Thread: BUG #16365: recieve smart shutdown after system start

The following bug has been logged on the website:

Bug reference:      16365
Logged by:          Roman Pivnitskiy
Email address:      prd87@mail.ru
PostgreSQL version: 9.5.7
Operating system:   Debian GNU/Linux 8.8 (jessie)
Description:

Hello.
On some servers, we began to encounter the "strange" behavior of
postgresql.
When shutting down the server at the command of UPS (shutdown now). The next
time you turn on postgresql, it gets smart shutdown immediately at startup
In logs pg
when server shutdown
2020-04-15 05:26:48 MSK [936]: [4-1] db=,user=,app=,client= LOG:  received
fast shutdown request
2020-04-15 05:26:48 MSK [936]: [5-1] db=,user=,app=,client= LOG:  aborting
any active transactions
2020-04-15 05:26:48 MSK [1338]: [2-1] db=,user=,app=,client= LOG:
autovacuum launcher shutting down
2020-04-15 05:26:53 MSK [1335]: [7664-1] db=,user=,app=,client= LOG:
checkpoint complete: wrote 13604 buffers (23.6%); 0 transaction log file(s)
added, 0 removed, 0 recycled; write=472.958 s, sync=2.984 s, total=476.084
s; sync files=251, longest=1.098 s, average=0.011 s; distance=125092 kB,
estimate=203540 kB
2020-04-15 05:26:53 MSK [1335]: [7665-1] db=,user=,app=,client= LOG:
shutting down
2020-04-15 05:26:54 MSK [1335]: [7666-1] db=,user=,app=,client= LOG:
checkpoint starting: shutdown immediate
2020-04-15 05:26:54 MSK [1335]: [7667-1] db=,user=,app=,client= LOG:
checkpoint complete: wrote 79 buffers (0.1%); 0 transaction log file(s)
added, 8 removed, 0 recycled; write=0.001 s, sync=0.393 s, total=0.552 s;
sync files=18, longest=0.104 s, average=0.021 s; distance=15863 kB,
estimate=184772 kB
2020-04-15 05:26:54 MSK [1335]: [7668-1] db=,user=,app=,client= LOG:
database system is shut down

after power on and start OS
2020-04-15 07:49:08 MSK [947]: [3-1] db=,user=,app=,client= LOG:  received
smart shutdown request
2020-04-15 07:49:08 MSK [1524]: [1-1] db=,user=,app=,client= LOG:  database
system was shut down at 2020-04-15 05:26:54 MSK
2020-04-15 07:49:15 MSK [1524]: [2-1] db=,user=,app=,client= LOG:  MultiXact
member wraparound protections are now enabled
2020-04-15 07:49:15 MSK [2398]: [1-1] db=,user=,app=,client= LOG:  shutting
down
2020-04-15 07:49:16 MSK [2398]: [2-1] db=,user=,app=,client= LOG:
checkpoint starting: shutdown immediate
2020-04-15 07:49:16 MSK [2398]: [3-1] db=,user=,app=,client= LOG:
checkpoint complete: wrote 0 buffers (0.0%); 0 transaction log file(s)
added, 1 removed, 0 recycled; write=0.001 s, sync=0.000 s, total=0.779 s;
sync files=0, longest=0.000 s, average=0.000 s; distance=16384 kB,
estimate=16384 kB
2020-04-15 07:49:16 MSK [2398]: [4-1] db=,user=,app=,client= LOG:  database
system is shut down

after manual start all ok (Even if you just restart the server in any way -
auto start will happen correctly)
2020-04-15 09:27:49 MSK [21204]: [1-1] db=,user=,app=,client= LOG:  database
system was shut down at 2020-04-15 07:49:16 MSK
2020-04-15 09:27:49 MSK [21204]: [2-1] db=,user=,app=,client= LOG:
MultiXact member wraparound protections are now enabled
2020-04-15 09:27:49 MSK [21208]: [1-1] db=,user=,app=,client= LOG:
autovacuum launcher started
2020-04-15 09:27:49 MSK [21202]: [3-1] db=,user=,app=,client= LOG:  database
system is ready to accept connections
2020-04-15 09:27:50 MSK [21213]: [1-1]
db=[unknown],user=[unknown],app=[unknown],client=[local] LOG:  connection
received: host=[local]

PG Bug reporting form <noreply@postgresql.org> writes:
> On some servers, we began to encounter the "strange" behavior of
> postgresql.
> When shutting down the server at the command of UPS (shutdown now). The next
> time you turn on postgresql, it gets smart shutdown immediately at startup

Something is sending the server a SIGTERM signal.  This is not a
Postgres bug, it's something else in your system doing something
it shouldn't.

            regards, tom lane