fifottl queue entries with an explicit ttl get stuck in an odd state

when this syntax is used to put data in the buffer

queue.tube.test_buffer:put(data, {ttr=60, ttl=15768000000000000})

You end up with entries that have a ttl of 0 and a large next_event time

issue after the test fails the buffer has entries in a state where the ttl is 0, but next event is large [[select "task_id" from "test_buffer" where "ttl" = 0;]] the entries are not returned from take as they are expired, but they should not have expired

using docker image tarantool/tarantool:2.8.4
and a nodejs client

Place tarantool --version output here.
Tarantool 2.8.4-0-g47e6bd362
Target: Linux-x86_64-RelWithDebInfo
Build options: cmake . -DCMAKE_INSTALL_PREFIX=/usr/local -DENABLE_BACKTRACE=ON
Compiler: /usr/bin/cc /usr/bin/c++
C_FLAGS: -fexceptions -funwind-tables -fno-omit-frame-pointer -fno-stack-protector -fno-common -fopenmp -msse2 -std=c11 -Wall -Wextra -Wno-strict-aliasing -Wno-char-subscripts -Wno-format-truncation -Wno-gnu-alignof-expression -fno-gnu89-inline -Wno-cast-function-type
CXX_FLAGS: -fexceptions -funwind-tables -fno-omit-frame-pointer -fno-stack-protector -fno-common -fopenmp -msse2 -std=c++11 -Wall -Wextra -Wno-strict-aliasing -Wno-char-subscripts -Wno-format-truncation -Wno-invalid-offsetof -Wno-gnu-alignof-expression -Wno-cast-function-type

Steps to reproduce
see the project at
https://github.com/mannnick24/tarantool-fifottl-explicit-ttl/blob/main/docker/Dockerfile
this code uses node tarantool connection to concurrently call queue.tube.test_buffer:take, queue.tube.test_buffer:put and queue.tube.test_buffer:ack

How to run:

npm i

build.sh (in docker)

npm run test

Actual behavior
after the test fails the buffer has entries in a state where the ttl is 0, but next event is large [[select "task_id" from "test_buffer" where "ttl" = 0;]] the entries are not returned from take as they are expired, but they should not have expired

e.g. box.execute([[select "task_id", "status", "created", "ttr", "ttl", "next_event" from "el_buffer" where "ttl" = 0 limit 20;]])

metadata:

name: task_id type: unsigned

name: status type: string

name: created type: unsigned

name: ttr type: unsigned

name: ttl type: unsigned

name: next_event type: unsigned rows:

[4, 'r', 1662662470225835, 60000000, 0, 18446744073709551615]

Expected behavior
the buffer would be empty

Start with the linked reproduction project and its docker/Dockerfile, then run npm i, build.sh in Docker, and npm run test. Inspect the concurrent queue.tube.test_buffer:take, put, and ack calls, especially the explicit ttl case. Done means entries are not left with ttl 0 and a large next_event, and the reproduction no longer leaves incorrectly expired tasks unavailable from take.