Merge branch 'maint', 4.4.2 release

Author: hedenface

Changelog

@@ -2,6 +2,20 @@
 Nagios Core 4 Change Log
 ########################
 
+4.4.2 - 2018-08-16
+------------------
+FIXES
+* Fix comment data being duplicated after a `service nagios reload` or similar (#549) (Bryan Heden)
+* Fix check_interval and retry_interval not changing at the appropriate times (#551) (Scott Wilkerson)
+* Fixed passive checks sending recovery email when host was previously UP (#552) (Scott Wilkerson)
+* Fixed flapping comments duplication on nagios reload (#554) (Christian Jung)
+* Fix for CVE-2018-13441, CVE-2018-13458, CVE-2018-13457 null pointer dereference (Trevor McDonald)
+* Fixed syntax error in file: default-init.in (#558) (Christian Zettel)
+* Reset current notification number and state flags when the host recovers, reset all service variables when they recover fixes (#557) (Scott Wilkerson)
+* Fixed wrong counting of service status totals when showing servicegroup details (#548) (Christian Zettel, Bryan Heden)
+* Fixed avail.cgi not printing CSV output when checkbox is checked (for any type: host/service/hostgroup/servicegroup) (#570) (Bryan Heden)
+* Fixed nagios not logging SOFT state changes after 1 (Scott Wilkerson)
+
 4.4.1 - 2018-06-25
 ------------------
 FIXES

THANKS

@@ -54,6 +54,7 @@ wrong, please let me know.
 * Chris Kolquist
 * Chris Rothecker
 * Chris Witterholt
+* Christian Jung
 * Christian Masopust
 * Christian Mies
 * Christian Zettel

base/checks.c

@@ -893,6 +893,9 @@ static inline void service_state_or_hard_state_type_change(service * svc, int st
 		if (svc->current_state == STATE_OK) {
 			svc->last_problem_id = svc->current_problem_id;
 			svc->current_problem_id = 0L;
+            svc->current_attempt = 1;
+            svc->current_notification_number = 0;
+            svc->host_problem_at_last_check = FALSE;
 		}
 
 		svc->state_type = SOFT_STATE;
@@ -911,6 +914,11 @@ static inline void service_state_or_hard_state_type_change(service * svc, int st
 
 	if (state_or_type_change) {
 
+		/* check if service should go into downtime from flexible downtime */
+		if (svc->pending_flex_downtime > 0) {
+			check_pending_flex_service_downtime(svc);
+		}
+
 		/* reset notification times and suppression option */
 		svc->last_notification = (time_t)0;
 		svc->next_notification = (time_t)0;
@@ -941,8 +949,12 @@ static inline void host_state_or_hard_state_type_change(host * hst, int state_ch
 
 		log_debug_info(DEBUGL_CHECKS, 2, "Check type passive and passive host checks aren't false\n");
 
-		hst->current_attempt = 1;
-		hard_state_change = TRUE;
+		if (state_change == TRUE) {
+            hst->current_attempt = 1;
+            hard_state_change = TRUE;
+        }
+        
+		hst->state_type = HARD_STATE;
 	}
 
 	/* update event and problem ids */
@@ -989,6 +1001,9 @@ static inline void host_state_or_hard_state_type_change(host * hst, int state_ch
 
 	if (state_or_type_change) {
 
+		/* check if host should go into downtime from flexible downtime */
+		check_pending_flex_host_downtime(hst);
+
 		/* reset notification times and suppression option */
 		hst->last_notification = (time_t)0;
 		hst->next_notification = (time_t)0;
@@ -1228,7 +1243,7 @@ int handle_async_service_check_result(service *svc, check_result *cr)
 	next_check = (time_t)(svc->last_check + (svc->check_interval * interval_length));
 
 	/***********************************************/
-	/********** SCHEDULE HOST CHECK LOGIC **********/
+	/********** SCHEDULE SERVICE CHECK LOGIC **********/
 	/***********************************************/
 	if (svc->current_state == STATE_OK) {
 
@@ -1269,6 +1284,7 @@ int handle_async_service_check_result(service *svc, check_result *cr)
 
 			svc->host_problem_at_last_check = TRUE;
 		}
+        
 	}
 	else {
 
@@ -1368,6 +1384,9 @@ int handle_async_service_check_result(service *svc, check_result *cr)
 		else {
 
 			log_debug_info(DEBUGL_CHECKS, 1, "Service is a non-OK state (%s)!", service_state_name(svc->current_state));
+            
+            svc->state_type = SOFT_STATE;
+			svc->current_attempt = 1;
 
 			handle_event = TRUE;
 		}
@@ -1395,6 +1414,12 @@ int handle_async_service_check_result(service *svc, check_result *cr)
 
 				log_debug_info(DEBUGL_CHECKS, 1, "Service experienced a SOFT recovery.\n");				
 			}
+            
+            
+            /* reset all service variables because its ok now... */
+            svc->state_type = HARD_STATE;
+            state_change = TRUE;
+            hard_state_change = TRUE;
 		}
 
