Program terminated with signal 6, Aborted,有可能啥原因呢？其中一種原因就是事實上的OOM（雖然/var/log/message中沒有標明操作系統kill了進行，應該是進程內部初始化已申請內存時報錯了，因為malloc的申請會被OS盡可能延后的分配，所以很有可能已經申請的內存早就OOM了，但是程序還可以運行一段時間，甚至很久，除非設置了lock in memory，比如jdk和oracle、mysql都支持這么做）
但是也不一定，如下：
2022-04-01 06:45:04.269332C mysub zjh@postgres ::1(62148) walsender idle 55006[2022-04-01 06:45:04 UTC] 0 [88793] ERROR: replication slot "mysub" is active for PID 192415 2022-04-01 06:45:34.277434C mysub zjh@postgres ::1(62484) walsender idle 55006[2022-04-01 06:45:34 UTC] 0 [88976] ERROR: replication slot "mysub" is active for PID 192415 2022-04-01 06:48:30.667272C mysub zjh@postgres ::1(63883) walsender startup XX000[2022-04-01 06:46:04 UTC] 0 [89179] PANIC: stuck spinlock detected at InitWalSenderSlot, walsender.c:2415 2022-04-01 06:48:39.594554C @ postmaster 00000[2022-03-29 11:23:45 UTC] 0 [69880] LOG: server process (PID 89179) was terminated by signal 6: Aborted 2022-04-01 06:48:39.594567C @ postmaster 00000[2022-03-29 11:23:45 UTC] 0 [69880] LOG: terminating any other active server processes 2022-04-01 06:48:39.594624C pgbench_sub zjh@postgres ::1(63882) walsender idle 57P02[2022-04-01 05:19:52 UTC] 0 [231421] WARNING: terminating connection because of crash of another server process 2022-04-01 06:48:39.594624C pgbench_sub zjh@postgres ::1(63882) walsender idle 57P02[2022-04-01 05:19:52 UTC] 0 [231421] DETAIL: The postmaster has commanded this server process to roll back the current transaction and exit, because another server process exited abnormally and possibly corrupted shared memory. 2022-04-01 06:48:39.594624C pgbench_sub zjh@postgres ::1(63882) walsender idle 57P02[2022-04-01 05:19:52 UTC] 0 [231421] HINT: In a moment you should be able to reconnect to the database and repeat your command. 2022-04-01 06:48:39.594631C ltsql zjh@postgres [local] client backend idle 57P02[2022-04-01 06:14:36 UTC] 0 [254275] WARNING: terminating connection because of crash of another server process 2022-04-01 06:48:39.594631C ltsql zjh@postgres [local] client backend idle 57P02[2022-04-01 06:14:36 UTC] 0 [254275] DETAIL: The postmaster has commanded this server process to roll back the current transaction and exit, because another server process exited abnormally and possibly corrupted shared memory. 2022-04-01 06:48:39.594631C ltsql zjh@postgres [local] client backend idle 57P02[2022-04-01 06:14:36 UTC] 0 [254275] HINT: In a moment you should be able to reconnect to the database and repeat your command.
在postgresql/lightdb中，是因為默認（通過編譯時宏S_LOCK_TEST控制）會限制自旋鎖等待的時間，如果超過指定閾值（1ms-1s*1000），就會觸發stuck spinlock，然后導致lightdb/postgresql自動重啟。