Undeliverable mail: SMTP delivery failure

Fri Feb 15 07:06:02 AEST 1991

Return-path: <postmaster at VAX1.Mankato.MSUS.EDU>
Date: Thu, 14 Feb 1991 13:46 CST
From: PMDF Mail Server <Postmaster at VAX1.Mankato.MSUS.EDU>
Subject: Undeliverable mail: SMTP delivery failure
To: "MSUS1::IN%\"UNIX-WIZARDS at BRL.MIL\""@VAX1.Mankato.MSUS.EDU
Message-id: <8AD0265CE0204963 at VAX1.Mankato.MSUS.EDU>

The message could not be delivered to:

Addressee: BD6 at MKATT1.MANKATO.MSUS.EDU
Reason: Illegal host/domain name found.

----------------------------------------

Received: from DECNET-MAIL by VAX1.Mankato.MSUS.EDU with PMDF#10000; Thu, 14
 Feb 1991 13:44 CST
Date: Thu, 14 Feb 1991 13:44 CST
From: "MSUS1::IN%\"UNIX-WIZARDS at BRL.MIL\""@VAX1.Mankato.MSUS.EDU
Subject: UNIX-WIZARDS Digest  V12#018
To: BD6 at MKATT1.MANKATO.MSUS.EDU
Message-id: <8A91244C0020523B at VAX1.Mankato.MSUS.EDU>
X-VMS-To: "Brian D. Goecke" <bd6%MKVAX1.DECNET at MSUS1.BITNET>

Date: Thu, 14 Feb 91 05:45:01 EST
From: Mike Muuss The Moderator <Unix-Wizards-Request at BRL.MIL>
Subject: UNIX-WIZARDS Digest  V12#018
Sender: Unix-Wizards Mailing List <UNIX-WIZ at NDSUVM1.BITNET>
To: "Brian D. Goecke" <bd6%MKVAX1.DECNET at MSUS1.BITNET>
Reply-to: UNIX-WIZARDS at BRL.MIL
Message-id: <4659BFF0A0001A42 at MSUS1.MSUS.EDU>

UNIX-WIZARDS Digest          Thu, 14 Feb 1991              V12#018

Today's Topics:
                        Re: What is sigcleanup?
                   Re: SIGCONT occurs after a SIGTERM
             Re: dynamic linking C code with ld link editor
             Re: Help!  There's a slash '/' in my filename.
                             cfreelist info
                     gettytab, entries f0, f1 & f2
              Loading and Executing Object Code at Runtime
                                Re: dup2
 Re: Stuck-up Wizards (Re: Summary and Fix for "slashes in filenames")
                       Re: Slashes in file names

-----------------------------------------------------------------

From: Root Boy Jim <rbj at uunet.uu.net>
Subject: Re: What is sigcleanup?
Date: 25 Jan 91 00:38:21 GMT
To:       unix-wizards at sem.brl.mil

In article <1991Jan22.153802.9869 at daimi.aau.dk> kaja at daimi.aau.dk (Kaja P.
 Christiansen) writes:
>What is "sigcleanup" doing?

I dunno for sure, but I'll hazard a guess. It could be a mutation of
4.3BSD's sigreturn call with new bells and whistles. RTFM sez
"sigreturn allows users to atomicly unmask, switch stacks, and
return from a signal context". Since signals are more complex
than they used to be, s system call is required to exit from
them as well as to enter them. When a signal handler returns,
it executes what is known as the signal trampoline code, which
actually calls sigreturn. Setjmp/longjmp have been extended to
save and restore the signal mask as well. However, _setjmp/_longjmp
exist for speedier execution when signals are involved.

Now sing with me to the tune of a Beatles song:

	We all live in a signal trampoline,
	A signal trampoline, a signal trampoline!

>Kaja P. Christiansen                     Computer Science Department
>(kpchristiansen at daimi.aau.dk)            Aarhus University
>Phone: +45 86 12 71 88                   Ny Munkegade 116,
>Fax:   +45 86 13 57 25                   DK-8000 Aarhus C., Denmark

--

	Root Boy Jim Cottrell <rbj at uunet.uu.net>
	Close the gap of the dark year in between

-----------------------------

From: Edward Conger <conger at hpcupt1.cup.hp.com>
Subject: Re: SIGCONT occurs after a SIGTERM
Date: 12 Feb 91 22:23:14 GMT
To:       unix-wizards at sem.brl.mil

/ hpcupt1:comp.unix.wizards / coleman at cam.nist.gov (Sean Sheridan Coleman X5672)
 /  9:06 am  Feb 11, 1991 /
>Please explain to me why a SIGCONT is sent to a process after
>SIGTERM is sent to my process. It doesn't compute because TERM
>means to terminate the the process.

