SUNRPC: Add a new error code for retry waiting for another binder
authorChuck Lever <chuck.lever@oracle.com>
Tue, 11 Sep 2007 22:00:41 +0000 (18:00 -0400)
committerTrond Myklebust <Trond.Myklebust@netapp.com>
Tue, 9 Oct 2007 21:16:53 +0000 (17:16 -0400)
Add new error code processing to the kernel's rpcbind client and to
call_bind_status() to distinguish two cases:

Case 1: the remote has replied that the program/version tuple is not
registered (returns -EACCES)

Case 2: another process is already in the middle of binding on this
transport (now returns -EAGAIN)

This change allows more specific retry processing for each of these two
cases.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
net/sunrpc/clnt.c
net/sunrpc/rpcb_clnt.c

index 7a7cd3b..e53d718 100644 (file)
@@ -950,6 +950,12 @@ call_bind_status(struct rpc_task *task)
        }
 
        switch (task->tk_status) {
+       case -EAGAIN:
+               dprintk("RPC: %5u rpcbind waiting for another request "
+                               "to finish\n", task->tk_pid);
+               /* avoid busy-waiting here -- could be a network outage. */
+               rpc_delay(task, 5*HZ);
+               goto retry_timeout;
        case -EACCES:
                dprintk("RPC: %5u remote rpcbind: RPC program/version "
                                "unavailable\n", task->tk_pid);
Simple merge