Until yesterday, we expired passwords in 120 days.  122 days ago I
registered a bunch of client nodes.  Yesterday, our ADSM/VM server
brought VM's TCP/IP to it's knees.  The passwords for several nodes
had expired.  When ADSM's overnight schedule kicked in, it couldn't
complete the scheduled backup for these nodes.  (I presume the
schedules were retried several times).  By morning, our VM TCP/IP
machine was accepting no more connections from anyone.  Our TCP/IP
maintainer quickly saw the problem as hundreds of connections from the
ADSM/VM server in state FIN-WAIT2.

Those connections were dropped and all was well with our world again.
(We also saw this problem a few months ago when I added several nodes
to a schedule, but hadn't completed the client node installation.  I'd
hoped that fairly recent service installation would cure this, but alas
...)

Sorry this is a rather inexact explanation, but perhaps it will help
someone else avoid the problem or solve it more quickly.  (I have not
(otherwise) reported this to IBM).

cheers,

cheers,

Wayne T. Smith
Systems Group -- CAPS        internet: wts AT maine.maine DOT edu
University of Maine System   BITNET/CREN: WTS@MAINE