The distinction is that sending a signal to a process (usually|often)
is implemented by setting a bit in a flag word associated with the
"victim process".  The action of *send*ing the signal doesn't
terminate the process, rather, it says, "when next you run in the
kernel (either via a system call or a timeslice (usually ~ 1/100 sec)),
you should go handle this signal."  In the case of SIGTERM, the default
behaviour is to TERMinate.

Now suppose the victim process is stopped (either by job control, SIGSTOP,
or via debugging), it will NOT see the bit set in the flag word until it
runs again.  The SIGCONT gets it unstopped and it runs long enough to
terminate.

Your mileage (and implementation) may vary, but this is the general gist of
the problem.

>Thanks

>Sean Coleman
>coleman at bldrdoc.gov
>NIST
>Boulder, CO
>----------

Hope this helps,

-Ed.

===========================================================================
The above is an official statement of MeMyself & I Inc.  It should not
be interpreted to be an official statement of any other likely targets,
including, but not limited to, Hewlett-Packard Co., ACME Rockets, ACME Rubber
Bands, ACME Consolidated Mining Engineering, or the Home for Damaged Coyotes.

-----------------------------

From: Steve Alter <alter at ttidca.tti.com>
Subject: Re: dynamic linking C code with ld link editor
Date: 13 Feb 91 03:09:27 GMT
To:       unix-wizards at sem.brl.mil

In article <23713 at netcom.COM> aed at netcom.COM (Andrew Davidson) writes:
} I am trying to fiqure out how to use the ld link editor to dynamicly link
} some C code.  Are there any refference? All I have are the man pages
} which are less than usefull for the beginer
}
} I am working on a sun workstation, but must also get this to run on a
} sys V unix(SCO ODT UNIX sysV 3.2.0).

This info is for SunOS (4.1, but I believe also applies to 4.0.)

Load the "shlib.etc" package from your distrubution media (under 4.1
the command to do this is "/usr/etc/install/add_services") and then
look in directory /usr/lib/shlib.etc at the README and Makefile files.
That stuff only describes how to build a new version of the shared libc
library, but it shows some of the basics that might be useful/needed in
building your own library.

Highlights:

-- You have to build all of your object modules (for this library) with
   the "-pic" option to generate position-independent (a.k.a.
   relocatable) code.

-- The Makefile uses the "-assert pure-text" option on the "ld" command
   to ensure (I believe) that everything in the library is read-only,
   i.e. no writable global data areas because read-only is a must for
   anything that is shared between anonymous users.  If you want a
   global writable buffer then you've got to create a shared-memory
   segment.

-- The Makefile also plays some games to ensure that the object modules
   go into the library in the correct order.  The concepts of ranlib and
   the __.SYMDEF module don't apply to shared libraries because they're
   mapped directly into the address space of the process, whereas a
   normal library is just a collection of .o files with a symbol table
   up front and .o files still have to be linked with ld.

Read the "ld" manual-page very carefully as it explains a lot of how
the ld and ld.so programs, the LD_LIBRARY_PATH, and the version numbers
on the /usr/lib/lib*.s[ao].* files work together.

My ultimate suggestion: buy a set of the manuals; there's a whole
chapter on shared libraries in there.
--
Steve Alter        <alter at ttidca.tti.com>
{philabs,psivax,pyramid,quad1,rdlvax,retix,rutgers}!ttidca!alter
Transaction Technology Inc. a subsidiary of Citicorp  (213) 450-9111 x2541

-----------------------------

From: Greg Kemnitz <kemnitz at gaia.berkeley.edu>
Subject: Re: dynamic linking C code with ld link editor
Date: 13 Feb 91 20:04:48 GMT
Sender: USENET Administrator <usenet at agate.berkeley.edu>
To:       unix-wizards at sem.brl.mil

I had this problem for several months until (after much hacking) I got my
dynamic loaders to work with Postgres.  It is a rather different idea than
shared libraries - the idea is to load and execute a function (whose name is
unknown beforehand) in an object file given by the user.  It is quite tricky,
since you have to know about the a.out format, and getting it to work requires
using lots of undocumented or almost undocumented features and options
(especially on DECstations).  Also, Postgres had to be able to unload and
reload a user function (since the user may have changed it) which is easy but
not entirely straightforward.

