pgsql: Avoid calling proc_exit() in processes forked by system(). - Mailing list pgsql-committers

From Nathan Bossart
Subject pgsql: Avoid calling proc_exit() in processes forked by system().
Date
Msg-id E1qsmEz-0011HW-Lu@gemulon.postgresql.org
Whole thread Raw
List pgsql-committers
Avoid calling proc_exit() in processes forked by system().

The SIGTERM handler for the startup process immediately calls
proc_exit() for the duration of the restore_command, i.e., a call
to system().  This system() call forks a new process to execute the
shell command, and this child process inherits the parent's signal
handlers.  If both the parent and child processes receive SIGTERM,
both will attempt to call proc_exit().  This can end badly.  For
example, both processes will try to remove themselves from the
PGPROC shared array.

To fix this problem, this commit adds a check in
StartupProcShutdownHandler() to see whether MyProcPid == getpid().
If they match, this is the parent process, and we can proc_exit()
like before.  If they do not match, this is a child process, and we
just emit a message to STDERR (in a signal safe manner) and
_exit(), thereby skipping any problematic exit callbacks.

This commit also adds checks in proc_exit(), ProcKill(), and
AuxiliaryProcKill() that verify they are not being called within
such child processes.

Suggested-by: Andres Freund
Reviewed-by: Thomas Munro, Andres Freund
Discussion: https://postgr.es/m/Y9nGDSgIm83FHcad%40paquier.xyz
Discussion: https://postgr.es/m/20230223231503.GA743455%40nathanxps13
Backpatch-through: 11

Branch
------
REL_14_STABLE

Details
-------
https://git.postgresql.org/pg/commitdiff/54fc9dca5b10e82cb85c7cd44ebe1eb0f28f795b

Modified Files
--------------
src/backend/postmaster/startup.c | 17 ++++++++++++++++-
src/backend/storage/ipc/ipc.c    |  4 ++++
src/backend/storage/lmgr/proc.c  |  8 ++++++++
src/backend/utils/error/elog.c   | 28 ++++++++++++++++++++++++++++
src/include/utils/elog.h         |  6 ++++++
5 files changed, 62 insertions(+), 1 deletion(-)


pgsql-committers by date:

Previous
From: Robert Haas
Date:
Subject: pgsql: Reword messages about impending (M)XID exhaustion.
Next
From: Tom Lane
Date:
Subject: pgsql: Dodge a compiler bug affecting timetz_zone/timetz_izone.