[comp.sys.apollo] security hole

dawson@apollo.HP.COM (Keith Dawson) (08/09/89)

In a recent posting Peter Lipp [plipp@tugiig.uucp] details a

 >  Possible Security Problem with DOMAIN-OS and the Display-Manager

This security problem has been fixed in SR10.2. Patches have been 
generated for earlier releases:

  SR9.7 -- patch # 184 on the June 1989 patch tape

  SR10.1 -- patch # m0048 on the August 1989 tape

These patches can be applied to /lib/streams on any version of SR9.7
and SR10.1, including the versions that support Domain/X11. Please
see your local support office to obtain copies of these patch tapes.

We regret the broad dissemination of detailed instructions for exploiting
a security hole.
____________________________________________________________  
Keith Dawson  Section Manager, Window Systems Group
              Hewlett Packard Co.         508-256-0176 x5739  
              Graphics Technology Division / East  
              300 Apollo Dr.  (CHR.03.DE)
              Chelmsford, MA  01824                      USA  
              Internet: dawson@apollo.hp.com                         
              UUCP: {mit-eddie,yale,uw-beaver}!apollo!dawson  

jim@eda.com (Jim Budler) (08/09/89)

dawson@apollo.HP.COM (Keith Dawson) writes:

>This security problem has been fixed in SR10.2. Patches have been 
>generated for earlier releases:

>  SR9.7 -- patch # 184 on the June 1989 patch tape

>  SR10.1 -- patch # m0048 on the August 1989 tape

>We regret the broad dissemination of detailed instructions for exploiting
>a security hole.
>____________________________________________________________  
>Keith Dawson  Section Manager, Window Systems Group
>              Hewlett Packard Co.         508-256-0176 x5739  
>              Graphics Technology Division / East  


We regret that it took such a blunt step to cause Apollo to bother to
tell us that the hole exists, and has fixes.

From Sun Microsystems?

	Every quarter, a thick book arrives in the mail, detailing
bug reports.

	Every month, a book arrives in the mail, containing
interesting Technical details.

	I forget which usually contains notices of patches, perhaps
both.

From Apollo?

	Every month, an invoice arrives. (Well, not any more 8^)

================

I fully agree with the posting of the bug. Look, INSTANT action.
Explicit mention of compatibility of /lib/streams. High awareness
in community of seriousness of bug.

Report it to software support. You get told about the patch.

Don't notice the problem. Remain ignorant.

Post a note to Usenet less explicitely saying there is a bug. Response
from Apollo may list patches, but people may not realize how serious
it is, no matter how serious the poster says it is. Result: Hacker
decides to explore area in question, finds hole, most people do not
install patches, perhaps because of the confusion caused by the fact
that almost everything Apollo fixes involves /lib/streams.

The party line answer will be "what about all the people not on
software support or Usenet"

Apollo needs to solve that problem, not hide the facts from the rest
of us.-- 
Jim Budler   address = uucp: ...!{decwrl,uunet}!eda!jim
					 domain: jim@eda.com
			 voice	 = +1 408 986-9585
			 fax	 = +1 408 748-1032

collins@nvpna1.prl.philips.nl (Donal O Coileain) (08/10/89)

In article <511@eda.com> jim@eda.com (Jim Budler) writes:
>From Apollo?
>
>	Every month, an invoice arrives. (Well, not any more 8^)

Apollo produces a patch tape every month. In the 9.7 patch tape for JUNE 89
months before this discussion was started I read :

 "Patch 184 APR DCB34 : A security hole existed in the pad_$dm_cmd
	   ..........................
  Now if the two user_ids are not equal, the command is disallowed and
  the following error status is returned:

	'operation is illegal when no display is attached'"

You cannot blame Apollo because you don't read the release notes or 
understand the bugs/fixes.

>I fully agree with the posting of the bug. Look, INSTANT action.
>Explicit mention of compatibility of /lib/streams. High awareness
>in community of seriousness of bug.

I see no problem in posting the problem, however I feel that it is not
necessary to post the source code as well.


Donal O Coileain.   collins@apolloway.prl.philips.nl   or
                    collins%nvpna1.prl.philips.nl@uunet.uu.nl
-- And out of the gloom a voice said, 'Smile and be happy for things could
   be a lot worse'. So I smiled and was happy and behold, things got worse --

kts@quintro.uucp (Kenneth T. Smelcer) (08/11/89)

In article <511@eda.com> jim@eda.com (Jim Budler) writes:
>
>We regret that it took such a blunt step to cause Apollo to bother to
>tell us that the hole exists, and has fixes.
>
>From Sun Microsystems?
>
>	Every quarter, a thick book arrives in the mail, detailing
>bug reports.
>
>	Every month, a book arrives in the mail, containing
>interesting Technical details.
>
Mentor Graphics does the same thing for its Apollo customers.  Every month
we get a book with technical tips for Sys Admins, followed by a list of
current bugs, work-arounds, and bug fixes for their entire product line.

It would be great if Apollo would do something like this.  It would
save them money in fewer calls to the Hotline, and it would enhance
their image (I tend to get real ticked off when I spend 2-3 days trying
to fix something that Apollo already knows doesn't work.)