I have two versions - one for Sun 3's, Sparcs, and Sequents, and another version
for DECstations running Ultrix >= 4.0.  They all use "ld -A", but there's lots
of stuff you need to know that you only discover after much inspection.  Nobody
has a man page that is at all descriptive on "ld -A"'s use.  If there's
interest, I'll post them to the net.

 -----------------------------------------------------------------------
Greg Kemnitz                  |      "I ran out of the room - I
Postgres Chief Programmer     |      didn't want to be killed by a pile
278 Cory Hall, UCB            |      of VMS manuals" :-)
(415) 642-7520                |
kemnitz at postgres.berkeley.edu |      --A friend at DEC Palo Alto in the Quake

-----------------------------

From: s082 at brems.ii.uib.no
Subject: Re: Help!  There's a slash '/' in my filename.
Date: 13 Feb 91 12:02:01 GMT
Sender: Usenet posting account <usenet at eik.ii.uib.no>
To:       unix-wizards at sem.brl.mil

There seems to be a rather lengthy discussion going on about slashes in
filenames,
how to get them there, and whether to ask questions about them...
Just one thing: Why would anyone want a slash (or any such character) a
filename in the first place?

-km

-----------------------------

From: John F Haugh II <jfh at rpp386.cactus.org>
Subject: cfreelist info
Date: 13 Feb 91 16:45:41 GMT
X-Clever-Slogan: Recycle or Die.
To:       unix-wizards at sem.brl.mil

in various system v's the cblock freelist is declared
something like

struct chead cfreelist;

and (struct chead) is

struct chead {
	struct	cblock	*c_next;
	int	c_size;
	int	c_flag;
}

what the hell is cfreelist.c_size =supposed= to contain?  it
sure doesn't contain the total number of characters available
on the cblock freelist right now.  my x.29 tty driver thanks
you ...

ps - no fair peeking at the source code.
--
John F. Haugh II                             UUCP: ...!cs.utexas.edu!rpp386!jfh
Ma Bell: (512) 832-8832                           Domain: jfh at rpp386.cactus.org
"I've never written a device driver, but I have written a device driver manual"
                -- Robert Hartman, IDE Corp.

-----------------------------

From: Thomas Omerzu <omerzu at quando.quantum.de>
Subject: gettytab, entries f0, f1 & f2
Date: 13 Feb 91 20:13:17 GMT
To:       unix-wizards at sem.brl.mil

A little but nasty problem:

All bsd-based Unix-Implementations of getty (at least as far as known to me)
support gettytab attributes f0, f1 and f2.

According to TFM(gettytab) they take numeric values, where f0, f1 & f2
represent the tty mode flags "to write messages", "to read login name",
and "to leave terminal in".

"... Terminal modes to be used for the output of the message,
for input of the login name, and to leave ther terminal set as upon
completion, are derived from the Boolean flags specified. If the derivation
should prove inadequate, any (or all) of these three may be overridden
with one of the f0, f1, or f2 numeric specifications, which can be
used to specify (usually in octal, with a leading 0) the exact values
of the flags. Local (new tty) flags are set in the top 16 bits of this
(32-bit) value. ..."  [ From Dec Ultrix V4.0 Ref. Man. Vol 6 ]

Now my question: how do the terminal mode flags (as know from termio or stty)
correspond to the bits in f0 (or f1, f2)?

Obviously it cannot be a 1:1 corresponency, since the struct termio
contains 4 short values (c_iflag, c_oflag, c_cflag, c_lflag),
or - in other words - 64 bits.

Any ideas/experiences anyone?

I was trying to use this on a DECstation 5000/200 running Ultrix 4.0,
although other vendors seem not to supply much more information.

Thanks in advance for your help,
	Thomas

--
*--------------------------------------------------------------------------*
   Thomas Omerzu.       Internet: omerzu at quantum.de
    Quantum GmbH        Bitnet:   omerzu%quando at UNIDO.bitnet
  Dortmund,Germany      UUCP:     ...!unido!quando!omerzu

-----------------------------

From: Ian Flanigan <flan at cics.berkeley.edu>
Subject: Loading and Executing Object Code at Runtime
Keywords: help, load, run, .o, LISP
Date: 13 Feb 91 21:27:04 GMT
Sender: USENET News System <news at cec1.wustl.edu>
To:       unix-wizards at sem.brl.mil

Hi,

I tried posting this to comp.unix.questions first, but as I got no answer
I figured this must be a Wizard level question:

I'm wondering how to load an object file (my_functions.o) at execution
time and execute a function contained therein.  I know this is possible
since many flavors of LISP allow you to compile your functions and then
load the compiled versions later.

