[SUMMARY] Patch OSF350-048 & CDE

From: Tim Mooney <mooney_at_dogbert.cc.ndsu.NoDak.edu>
Date: Tue, 30 Jan 1996 16:26:33 -0600 (CST)

All-

This is the summary regarding problems I had with the CDE under DU 3.2c,
after installing some patches on 13-Dec-95 (sorry it took so long to get this
to you, but I've just gotten final comfirmation of the problem).

As indicated in the original message (which is attached below, to refresh
your memory...) after applying a number of patches CDE ceased to function.
I suspected patch OSF350-048 -- that was *not* the problem.

The problem is with patch OSF350-004. Installing this patch will cause CDE
to quit functioning under 3.2c -- in addition this patch has been rolled into
3.2d; hence the recent message to this list from Dirk Grunwald
(grunwald_at_foobar.cs.colorado.edu) indicating the fact that the CDE doesn't work
under 3.2d.

Thanks to Jeff Ford of Digital's Atlanta CSC for his help in reproducing this
problem and especially for keeping me informed as he investigated it.

With DU 4.0 due in 2 months I will forego the `upgrade' to 3.2d and stay at
3.2c until I can move into the fully supported CDE environment.

Again, sorry this summary was overdue.

Tim

-- 
Tim Mooney                              mooney_at_toons.cc.ndsu.NoDak.edu
Information Technology Services         (701) 231-1076 (Voice)
Room 242-J6, IACC Bldg.                 (701) 231-8541 (FAX)
North Dakota State University, Fargo, ND 58105
---- original message follows ----
All-
I've run into a bit of a problem tonight while installing some patches for
Digital Unix 3.2c on my AlphaStation 200 (it's the guinea pig before we put
any of these patches on a general purpose system).  I've installed a number
of patches (mostly security related) and one of them has caused the Common
Desktop Environment to quit working for normal users (but one can still log
in as root).
Here's the relevant information:
OS: 3.2c
Patches installed tonight:
	OSF 350X-001
	OSF 350X-002
	OSF 350-015
	OSF 350-048
	OSF 350-064
	OSF 350-070
	OSF 350-061
	OSF 350-018
	OSF 350X-004
	OSF 350X-005
	OSF 350-004
Error message:  When a normal (non-root) user tries to log in via dtlogin
to the Common Desktop Environment, he/she receives the error message:
	"Sorry. Maximum number of users already logged in."
That message is listed at the bottom of the man page for `dtlogin', but no
reason is given for what causes that message (I'm assuming it could only
be licensing issues or pseudo-terminal problems that would cause it).
Normal users can log in just fine via telnet/rlogin or if xdm is used instead
of the CDE.  Also, root can log in via dtlogin, so it would seem to be
a permissions problem.
I have read the summary regarding `telnet-all ports full' sent to this list
by "William H. Magill" <magill_at_isc.upenn.edu>, but the problem isn't that
I'm out of pseudo-terminals (though I believe the problem *is* that I'm
having problems accessing them via dtlogin) because I can access the machine
via all means except the CDE.  Also, using 'sysconfig -q pts' I see ntpys=255, 
which has always been plenty for my personal workstation before.
Of the patches listed above, only
	OSF 350X-001
	OSF 350X-002
	OSF 350-015
	OSF 350-048
	OSF 350-064
	OSF 350-061
are security related, and I'm fairly sure that the problem was created by
patch OSF350-048, which corrects a security problem in the /dev directory
(I'm not certain how much I should divulge, though the patch information
can be had by anyone, as far as I know) by remaking some device files.
I have tried remaking the old (insecure) device files, but the problem
persists.  /var/dt/Xerrors and $HOME/.dt/errorlog contain nothing enlightening,
and running /usr/dt/bin/dtlogin with `-debug' enabled only indicates that I
am in the `ERROR' state after leaving the `CHALLENGE' (passwd prompt) state.
I have stopped and restarted the Common Desktop Environment, and I've rebooted
several times.
I will be in touch with Digital's CSC in the morning, but I thought I would
tap the collective experience of the list to see if anyone else has run into
(and hopefully over) the problem I'm experiencing.  As I said, I have tried
remaking the device files using the pre-patch version of `MAKEDEV', but that
does not correct the problem, so it appears that it may not be possible to
`back out' the patch.
I will summarize to the list.
Thanks,
Tim
-- 
Tim Mooney                              mooney_at_toons.cc.ndsu.NoDak.edu
Information Technology Services         (701) 231-1076 (Voice)
Room 242-J6, IACC Bldg.                 (701) 231-8541 (FAX)
North Dakota State University, Fargo, ND 58105

Received on Tue Jan 30 1996 - 23:45:20 NZDT