From: dzogin <dzogin>
Date: Mon, 21 Sep 2009 15:54:05 +0000 (+0000)
Subject:  Branch b1_8
X-Git-Tag: v1_8_2_01~1^2~85
X-Git-Url: https://git.whamcloud.com/?a=commitdiff_plain;h=7e151969bc93cccd46372da1fd3518e81d4489bc;p=fs%2Flustre-release.git

 Branch b1_8
 b=18674
 i=johann
 i=andrew.perepechko
----------------------------------------------------------------------
 Modified Files:
  Tag: b1_8
 	lustre/ChangeLog lustre/include/lustre_import.h
 	lustre/ptlrpc/import.c
----------------------------------------------------------------------
Details    : The client also now retries to reconnect to the same server if a
	     connect request failed with EBUSY or -EAGAIN. The patch has not
	     been landed in the previous commits.
---

diff --git a/lustre/ChangeLog b/lustre/ChangeLog
index 3ebd312..7cc1ca7 100644
--- a/lustre/ChangeLog
+++ b/lustre/ChangeLog
@@ -16,6 +16,14 @@ tbd Sun Microsystems, Inc.
 	  more information, please refer to bugzilla 17630.
 
 Severity   : normal
+Bugzilla   : 18674
+Frequency  : occasionally during network problems
+Description: client not allowed to reconnect to OST because of active request
+Details    : The client also now retries to reconnect to the same server if a 
+	     connect request failed with EBUSY or -EAGAIN. The patch has not
+	     been landed in the previous commits.
+
+Severity   : normal
 Bugzilla   : 19526
 Description: can't stat file in some situation.
 Details    : improve initialize osc date when target is added to mds and 
diff --git a/lustre/include/lustre_import.h b/lustre/include/lustre_import.h
index 8fdb36e..f1e3799 100644
--- a/lustre/include/lustre_import.h
+++ b/lustre/include/lustre_import.h
@@ -176,7 +176,8 @@ struct obd_import {
                                   imp_pingable:1,         /* pingable */
                                   imp_resend_replay:1,    /* resend for replay */
                                   imp_recon_bk:1,         /* turn off reconnect if all failovers fail */
-                                  imp_last_recon:1;       /* internally used by above */
+                                  imp_last_recon:1,       /* internally used by above */
+                                  imp_force_reconnect:1;  /* import must be reconnected instead of chouse new connection */
         __u32                     imp_connect_op;
         struct obd_connect_data   imp_connect_data;
         __u64                     imp_connect_flags_orig;
diff --git a/lustre/ptlrpc/import.c b/lustre/ptlrpc/import.c
index 8ea8799..c4d4222 100644
--- a/lustre/ptlrpc/import.c
+++ b/lustre/ptlrpc/import.c
@@ -482,7 +482,7 @@ static int import_select_connection(struct obd_import *imp)
         }
 
         /* if not found, simply choose the current one */
-        if (!imp_conn) {
+        if (!imp_conn || imp->imp_force_reconnect) {
                 LASSERT(imp->imp_conn_current);
                 imp_conn = imp->imp_conn_current;
                 tried_all = 0;
@@ -761,6 +761,11 @@ static void ptlrpc_maybe_ping_import_soon(struct obd_import *imp)
         EXIT;
 }
 
+static int ptlrpc_busy_reconnect(int rc)
+{
+        return (rc == -EBUSY) || (rc == -EAGAIN);
+}
+
 static int ptlrpc_connect_interpret(struct ptlrpc_request *request,
                                     void * data, int rc)
 {
@@ -777,18 +782,23 @@ static int ptlrpc_connect_interpret(struct ptlrpc_request *request,
                 spin_unlock(&imp->imp_lock);
                 RETURN(0);
         }
-        spin_unlock(&imp->imp_lock);
 
-        if (rc)
+        if (rc) {
+                /* if this reconnect to busy export - not need select new target
+                 * for connecting*/
+                if (ptlrpc_busy_reconnect(rc))
+                        imp->imp_force_reconnect = 1;
+                spin_unlock(&imp->imp_lock);
                 GOTO(out, rc);
+        }
 
         LASSERT(imp->imp_conn_current);
 
         msg_flags = lustre_msg_get_op_flags(request->rq_repmsg);
 
         /* All imports are pingable */
-        spin_lock(&imp->imp_lock);
         imp->imp_pingable = 1;
+        imp->imp_force_reconnect = 0;
 
         if (aa->pcaa_initial_connect) {
                 if (msg_flags & MSG_CONNECT_REPLAYABLE) {