Ankur Mhatre
eaa8171966
Three replicas of one application against one PostgreSQL database, proving duplicate @Scheduled execution and then removing it with ShedLock: 24 executions where 8 were due, then 7 for 7 ticks. Also measured: @SchedulerLock without @EnableSchedulerLock does nothing and warns about nothing; spring.task.scheduling.pool.size=1 does not starve a fixedRate job but delays it and fires 35 of 40 executions in a burst; and a node whose clock is 40 seconds fast takes a live lock unless the provider uses usingDbTime().
19 lines
772 B
SQL
19 lines
772 B
SQL
-- Both tables are created by every replica at startup, so every statement is IF NOT EXISTS.
|
|
-- The shedlock table's shape is fixed by ShedLock: name, lock_until, locked_at, locked_by.
|
|
-- lock_until and locked_at must be timestamps; with usingDbTime() they are written by the
|
|
-- database, so their precision and time zone are the database's, not the JVM's.
|
|
create table if not exists shedlock (
|
|
name varchar(64) not null,
|
|
lock_until timestamp not null,
|
|
locked_at timestamp not null,
|
|
locked_by varchar(255) not null,
|
|
primary key (name)
|
|
);
|
|
|
|
create table if not exists job_execution (
|
|
id serial primary key,
|
|
job_name varchar(64) not null,
|
|
instance_id varchar(64) not null,
|
|
started_at timestamp not null
|
|
);
|