If it's in TFM, could someone point me in the right direction, or, if it's
trivially simple, could someone please explain how to go about it?

Thanks much.

--
Ian Flanigan

flan at cics.wustl.edu              "You can never have too many napkins."
wucs1.wustl.edu!cics!flan at uucp

-----------------------------

From: Doug Gwyn <gwyn at smoke.brl.mil>
Subject: Re: dup2
Date: 13 Feb 91 22:41:02 GMT
To:       unix-wizards at sem.brl.mil

In article <122373 at uunet.UU.NET> rbj at uunet.UU.NET (Root Boy Jim) writes:
-In article <15136 at smoke.brl.mil> gwyn at smoke.brl.mil (Doug Gwyn) writes:
->/*
->	dup2 -- 7th Edition UNIX system call emulation for UNIX System V
->*/
-Do you mean that V7 had dup2 but System V doesn't?
-And what about systems that don't have fcntl?

Geez, what do you think the comment means, anyway?

-----------------------------

From: David Elliott <dce at smsc.sony.com>
Subject: Re: Stuck-up Wizards (Re: Summary and Fix for "slashes in filenames")
Date: 14 Feb 91 01:35:27 GMT
To:       unix-wizards at sem.brl.mil

In article <=H_&Z9#@rpi.edu> rodney at sun.ipl.rpi.edu (Rodney Peck II) writes:
>and finally, just because a question comes back again and again, doesn't
>mean that all the wrong answers have to be dragged out again.  that's what
>the faq is for -- if someone wants to say "hey I read these answers in
>the faq, and I didn't see this idea .... why won't this work?" -- that,
>in my opinon, would be fine.

I certainly agree with you, but let me make a suggestion.

Instead of people complaining about people sending out wrong answers,
or even answers at all, let's all try doing this:

	If someone posts a frequently-asked question, mail them a copy
	of the FAQ posting, and point out that the answer is there.

	If someone posts an answer, be it correct or incorrect, mail
	them a copy of the FAQ posting to remind them that the answer
	is already available to everyone in a consistent format.

I did this myself for a while, made some friends in the first case, and
got some pretty interesting excuses in the second.

-----------------------------

From: Steve Summit <scs at adam.mit.edu>
Subject: Re: Slashes in file names
Date: 14 Feb 91 07:05:12 GMT
Sender: News system <news at athena.mit.edu>
To:       unix-wizards at sem.brl.mil

I begin to understand why people have such passionately awful
things to say about NFS.

As far as I'm concerned, if the NFS server code can create an
illegal file (where "illegal" is defined by the host on which the
server is running), then that's a bug in the NFS server, period.
If the protocol doesn't include an error return of "I can't
create a file of that name" (not surprising; <errno.h> doesn't
define that specific error either, although it probably should),
then it should just say something halfway like "no such file or
directory."

If the client can help out by trying not to ask for illegal files
in the first place, that's fine; but in the final analysis the
server has got to protect itself.  Having the client not ask for
illegal files is never going to work perfectly, anyway: even if a
negotiation is defined by which the client can learn which
characters are illegal, some operating system somewhere is going
to have rules that can't be encoded that simply.

Has anyone ever considered modifying namei so that it essentially
checks to see if the current path character matches a target
filename character *before* checking to see if the current path
character is a '/'?  That would make the filesystem robust
against inadvertent /'s in filenames, because unlink("a/b/c")
would in fact unlink a file called "b/c" in directory "a", if one
somehow existed.

(To be sure, there would be idiosyncrasies associated with such
an algorithm.  In the above example, if directory "a" also
contained a subdirectory "b" as well as a file "b/c" the behavior
would depend on which came first, or something.  Still, it might
be a nice escape hatch.  Note that by only attempting to match
existing files in this way, it would still never create a
filename with a slash.)

                                            Steve Summit
                                            scs at adam.mit.edu

-----------------------------

End of UNIX-WIZARDS Digest
**************************