NFSv4.1: Fix the client behaviour on NFS4ERR_SEQ_FALSE_RETRY

If the server returns NFS4ERR_SEQ_FALSE_RETRY or NFS4ERR_RETRY_UNCACHED_REP,
then it thinks we're trying to replay an existing request. If so, then
let's just bump the sequence ID and retry the operation.

Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index c5c5d6c..ed45090 100644 (file)
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -775,6 +775,13 @@ static int nfs41_sequence_process(struct rpc_task *task,
                        slot->slot_nr,
                        slot->seq_nr);
                goto out_retry;
+       case -NFS4ERR_RETRY_UNCACHED_REP:
+       case -NFS4ERR_SEQ_FALSE_RETRY:
+               /*
+                * The server thinks we tried to replay a request.
+                * Retry the call after bumping the sequence ID.
+                */
+               goto retry_new_seq;
        case -NFS4ERR_BADSLOT:
                /*
                 * The slot id we used was probably retired. Try again
@@ -799,10 +806,6 @@ static int nfs41_sequence_process(struct rpc_task *task,
                        goto retry_nowait;
                }
                goto session_recover;
-       case -NFS4ERR_SEQ_FALSE_RETRY:
-               if (interrupted)
-                       goto retry_new_seq;
-               goto session_recover;
        default:
                /* Just update the slot sequence no. */
                slot->seq_done = 1;