[Bug 260011] Unresponsive NFS mount on AWS EFS

https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=260011

Rick Macklem <rmacklem@FreeBSD.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
 Attachment #234129|0                           |1
        is obsolete|                            |

--- Comment #15 from Rick Macklem <rmacklem@FreeBSD.org> ---
Created attachment 234164
  --> https://bugs.freebsd.org/bugzilla/attachment.cgi?id=234164&action=edit
do not mark recovery needed after nfsrpc_renew() for BADSESSION

This is a much simpler patch that might fix the problem you saw.
I am not sure, but without this patch, the "NFSCLFLAGS_RECOVER"
would be set twice when nfsrpc_renew() gets a NFSERR_BADSESSION
error reply.  I still can't see how that would cause two recovery
cycles, but if it did, that would explain how a broken session
might not be marked "defunct".

The patch also adds a diagnostic printf() for the case where the
broken session is not marked "defunct".

The previous patch would not have been acceptable for a commit to
FreeBSD.

I will once more mention that "soft" mounts will never work correctly
for NFSv4.1.

-- 
You are receiving this mail because:
You are the assignee for the bug.