UCT/TCP: Increase keepalive interval from 1 to 2 seconds and use syst…

…em default for keepalive retries count
openucx · Jul 14, 2021 · 7b3112d · 7b3112d
1 parent 8d5552e
commit 7b3112d
Show file tree

Hide file tree

Showing 3 changed files with 13 additions and 11 deletions.
diff --git a/src/uct/tcp/tcp.h b/src/uct/tcp/tcp.h
@@ -65,7 +65,7 @@
 #define UCT_TCP_EP_DEFAULT_KEEPALIVE_IDLE    10
 
 /* The seconds between individual keepalive probes */
-#define UCT_TCP_EP_DEFAULT_KEEPALIVE_INTVL   1
+#define UCT_TCP_EP_DEFAULT_KEEPALIVE_INTVL   2
 
 
 /**
@@ -402,7 +402,7 @@ typedef struct uct_tcp_iface {
             ucs_time_t            idle;              /* The time the connection needs to remain
                                                       * idle before TCP starts sending keepalive
                                                       * probes (TCP_KEEPIDLE socket option) */
-            unsigned              cnt;               /* The maximum number of keepalive probes TCP
+            size_t                cnt;               /* The maximum number of keepalive probes TCP
                                                       * should send before dropping the connection
                                                       * (TCP_KEEPCNT socket option). */
             ucs_time_t            intvl;             /* The time between individual keepalive
@@ -440,7 +440,7 @@ typedef struct uct_tcp_iface_config {
     ucs_range_spec_t               port_range;
     struct {
         ucs_time_t                 idle;
-        unsigned                   cnt;
+        size_t                     cnt;
         ucs_time_t                 intvl;
     } keepalive;
 } uct_tcp_iface_config_t;

diff --git a/src/uct/tcp/tcp_ep.c b/src/uct/tcp/tcp_ep.c
@@ -526,11 +526,13 @@ static ucs_status_t uct_tcp_ep_keepalive_enable(uct_tcp_ep_t *ep)
         return status;
     }
 
-    status = ucs_socket_setopt(ep->fd, IPPROTO_TCP, TCP_KEEPCNT,
-                               &iface->config.keepalive.cnt,
-                               sizeof(iface->config.keepalive.cnt));
-    if (status != UCS_OK) {
-        return status;
+    if (iface->config.keepalive.cnt != UCS_ULUNITS_AUTO) {
+        status = ucs_socket_setopt(ep->fd, IPPROTO_TCP, TCP_KEEPCNT,
+                                   &iface->config.keepalive.cnt,
+                                   sizeof(iface->config.keepalive.cnt));
+        if (status != UCS_OK) {
+            return status;
+        }
     }
 
     status = ucs_socket_setopt(ep->fd, IPPROTO_TCP, TCP_KEEPIDLE,

diff --git a/src/uct/tcp/tcp_iface.c b/src/uct/tcp/tcp_iface.c
@@ -93,11 +93,11 @@ static ucs_config_field_t uct_tcp_iface_config_table[] = {
    ucs_offsetof(uct_tcp_iface_config_t, keepalive.idle),
                 UCS_CONFIG_TYPE_TIME_UNITS},
 
-  {"KEEPCNT", "3",
+  {"KEEPCNT", "auto",
    "The maximum number of keepalive probes TCP should send before "
    "dropping the connection.",
    ucs_offsetof(uct_tcp_iface_config_t, keepalive.cnt),
-                UCS_CONFIG_TYPE_UINT},
+                UCS_CONFIG_TYPE_ULUNITS},
 
   {"KEEPINTVL", UCS_PP_MAKE_STRING(UCT_TCP_EP_DEFAULT_KEEPALIVE_INTVL) "s",
    "The time between individual keepalive probes.",
@@ -872,7 +872,7 @@ int uct_tcp_keepalive_is_enabled(uct_tcp_iface_t *iface)
 {
 #ifdef UCT_TCP_EP_KEEPALIVE
     return (iface->config.keepalive.idle != UCS_TIME_INFINITY) &&
-           (iface->config.keepalive.cnt != 0) &&
+           (iface->config.keepalive.cnt != UCS_ULUNITS_INF) &&
            (iface->config.keepalive.intvl != UCS_TIME_INFINITY);
 #else /* UCT_TCP_EP_KEEPALIVE */
     return 0;