 		/***** SERVICE IS STILL IN PROBLEM STATE *****/
@@ -1418,6 +1443,14 @@ int handle_async_service_check_result(service *svc, check_result *cr)
 			}
 		}
 	}
+    
+    /* soft states should be using retry_interval */
+    if (svc->state_type == SOFT_STATE) {
+        
+		log_debug_info(DEBUGL_CHECKS, 2, "Service state type is soft, using retry_interval\n");
+
+		next_check = (unsigned long) (current_time + svc->retry_interval * interval_length);
+    }
 
 	/* check for a state change */
 	if (svc->current_state != svc->last_state || (svc->current_state == STATE_OK && svc->state_type == SOFT_STATE)) {
@@ -1454,6 +1487,8 @@ int handle_async_service_check_result(service *svc, check_result *cr)
 	if (svc->current_attempt >= svc->max_attempts && svc->current_state != svc->last_hard_state) {
 
 		log_debug_info(DEBUGL_CHECKS, 2, "Service had a HARD STATE CHANGE!!\n");
+        
+        next_check = (unsigned long)(current_time + (svc->check_interval * interval_length));
 
 		hard_state_change = TRUE;
 
@@ -1464,6 +1499,11 @@ int handle_async_service_check_result(service *svc, check_result *cr)
 	/* handle some acknowledgement things and update last_state_change */
 	service_state_or_hard_state_type_change(svc, state_change, hard_state_change, &log_event, &handle_event);
 
+	/* fix edge cases where log_event wouldn't have been set or won't be */
+	if (svc->current_state != STATE_OK && svc->state_type == SOFT_STATE) {
+		log_event = TRUE;
+	}
+
 	record_last_service_state_ended(svc);
 
 	check_for_service_flapping(svc, TRUE, TRUE);
@@ -2197,6 +2237,9 @@ int handle_async_host_check_result(host *hst, check_result *cr)
 		else {
 
 			log_debug_info(DEBUGL_CHECKS, 1, "Host is no longer UP (%s)!\n", host_state_name(hst->current_state));
+            
+            hst->state_type = SOFT_STATE;
+            hst->current_attempt = 1;
 
 			/* propagate checks to immediate parents if they are UP */
 			host_propagate_checks_to_immediate_parents(hst, FALSE, current_time);
@@ -2276,7 +2319,9 @@ int handle_async_host_check_result(host *hst, check_result *cr)
 	if (hst->current_state != HOST_UP && (hst->check_type == CHECK_TYPE_ACTIVE || translate_passive_host_checks == TRUE)) {
 
 		hst->current_state = determine_host_reachability(hst);
-		next_check = (unsigned long)(current_time + (hst->retry_interval * interval_length));
+		if (hst->state_type == SOFT_STATE)
+            next_check = (unsigned long)(current_time + (hst->retry_interval * interval_length));
+        
 	}
 
 	/* check for state change */
@@ -2310,7 +2355,9 @@ int handle_async_host_check_result(host *hst, check_result *cr)
 
 		log_debug_info(DEBUGL_CHECKS, 2, "Host had a HARD STATE CHANGE!!\n");
 
-		hard_state_change = TRUE;
+		next_check = (unsigned long)(current_time + (hst->check_interval * interval_length));
+        
+        hard_state_change = TRUE;
 		send_notification = TRUE;
 	}
 
@@ -2372,6 +2419,12 @@ int handle_async_host_check_result(host *hst, check_result *cr)
 		}
 	}
 
+    /* the host recovered, so reset the current notification number and state flags (after the recovery notification has gone out) */
+    if(hst->current_state == HOST_UP && hst->state_type == HARD_STATE && hard_state_change == TRUE) {
+        hst->current_notification_number = 0;
+        hst->notified_on = 0;
+        }
+        
 	if (obsess_over_hosts == TRUE) {
 		obsessive_compulsive_host_check_processor(hst);
 	}

base/flapping.c

@@ -308,10 +308,12 @@ void set_service_flap(service *svc, double percent_change, double high_threshold
 	/* log a notice - this one is parsed by the history CGI */
 	logit(NSLOG_RUNTIME_WARNING, FALSE, "SERVICE FLAPPING ALERT: %s;%s;STARTED; Service appears to have started flapping (%2.1f%% change >= %2.1f%% threshold)\n", svc->host_name, svc->description, percent_change, high_threshold);
 
