On Wed, 14 Mar 2007 16:50:04 -0500, Ed Rabara <[EMAIL PROTECTED]>
wrote:
>
>Gil, your situation is a close match to mine. Unfortunately, my Servers is
>in Production status. We push a lot of files back and forth. Before this
>change, i.e., going to a more current AIX and faster P-series servers, we
>ran the MVSNFS Servers for weeks to months without recycling the NFS
>Servers.
>
>I have a PMR with IBM. If we find anything, I'll report back here.
>
>Ed
IBM opened APAR OA20339 to address my problem. Here's the description:
ERROR DESCRIPTION:
NFS Server allocates mounted mvs datasets causing an enqueue
on the dataset while being processed. We increment the
ref_count to show the file in use, however, there is an error
path taken which does not decrement the ref count. When this
happens, the enqueue on the dataset is never released until the
NFS server is cancelled and restarted.
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
