Tone down shutdown messages for socket activated responders

Signed-off-by: Fabiano Fidêncio <fidencio@redhat.com>

0019-SERVER-Tone-down-shutdown-messages-for-socket-activa.patch

@@ -0,0 +1,46 @@
+From 778f7c61b8d55e0b8d8eccd2cf8649d730e7d4a5 Mon Sep 17 00:00:00 2001
+From: =?UTF-8?q?Fabiano=20Fid=C3=AAncio?= <fidencio@redhat.com>
+Date: Tue, 3 Apr 2018 21:43:28 +0200
+Subject: [PATCH] SERVER: Tone down shutdown messages for socket-activated
+ responders
+MIME-Version: 1.0
+Content-Type: text/plain; charset=UTF-8
+Content-Transfer-Encoding: 8bit
+
+When dealing with socket-activated responders, those may be shut
+themselves down after some inactivy period. And that's completely normal
+and expected, thus should not be logged as an fatal error.
+
+For the case when the responder is started by the monitor, however, it
+still makes sense to keep the code as it is as the responders won't shut
+themselves down in any normal scenario.
+
+Signed-off-by: Fabiano Fidêncio <fidencio@redhat.com>
+
+Reviewed-by: Jakub Hrozek <jhrozek@redhat.com>
+(cherry picked from commit 519354d079731e673244a8e3851e5c5522d1b45e)
+---
+ src/util/server.c | 6 +++++-
+ 1 file changed, 5 insertions(+), 1 deletion(-)
+
+diff --git a/src/util/server.c b/src/util/server.c
+index 62e09314c..f34bf49f6 100644
+--- a/src/util/server.c
++++ b/src/util/server.c
+@@ -248,8 +248,12 @@ void orderly_shutdown(int status)
+ {
+ #if HAVE_GETPGRP
+     static int sent_sigterm;
++    int debug;
++
+     if (sent_sigterm == 0 && getpgrp() == getpid()) {
+-        DEBUG(SSSDBG_FATAL_FAILURE, "SIGTERM: killing children\n");
++        debug = is_socket_activated() ? SSSDBG_TRACE_INTERNAL
++                                      : SSSDBG_FATAL_FAILURE;
++        DEBUG(debug, "SIGTERM: killing children\n");
+         sent_sigterm = 1;
+         kill(-getpgrp(), SIGTERM);
+     }
+-- 
+2.14.3
+

sssd.spec

@@ -60,6 +60,7 @@ Patch0015: 0015-KCM-Only-print-the-number-of-found-items-after-we-ha.patch
 Patch0016: 0016-SYSDB-When-marking-an-entry-as-expired-also-set-the-.patch
 Patch0017: 0017-sudo-ldap-do-not-store-rules-without-sudoHost-attrib.patch
 Patch0018: 0018-sysdb-custom-completely-replace-old-object-instead-o.patch
+Patch0019: 0019-SERVER-Tone-down-shutdown-messages-for-socket-activa.patch
 
 Patch0502: 0502-SYSTEMD-Use-capabilities.patch
 Patch0503: 0503-Disable-stopping-idle-socket-activated-responders.patch
@@ -1265,6 +1266,7 @@ fi
                             with the cleanup task, but the group does not
                             change
 - Resolves: upstream#3558 - sudo: report error when two rules share cn
+- Tone down shutdown messages for socket activated responders
 
 * Fri Mar 30 2018 Fabiano Fidêncio <fidencio@fedoraproject.org> - 1.16.1-2
 - Resolves: upstream#3573 - sssd won't show netgroups with blank domain