Who knows, maybe with HP taking control, Apollo will show more interest
in taking care of their current customer base, as well as trying to open
new markets.

-- 
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Ken Smelcer        Quintron Corp.           quintro!kts@lll-winken 
                   Quincy,  IL              tiamat!quintro!kts@uunet

jwright@atanasoff.cs.iastate.edu (Jim Wright) (08/11/89)

In article <641@prles2.UUCP> collins@nvpna1.UUCP (Donal O Coileain) writes:
| 
| Apollo produces a patch tape every month.
| 
| You cannot blame Apollo because you don't read the release notes or 
| understand the bugs/fixes.

I don't happen to have the clout of Phillips Research Labs.  I don't get
a patch tape every month, or even an announcement.  In fact, this is the
first I've heard of it.  But enough whining.  I'd like to offer some
constructive help.

The last time the patch issue was raised, there were some noises about
the University of Iowa setting up an anonymous ftp site which archived
all the patches.  Apparently it has died; I haven't heard a peep out of
them since.

As part of the anti-viral archives, in conjunction with comp.virus/VIRUS-L,
we are setting up a site for Unix systems.  To start this, we already
have bug fixes for Sun, Pyramid and DEC.  We would be pleased to make the
Apollo patches available too.  If someone from Apollo could respond here,
contact me, or give me a lead as to who to contact at Apollo, I would
appreciate it.

Also, opinions from the net as to whether this is good or bad would be
helpful.  Perhaps if enough interest is shown, it will happen.  I know
that anonymous ftp is not available to everyone.  But as it stands now,
you apparently have to be a mega-corporation with an inside track to
Apollo to get access to bug fixes.  (My perception; am I wrong?)

-- 
Jim Wright
jwright@atanasoff.cs.iastate.edu

dente@els.uucp (Colin Dente) (08/11/89)

In article <44e9d7d4.c4b0@apollo.HP.COM> dawson@apollo.HP.COM (Keith Dawson) writes:
>In a recent posting Peter Lipp [plipp@tugiig.uucp] details a
> >  Possible Security Problem with DOMAIN-OS and the Display-Manager

>[Gives patch numbers for 9.7 and 10.1]

>These patches can be applied to /lib/streams on any version of SR9.7
>and SR10.1, including the versions that support Domain/X11. Please
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

Well, I 'phoned my local support office, and they told me that the
release notes for the 9.7 patch (they haven't got the 10.1 patch yet)
state that it should *not* be applied if you are running Domain/X11.
Naturally, I was a bit miffed at this - so I 'phoned Keith to see what
he had to say.  He seemed to think that the patch was, infact, more
than one patch in one bundle, and if I applied just the /lib/streams
part, I'd be okay.  Anyway, I'm waiting for a floppy to land on my
doormat with the patch on it, and when it does, I'll try it and let
you know.  Just thought I'd try to clear the confusion in the
meantime.

>We regret the broad dissemination of detailed instructions for exploiting
>a security hole.
So do I...... |-(

Colin

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
| Colin Dente                      | JANET: dente@uk.ac.man.ee.els            |
| Dept. of Electrical Engineering  | ARPA:  dente@els.ee.man.ac.uk            |
| University of Manchester         | UUCP:  ...!mcvax!ukc!man.ee.els!dente    |
| England                          | These might work now, but then again...  |
|-----------------------------------------------------------------------------|
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=

jim@eda.com (Jim Budler) (08/15/89)

collins@nvpna1.prl.philips.nl (Donal O Coileain) writes:

>In article <511@eda.com> jim@eda.com (Jim Budler) writes:
>>From Apollo?
>>
>>	Every month, an invoice arrives. (Well, not any more 8^)

>Apollo produces a patch tape every month. In the 9.7 patch tape for JUNE 89
>months before this discussion was started I read :

> "Patch 184 APR DCB34 : A security hole existed in the pad_$dm_cmd
>	   ..........................
>  Now if the two user_ids are not equal, the command is disallowed and
>  the following error status is returned:

>	'operation is illegal when no display is attached'"

>You cannot blame Apollo because you don't read the release notes or 
>understand the bugs/fixes.

Wanta bet? In two years of paying Apollo for support I did not
receive ONE patch tape that I didn't have to ask about myself.
I can't read the release notes if I don't get them. I call in a bug
and get told "fixed in the XXXX patch tape", and that is frequently
months old, as you mention.

It doesn't do me any good to have Apollo produce a monthly patch tape
if they never inform me of it's existance.

You didn't read my posting very well or you would have realized
that. You even quoted my statement that the only monthly mailings I got
from Apollo were bills.

>Donal O Coileain.   collins@apolloway.prl.philips.nl   or
>                    collins%nvpna1.prl.philips.nl@uunet.uu.nl
>-- And out of the gloom a voice said, 'Smile and be happy for things could
>   be a lot worse'. So I smiled and was happy and behold, things got worse --

jim
-- 
Jim Budler   address = uucp: ...!{decwrl,uunet}!eda!jim
					 domain: jim@eda.com
			 voice	 = +1 408 986-9585
			 fax	 = +1 408 748-1032