-	/* add a non-persistent comment to the service */
-	asprintf(&temp_buffer, "Notifications for this service are being suppressed because it was detected as having been flapping between different states (%2.1f%% change >= %2.1f%% threshold).  When the service state stabilizes and the flapping stops, notifications will be re-enabled.", percent_change, high_threshold);
-	add_new_service_comment(FLAPPING_COMMENT, svc->host_name, svc->description, time(NULL), "(Nagios Process)", temp_buffer, 0, COMMENTSOURCE_INTERNAL, FALSE, (time_t)0, &(svc->flapping_comment_id));
-	my_free(temp_buffer);
+	if (svc->flapping_comment_id == 0) {
+		/* add a non-persistent comment to the service */
+		asprintf(&temp_buffer, "Notifications for this service are being suppressed because it was detected as having been flapping between different states (%2.1f%% change >= %2.1f%% threshold).  When the service state stabilizes and the flapping stops, notifications will be re-enabled.", percent_change, high_threshold);
+		add_new_service_comment(FLAPPING_COMMENT, svc->host_name, svc->description, time(NULL), "(Nagios Process)", temp_buffer, 0, COMMENTSOURCE_INTERNAL, FALSE, (time_t)0, &(svc->flapping_comment_id));
+		my_free(temp_buffer);
+		}
 
 	/* set the flapping indicator */
 	svc->is_flapping = TRUE;
@@ -396,10 +398,12 @@ void set_host_flap(host *hst, double percent_change, double high_threshold, doub
 	/* log a notice - this one is parsed by the history CGI */
 	logit(NSLOG_RUNTIME_WARNING, FALSE, "HOST FLAPPING ALERT: %s;STARTED; Host appears to have started flapping (%2.1f%% change > %2.1f%% threshold)\n", hst->name, percent_change, high_threshold);
 
-	/* add a non-persistent comment to the host */
-	asprintf(&temp_buffer, "Notifications for this host are being suppressed because it was detected as having been flapping between different states (%2.1f%% change > %2.1f%% threshold).  When the host state stabilizes and the flapping stops, notifications will be re-enabled.", percent_change, high_threshold);
-	add_new_host_comment(FLAPPING_COMMENT, hst->name, time(NULL), "(Nagios Process)", temp_buffer, 0, COMMENTSOURCE_INTERNAL, FALSE, (time_t)0, &(hst->flapping_comment_id));
-	my_free(temp_buffer);
+	if (hst->flapping_comment_id == 0) {
+		/* add a non-persistent comment to the host */
+		asprintf(&temp_buffer, "Notifications for this host are being suppressed because it was detected as having been flapping between different states (%2.1f%% change > %2.1f%% threshold).  When the host state stabilizes and the flapping stops, notifications will be re-enabled.", percent_change, high_threshold);
+		add_new_host_comment(FLAPPING_COMMENT, hst->name, time(NULL), "(Nagios Process)", temp_buffer, 0, COMMENTSOURCE_INTERNAL, FALSE, (time_t)0, &(hst->flapping_comment_id));
+		my_free(temp_buffer);
+	}
 
 	/* set the flapping indicator */
 	hst->is_flapping = TRUE;

base/nagios.c

@@ -878,6 +878,9 @@ int main(int argc, char **argv) {
 			/* clean up the scheduled downtime data */
 			cleanup_downtime_data();
 
+			/* clean up comment data */
+			free_comment_data();
+
 			/* clean up the status data if we are not restarting */
 			if(sigrestart == FALSE) {
 				cleanup_status_data(TRUE);

base/notifications.c

@@ -567,29 +567,6 @@ int check_service_notification_viability(service *svc, int type, int options) {
 		return ERROR;
 		}
 
