Received-SPF: pass (google.com: best guess record for domain of linux-nfs-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
MIME-Version: 1.0
From:   Gefei Li <gefeili.2013@gmail.com>
Date:   Sat, 28 Sep 2019 00:56:34 +0800
Message-ID: <CANidX5SdiM3fZ+A1N_Z4iNhMU8KSvSHqwRXciyX32ZqDzLH32w@mail.gmail.com>
Subject: Did NFS client hang all OPEN request after SEQ4_STATUS_EXPIRED_SOME_STATE_REVOKED
 flag is set?
To:     linux-nfs@vger.kernel.org
Cc:     "J. Bruce Fields" <bfields@fieldses.org>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-nfs-owner@vger.kernel.org
Precedence: bulk

Hi Kernel NFS devs,

From RFC 5661, I know that the flag
SEQ4_STATUS_EXPIRED_SOME_STATE_REVOKED may be a "time-consuming" flag
for client, when received this flag, client needs to check all
existing lock/state.

   SEQ4_STATUS_EXPIRED_SOME_STATE_REVOKED
      When set, indicates that some subset of the client's locks have
      been revoked due to expiration of the lease period followed by
      another client's conflicting LOCK operation.  This status bit
      remains set on all SEQUENCE replies until the loss of all such
      locks has been acknowledged by use of FREE_STATEID.

My question is: when client received this flag, did it hang all
incoming "open" syscall on NFS files before the lock/state check
finished?

The problem I met is that my linux VM(with kernel 5.2.11) gives me a
really slow response(sometimes after an hour) after my NFS
traffic(which is heavy) runs for a while, I want to know the root
cause for this slow syscall, and from some packet trace I can get this
flag in some response of SEQUENCE response.

Appreciate your help and thanks in advance.

Thanks,
Gefei