  Intra object redzone:    bb
  ASan internal:           fe
  Left alloca redzone:     ca
  Right alloca redzone:    cb
  Shadow gap:              cc
==1872112==ABORTING

Anything Else?

No response

Are you willing to submit a PR?

• [ ] I'm willing to submit a PR!

1. Luajit replace lua
2. rm test case "cmsgpack can pack and unpack circular references". Becase luajit's core code diffrent from lua. Also, different compilers produce different code execution effects (GCC&clang on Ubuntu).

Search before asking

• [X] I had searched in the issues and found no similar issues.

Motivation

Refer to #689 and #691, since we build the nightly version of the docker image we can automatically publish it to some registries like the docker hub, e.g. some tags like nightly, nightly-20220629.

Solution

No response

Are you willing to submit a PR?

• [ ] I'm willing to submit a PR!

Search before asking

• [X] I had searched in the issues and found no similar issues.

Motivation

As shown in the figure above, the retrying workflow was canceled due to a new commit. We may retry the kvrocks ci action of some commits in the unstable branch. At this time, if the unstable branch can only run one workflow at the same time, it will cause some trouble.

Solution

No response

Are you willing to submit a PR?

• [ ] I'm willing to submit a PR!

Search before asking

• [X] I had searched in the issues and found no similar issues.

Motivation

There are many manual memory (or other system resources like file) allocation and deallocation such as new/delete in kvrocks.

Memory management is a complex topic (after RC/tracing GC is involved quickly) and there are some smart pointers like unique_ptr, shared_ptr in C++ to abstract a specific way to manage resources.

For some trivial scenario which do not require to share objects, we can do it via unique_ptr, which is a zero-overhead abstraction.

We can switch some manual deallocations to unique_ptr to make the code more maintainable and less prone to memory leaks.

NOTE: We try to use unique_ptr instead of creating our own RAII class types because unique_ptr has good semantics: it prohibits copy construction and only allows move construction, which keeps us from accidentally sharing objects and causing Use After Free or Double Free. (And we should carefully to retrieve raw pointers via get() by checking that the callee do not require ownership)

Solution

No response

Are you willing to submit a PR?

• [ ] I'm willing to submit a PR!