Sergey Ostanevich

Apparently future object should keep a reference to the connection.

Need a test to confirm we're safe: the new leader should ignore the outdated PROMOTE from previous one, sending the latter to rebootstrap.

In progress at #53

I feel uneasy because of inconsistency between the console mode and the lua script mode. Namely, things one tried in console doesn't work in a script passed as an argument...

It is definitely not a bug, rather a feature request (semantic change even). We also didn't come to a resolution how it should behave in this case - it's not...

Doesn't reproduce: ``` dev1.tarantool.org:test $ python3 ../../test/test-run.py --builddir=/home/s.ostanevich/workspaces/t.sergos2/build --vardir=./var app-tap/gh-2717-no-quit-sigint.test.lua Started ../../test/test-run.py --builddir=/home/s.ostanevich/workspaces/t.sergos2/build --vardir=./var app-tap/gh-2717-no-quit-sigint.test.lua Running in parallel with 160 workers Timeout options: ------------------- REPLICATION_SYNC_TIMEOUT: 100 TEST_TIMEOUT: 110 NO_OUTPUT_TIMEOUT: 120...

The problem is propagator (apparently - folder) makes a wrong transformation in 0072 substituting the newly created string pointer (0051) with original string [0x40002498] so that the difference between them...

The culprit is **LJFOLDF(kfold_strref_snew)** Apparently, there should be a check that the **strref (new_string, b)** is the only use of the new string. And I'm too rookie to fix this...

Copied into https://github.com/LuaJIT/LuaJIT/issues/505 Proposed fix is: ```diff diff --git a/src/lj_ffrecord.c b/src/lj_ffrecord.c index 849d7a2..e879c2a 100644 --- a/src/lj_ffrecord.c +++ b/src/lj_ffrecord.c @@ -949,8 +949,9 @@ static void LJ_FASTCALL recff_string_find(jit_State *J, RecordFFData *rd) str->len-(MSize)start,...

Have a try - pushed new fixup.