-	/***** RECOVERY NOTIFICATIONS ARE GOOD TO GO AT THIS POINT *****/
-	if(svc->current_state == STATE_OK)
-		return OK;
-
-	/* don't notify contacts about this service problem again if the notification interval is set to 0 */
-	if(svc->no_more_notifications == TRUE) {
-		log_debug_info(DEBUGL_NOTIFICATIONS, 1, "We shouldn't re-notify contacts about this service problem.\n");
-		return ERROR;
-		}
-
-	/* if the host is down or unreachable, don't notify contacts about service failures */
-	if(temp_host->current_state != STATE_UP && temp_host->state_type == HARD_STATE) {
-		log_debug_info(DEBUGL_NOTIFICATIONS, 1, "The host is either down or unreachable, so we won't notify contacts about this service.\n");
-		return ERROR;
-		}
-
-	/* don't notify if we haven't waited long enough since the last time (and the service is not marked as being volatile) */
-	if((current_time < svc->next_notification) && svc->is_volatile == FALSE) {
-		log_debug_info(DEBUGL_NOTIFICATIONS, 1, "We haven't waited long enough to re-notify contacts about this service.\n");
-		log_debug_info(DEBUGL_NOTIFICATIONS, 1, "Next valid notification time: %s", ctime(&svc->next_notification));
-		return ERROR;
-		}
-
 	/* if this service is currently in a scheduled downtime period, don't send the notification */
 	if(svc->scheduled_downtime_depth > 0) {
 		log_debug_info(DEBUGL_NOTIFICATIONS, 1, "This service is currently in a scheduled downtime, so we won't send notifications.\n");
@@ -614,6 +591,29 @@ int check_service_notification_viability(service *svc, int type, int options) {
 		return ERROR;
 		}
 
+	/***** RECOVERY NOTIFICATIONS ARE GOOD TO GO AT THIS POINT *****/
+	if(svc->current_state == STATE_OK)
+		return OK;
+
+	/* don't notify contacts about this service problem again if the notification interval is set to 0 */
+	if(svc->no_more_notifications == TRUE) {
+		log_debug_info(DEBUGL_NOTIFICATIONS, 1, "We shouldn't re-notify contacts about this service problem.\n");
+		return ERROR;
+		}
+
+	/* if the host is down or unreachable, don't notify contacts about service failures */
+	if(temp_host->current_state != STATE_UP && temp_host->state_type == HARD_STATE) {
+		log_debug_info(DEBUGL_NOTIFICATIONS, 1, "The host is either down or unreachable, so we won't notify contacts about this service.\n");
+		return ERROR;
+		}
+
+	/* don't notify if we haven't waited long enough since the last time (and the service is not marked as being volatile) */
+	if((current_time < svc->next_notification) && svc->is_volatile == FALSE) {
+		log_debug_info(DEBUGL_NOTIFICATIONS, 1, "We haven't waited long enough to re-notify contacts about this service.\n");
+		log_debug_info(DEBUGL_NOTIFICATIONS, 1, "Next valid notification time: %s", ctime(&svc->next_notification));
+		return ERROR;
+		}
+
 	return OK;
 	}
 
@@ -1495,16 +1495,16 @@ int check_host_notification_viability(host *hst, int type, int options) {
 		return ERROR;
 		}
 
-	/***** RECOVERY NOTIFICATIONS ARE GOOD TO GO AT THIS POINT *****/
-	if(hst->current_state == HOST_UP)
-		return OK;
-
 	/* if this host is currently in a scheduled downtime period, don't send the notification */
 	if(hst->scheduled_downtime_depth > 0) {
 		log_debug_info(DEBUGL_NOTIFICATIONS, 1, "This host is currently in a scheduled downtime, so we won't send notifications.\n");
 		return ERROR;
 		}
 
+	/***** RECOVERY NOTIFICATIONS ARE GOOD TO GO AT THIS POINT *****/
+	if(hst->current_state == HOST_UP)
+		return OK;
+
 	/* check if we shouldn't renotify contacts about the host problem */
 	if(hst->no_more_notifications == TRUE) {
 		log_debug_info(DEBUGL_NOTIFICATIONS, 1, "We shouldn't re-notify contacts about this host problem.\n");

base/query-handler.c

@@ -26,7 +26,7 @@ static int qh_echo(int sd, char *buf, unsigned int len)
 {
 	int result = 0;
 
-	if (!strcmp(buf, "help")) {
+	if (buf == NULL || !strcmp(buf, "help")) {
 
 		nsock_printf_nul(sd,
 			"Query handler that simply echoes back what you send it.");
@@ -371,7 +371,7 @@ static int qh_help(int sd, char *buf, unsigned int len)
 {
 	struct query_handler *qh = NULL;
 
-	if (!*buf || !strcmp(buf, "help")) {
+	if (buf == NULL || !strcmp(buf, "help")) {
 		nsock_printf_nul(sd,
 			"  help <name>   show help for handler <name>\n"
 			"  help list     list registered handlers\n");
@@ -405,7 +405,7 @@ static int qh_core(int sd, char *buf, unsigned int len)
 {
 	char *space;
 
-	if (*buf == 0 || !strcmp(buf, "help")) {
+	if (buf == NULL || !strcmp(buf, "help")) {
 
 		nsock_printf_nul(sd, 
 			"Query handler for manipulating nagios core.\n"