From ae3ef964c1f75999499c2d35f7485aea1c0a70b1 Mon Sep 17 00:00:00 2001
From: antirez <antirez@gmail.com>
Date: Fri, 27 Sep 2019 11:39:40 +0200
Subject: [PATCH] Modules fork: improve SIGUSR1 handling, fix include.

We can't expect SIGUSR1 to have any specific value range, so let's
define an exit code that we can handle in a special way.
This also fixes an #include <wait.h> that is not standard.
---
 src/module.c |  2 +-
 src/server.c | 14 ++++++++++----
 src/server.h |  8 ++++++++
 3 files changed, 19 insertions(+), 5 deletions(-)

diff --git a/src/module.c b/src/module.c
index 854989e73..d3b37a3d8 100644
--- a/src/module.c
+++ b/src/module.c
@@ -31,7 +31,7 @@
 #include "cluster.h"
 #include "rdb.h"
 #include <dlfcn.h>
-#include <wait.h>
+#include <sys/wait.h>
 
 #define REDISMODULE_CORE 1
 #include "redismodule.h"
diff --git a/src/server.c b/src/server.c
index f38ed7897..046694e1f 100644
--- a/src/server.c
+++ b/src/server.c
@@ -1913,8 +1913,11 @@ int serverCron(struct aeEventLoop *eventLoop, long long id, void *clientData) {
             if (WIFSIGNALED(statloc)) bysignal = WTERMSIG(statloc);
 
             /* sigKillChildHandler catches the signal and calls exit(), but we
-             * must make sure not to flag lastbgsave_status, etc incorrectly. */
-            if (exitcode == SIGUSR1) {
+             * must make sure not to flag lastbgsave_status, etc incorrectly.
+             * We could directly terminate the child process via SIGUSR1
+             * without handling it, but in this case Valgrind will log an
+             * annoying error. */
+            if (exitcode == SERVER_CHILD_NOERROR_RETVAL) {
                 bysignal = SIGUSR1;
                 exitcode = 1;
             }
@@ -4618,11 +4621,14 @@ void setupSignalHandlers(void) {
     return;
 }
 
+/* This is the signal handler for children process. It is currently useful
+ * in order to track the SIGUSR1, that we send to a child in order to terminate
+ * it in a clean way, without the parent detecting an error and stop
+ * accepting writes because of a write error condition. */
 static void sigKillChildHandler(int sig) {
     UNUSED(sig);
-    /* this handler is needed to resolve a valgrind warning */
     serverLogFromHandler(LL_WARNING, "Received SIGUSR1 in child, exiting now.");
-    exitFromChild(SIGUSR1);
+    exitFromChild(SERVER_CHILD_NOERROR_RETVAL);
 }
 
 void setupChildSignalHandlers(void) {
diff --git a/src/server.h b/src/server.h
index d132cf09c..c701d6d21 100644
--- a/src/server.h
+++ b/src/server.h
@@ -179,6 +179,14 @@ typedef long long mstime_t; /* millisecond time type. */
 #define ACTIVE_EXPIRE_CYCLE_SLOW 0
 #define ACTIVE_EXPIRE_CYCLE_FAST 1
 
+/* Children process will exit with this status code to signal that the
+ * process terminated without an error: this is useful in order to kill
+ * a saving child (RDB or AOF one), without triggering in the parent the
+ * write protection that is normally turned on on write errors.
+ * Usually children that are terminated with SIGUSR1 will exit with this
+ * special code. */
+#define SERVER_CHILD_NOERROR_RETVAL    255
+
 /* Instantaneous metrics tracking. */
 #define STATS_METRIC_SAMPLES 16     /* Number of samples per metric. */
 #define STATS_METRIC_COMMAND 0      /* Number of commands executed. */