diff --git a/Documentation/Release-Notes/MaxScale-2.4.8-Release-Notes.md b/Documentation/Release-Notes/MaxScale-2.4.8-Release-Notes.md
index 97b0185a9..7872cad21 100644
--- a/Documentation/Release-Notes/MaxScale-2.4.8-Release-Notes.md
+++ b/Documentation/Release-Notes/MaxScale-2.4.8-Release-Notes.md
@@ -11,7 +11,6 @@ report on [our Jira](https://jira.mariadb.org/projects/MXS).
 ## Bug fixes
 
 * [MXS-2921](https://jira.mariadb.org/browse/MXS-2921) Memory leak in `alter maxscale`
-* [MXS-2920](https://jira.mariadb.org/browse/MXS-2920) Cannot set max_slave_replication_lag=0
 * [MXS-2919](https://jira.mariadb.org/browse/MXS-2919) max_slave_replication_lag uses servers with unknown replication lag
 * [MXS-2917](https://jira.mariadb.org/browse/MXS-2917) qc_sqlite leaks memory with complex CREATE TABLE query
 * [MXS-2907](https://jira.mariadb.org/browse/MXS-2907) Logrotate warnings when PID file does not exist
diff --git a/server/modules/routing/readwritesplit/readwritesplit.cc b/server/modules/routing/readwritesplit/readwritesplit.cc
index e4cc362aa..6811c13ca 100644
--- a/server/modules/routing/readwritesplit/readwritesplit.cc
+++ b/server/modules/routing/readwritesplit/readwritesplit.cc
@@ -502,12 +502,7 @@ extern "C" MXS_MODULE* MXS_CREATE_MODULE()
                 MXS_MODULE_OPT_NONE,
                 master_failure_mode_values
             },
-            {
-                "max_slave_replication_lag",
-                MXS_MODULE_PARAM_DURATION,
-                DEFAULT_MAX_SLAVE_REPLICATION_LAG_STR,
-                MXS_MODULE_OPT_DURATION_S
-            },
+            {"max_slave_replication_lag",  MXS_MODULE_PARAM_DURATION,  "0s",   MXS_MODULE_OPT_DURATION_S},
             {"max_slave_connections",      MXS_MODULE_PARAM_STRING,    MAX_SLAVE_COUNT},
             {"retry_failed_reads",         MXS_MODULE_PARAM_BOOL,      "true"},
             {"prune_sescmd_history",       MXS_MODULE_PARAM_BOOL,      "false"},
diff --git a/server/modules/routing/readwritesplit/readwritesplit.hh b/server/modules/routing/readwritesplit/readwritesplit.hh
index 4c23566a6..cde124550 100644
--- a/server/modules/routing/readwritesplit/readwritesplit.hh
+++ b/server/modules/routing/readwritesplit/readwritesplit.hh
@@ -124,9 +124,6 @@ static const char gtid_wait_stmt[] =
     "SET @maxscale_secret_variable=(SELECT CASE WHEN %s('%s', %s) = 0 "
     "THEN 1 ELSE (SELECT 1 FROM INFORMATION_SCHEMA.ENGINES) END);";
 
-static const char DEFAULT_MAX_SLAVE_REPLICATION_LAG_STR[] = "2147483647s";
-static const int DEFAULT_MAX_SLAVE_REPLICATION_LAG = 2147483647;
-
 /** Function that returns a "score" for a server to enable comparison.
  *  Smaller numbers are better.
  */
diff --git a/server/modules/routing/readwritesplit/rwsplit_route_stmt.cc b/server/modules/routing/readwritesplit/rwsplit_route_stmt.cc
index f39633cd4..3691c2396 100644
--- a/server/modules/routing/readwritesplit/rwsplit_route_stmt.cc
+++ b/server/modules/routing/readwritesplit/rwsplit_route_stmt.cc
@@ -735,7 +735,7 @@ int RWSplitSession::get_max_replication_lag()
     int conf_max_rlag = SERVER::RLAG_UNDEFINED;
 
     /** if there is no configured value, then longest possible int is used */
-    if (m_config.max_slave_replication_lag != DEFAULT_MAX_SLAVE_REPLICATION_LAG)
+    if (m_config.max_slave_replication_lag > 0)
     {
         conf_max_rlag = m_config.max_slave_replication_lag;
     }