Fix for a race in BGSAVE that may result in some data not being saved as soon as possible (when the configured saving triggers should fire). Also known as Issue 313, more details there in the google code issue. (backported from master)

1 files changed, 3 insertions, 1 deletions

diff --git a/redis.c b/redis.c
index 59f84f17d..fe6d88043 100644
--- a/redis.c
+++ b/redis.c
@@ -347,6 +347,7 @@ struct redisServer {
     int fd;
     redisDb *db;
     long long dirty;            /* changes to DB from the last save */
+    long long dirty_before_bgsave; /* used to restore dirty on failed BGSAVE */
     list *clients;
     list *slaves, *monitors;
     char neterr[ANET_ERR_LEN];
@@ -1320,7 +1321,7 @@ void backgroundSaveDoneHandler(int statloc) {
     if (!bysignal && exitcode == 0) {
         redisLog(REDIS_NOTICE,
             "Background saving terminated with success");
-        server.dirty = 0;
+        server.dirty = server.dirty - server.dirty_before_bgsave;
         server.lastsave = time(NULL);
     } else if (!bysignal && exitcode != 0) {
         redisLog(REDIS_WARNING, "Background saving error");
@@ -3790,6 +3791,7 @@ static int rdbSaveBackground(char *filename) {
 
     if (server.bgsavechildpid != -1) return REDIS_ERR;
     if (server.vm_enabled) waitEmptyIOJobsQueue();
+    server.dirty_before_bgsave = server.dirty;
     if ((childpid = fork()) == 0) {
         /* Child */
         if (server.vm_enabled) vmReopenSwapFile();