Results 1 to 1 of 1

Thread: Stale NFS v4 mount behavior

  1. #1
    Join Date
    Dec 2008
    Ubuntu Studio 12.04 Precise Pangolin

    Stale NFS v4 mount behavior

    No fix or work around for this? So we have devolved to windows behavior - reboot if things stop working?

    Post Edit: I think this is what's happening to me:

    I have a server that maintains a fairly constant mount from another server using NFS with ver=4 specfied in the mount parameters.

    /etc/export from the NFS server (32 bit 12.04 LTS if that makes any difference)
    /local_export	client_server_name(ro,sync,no_subtree_check)
    /etc/fstab on client (64 bit 12.04 LTS)
    server1:/local_export /local/mountpoint nfs ro,vers=4,,
    I occasionaly take down the Serve side of the NFS link for maintenance - did tonight to replace a failing data disk.

    The mountpoint on the client had gone stale -- in NFS v3, that meant a hung mount point as the retries went on .... but as root you could umount -f /local/mountpoint and recover, and remount when it became availble.

    Now what I'm seeing is the mount goes stale, the filessystem there goes away and things like 'df' no longer hang waiting on endless retries to complete -- that's great except I can no longer clear and remount.

    umount -f /local/mountpoint now returns this:

    # umount -f /local/mountpoint
    /local/mountpoint was not found in /proc/mounts
    /local/mountpoint was not found in /proc/mounts

    attempts to remount it look like this:
    mount -a
    # df
    df: `/local/mountpoint': Stale NFS file handle
    # mount server1:/local_export /local/mountpoint
    mount.nfs: Stale NFS file handle
    I see the mount in the local /etc/mtab ... How do I clear a stale NFS v4 mount so that I can remount without bouncing the client?
    Last edited by MakOwner; May 6th, 2013 at 10:48 PM. Reason: bump


Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts