Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.

Dismiss

DECUServe Journal January

9 views

Skip to first unread message

Sharon Frey

unread,

Jan 3, 1994, 1:09:37 PM1/3/94

The DECUServe Journal
=====================
January 1994

Hello all! I hope that everyone enjoyed the holidays and had a very
nice new year. This issue marks the first anniversary for the electronic
version of the DECUServe Journal. We're doing well, and I must tell you, the
reception by the Internet audience over the year has been pretty gratifying.
As usual, this issue is chock full of a wide range of interesting
topics. Please feel free to browse.

Sharon Frey
DECUServe Journal Editor
fr...@eisner.decus.org

Table of Contents
-----------------

Editor's Greeting ................................. 1
Technical Information ............................. 34
Articles
Word Processing Packages for Alpha-VMS? ........ 2
Batch ad-hoc report-writer for Oracle .......... 8
DECnmail (Nmail) R 9.3-4 ....................... 12
PostScript printing on HP Laserjet III ......... 15
High Speed modems for disaster recovery? ....... 22
crontab file format ............................ 24
Bootstrap problem of sorts ..................... 29

January, 1994 The DECUServe Journal Page: 1

Word Processing Packages for Alpha-VMS?
---------------------------------------

The following article is an extract of the DECUServe
3rd_Party_VMS_Software conference topic 349. The discussion
occurred between October 27, 1993 and December 23, 1993.

By Bob Comarow, Bill Mayhew, Ray Whitmer, Chuck McMichael, Patrick Stair,
Shawn Allin, Michael Connelly, Duncan Brown, Linwood Ferguson, Dale Coy,

(10/27/93 Comarow)
------------------
Any word processing packages currently available - or at least taking
orders for Alpha Open VMS?

(10/27/93 Mayhew)
-----------------
Yes... contact me via Email please. We have a stable, mid-range word
processing product that runs on VAX and are ready to move it to Alpha, but
release has been deferred pending customer interest (read: assurance of $$).

(10/29/93 Whitmer: WordPerfect 6.0 is also in the works: ask marketing.)
--------------------------------------------------------------------------
WordPerfect 6.0 is also in the process of being ported to VAX and AXP
simultaneously. I am not yet allowed to quote a time frame. But I understand
our marketing department does talk about it.
Disclaimer: I work for WordPerfect Corporation, specifically developing
WordPerfect 6.0 for VMS. I am strongly biased in its favor.

(11/01/93 McMichael: A rough estimate)
---------------------------------------
I was at an office automation show a couple of weeks ago, and the
marketing-type from WordPerfect was saying "2nd quarter of '94."

(11/01/93 Whitmer: Good. I wasn't allowed to say.)
----------------------------------------------------

(12/07/93 Stair: WP 6.0 on both VAX & Alpha? Good news!)
----------------------------------------------------------
Ray,
When you say "simultaneously", do you *really* mean **simultaneously**?
I hope so. We are dying to here some good news about 6.0 on the Alpha,
especially since we are about to buy our second and third Alpha boxes.
(By the way, do you have any news on when/if we can hope for WPOffice 4.x,
on either VAX or Alpha?)

January, 1994 The DECUServe Journal Page: 2

(12/07/93 Allin: Would love to see WPO on VMS too.)
----------------------------------------------------
Having heard from a WordPerfect person that WordPerfect Office 4.0 isn't
planned for VMS, Ray, is there anything that people like Patrick and I (who
was VERY disappointed to hear the above news) can do to persuade WP Corp. to
change their mind?

(12/08/93 Connelly: Is there life after ALL-IN-1?)
---------------------------------------------------
We, also, would love to see WPCORP commit to WordPerfect Office 4.0 (and
beyond) for VMS & Alpha. Here's our scenario...
We've been an ALL-IN-1 IOS shop for many years and switched from WPSplus to
WordPerfect several years back. While we have many 386/486 PCs, everyone uses
ALL-IN-1 and WordPerfect 5.1 on the VAX through dialup or lat connections.
All email (internal & Internet) and much word processing is performed in this
manner. Of course, many also use WordPerfect 5.2 for DOS and Windows and
greatly appreciate the file compatibility that exists between the VAX and PC
platforms.
Now we're entering into the realm of client/server applications. At
present, this is focused on our database environment and we've commited to
Uniface on PC/Windows and VAX and Rdb under VMS. Of course, users are rarely
statisfied with slow progress. They also want a completely integrated
client/server email and word processing environment that will work cleanly on
their PCs and/or via dialup to the VAX. WPCORP, do you see where I'm
heading...?
Right now we're on the sidelines. We're sticking to ALL-IN-1 IOS office
automation for at least another year but our long-term strategy is to identify
an appropriate technology that will augment, transition, and/or replace
ALL-IN-1. DEC may eventually get there, but it certainly seems that
WordPerfect Office would be a serious contender. Of course, a strong
committment to VMS/Alpha is an absolute requirement.
I would be amazed if other ALL-IN-1 shops are not also struggling with
similar issues.

(12/08/93 Brown: We were also disappointed to hear that)
---------------------------------------------------------
I'll toss in my vote for Office 4.0 on VMS! We're current Office on VMS
users who are jealous of the more capable PC version. The same situation
exists with WP 6.0, but at least we know there's an end in sight, and we too
can have all those neat features if we wait a few months.

(12/17/93 Allin: Statement from WP person re: Office on VMS)
-------------------------------------------------------------
I received a mailer from WPCorp last week, including a glossy on WP Office
4.0 saying it runs on everything but VMS. The sales guy included a business
card that had his e-mail address on it. So, I wrote, asking about the future
of Office and VMS. Here's his reply, including a uuencoded statement of
Office futures:

January, 1994 The DECUServe Journal Page: 3

Date: Thu, 16 Dec 1993 12:42:55 -0700
From: KEN...@WordPerfect.com
Subject: WordPerfect Office request -Reply
To: ACC...@KRDC.INT.ALCAN.CA
Message-id: <sd1058...@WordPerfect.com>
X-Mailer: WordPerfect Office 4.0

** Reply Requested by 2/6/2106 (Saturday) **
Hello Shawn,

It's nice to meet you--electronically! I hope we will have the opportunity
in person sometime. You are not alone in your desire to see Office 4.0 show
up under VMS. Unfortunately, there are no current plans for that to happen.
That said, however, I should point out that WordPerfect is committed to support
VMS. We have developed a VMS Mail Gateway and a VMS Message Router for Office
4.0 (the Gateway will access VMS Mail, and the Router accesses All-in-one).
Office 3.1 for VMS already does include some of the more advanced features of
Office 4.0 such as automatic forward/reply, WordPerfect 5.1 printing, public
mailboxes, and user-definable sort order (in the In/Out Box).
I appreciate your desire to see continued development for VMS, so I am
giving you a lobbying avenue. The Director of Office Marketing for WordPerfect
Corp is Eldon Greenwood. (He doesn' know I'm doing this, but...) If you would
like to place an official request with him, please feel free to send him a
message on Internet at el...@wordperfect.com. The more requests he receives,
the more likely he is to seek for increased development, so you may wish to
have the "other companies" you referred to mail him as well.
Attached is a copy of WordPerfect's statement of direction for Office for
VMS. It is somewhat dated, but it will give you the basic information on where
we still currently stand on VMS. Best of luck in you lobbying efforts. Please
feel free to contact me for any needs you may have.

Kent Wimmer

[Uuencoded file deleted, text version included below. - ED.]

(12/17/93 Allin: Any interest in pursuing this?)
-------------------------------------------------
...so, if anyone is interested in following this up, should we just all
send independent messages to the person mentioned, or should we try to come up
with a single note that we all send together?

(12/17/93 Ferguson: I'd say separate -- am writing now)
--------------------------------------------------------
If I know marketing people, the best result is a bunch of seemingly
independent contacts from different customers.

January, 1994 The DECUServe Journal Page: 4

(12/18/93 Stair: Text version of WPCorp policy?)
-------------------------------------------------
Shawn,

For those of us such as myself that are unable to decode a uuencoded file,
would it be possible to upload a text version of Kent Wimmer's statement that
he sent to you? (Or, could somebody enlighten me on the best way to decode
this? I tried, but got an invalid record length during the process.)
Before I compose my response to Eldon Greenwood, it might be a good idea
for me to read the official policy first.

(12/18/93 Coy: Text version)
-----------------------------
Of course, after uudecoding it, it's a WP file - so I had to convert to
text, etc. Sorry - it's wider than 80 columns. For those using OneKey, press
<Gold>Y to go to 132-column mode.

January 4, 1993

WordPerfect Office for VMS
Statement of Direction

Development of WordPerfect Office for VMS has been underway at WordPerfect
Corporation (WPCorp) since 1987. The first release of Office for VMS was
version 2.0, which was released in May 1988. Office 3.1 for VMS was released
in April 1992. Version 3.1 provides numerous enhancements for the VMS user of
WP Office. One of the most significant is the ability to exchange mail and
scheduler messages with users of other Office 3.x platforms-including DOS,
Windows, Macintosh, UNIX, and Data General. A bundled VMS Mail gateway and an
optional Message Router gateway extend the reach of WP Office 3.1 for VMS to
other mail systems available for VMS.

WPCorp has now announced the development of WP Office 4.0. Versions of Office
4.0 for DOS, Windows, and Macintosh are expected to be available by the middle
of 1993. Versions of Office 4.0 for OS/2 and UNIX systems are also under
development with releases planned for 1993.

Office 3.1 for VMS has several of the features being added to the 4.0 product.
Automatic forward/reply, WordPerfect 5.1 printing, public mailboxes, and
user-definable sort order (in the In/Out Box) are 4.0 features currently
implemented in Office 3.1 for VMS. In terms of features, Office 3.1 is very
competitive with other mail packages in the VMS environment. In addition to
the current ability to exchange mail and scheduler messages with other Office
3.x platforms, WP Office 3.1 for VMS will be able to exchange mail and
scheduler requests with systems running WP Office 4.x.

At this time, WPCorp is concentrating its new Office product development
efforts for the VMS environment on 4.0 level gateways rather than on a full
Office 4.0 product for VMS. Gateways to Message Router and VMS Mail should be
available at the release of WP Office 4.0 for DOS, Windows, and Macintosh.
These gateways will be directly usable by WP Office 4.0 to communicate with

January, 1994 The DECUServe Journal Page: 5

users of ALL-IN-1, VMS Mail, and other mail systems accessible through Message
Router.

The decision on when to develop Office 4.x for VMS will be based on the
following factors: 1) the market acceptance of Alpha and OpenVMS, 2) the
overall health of the OpenVMS software market, and 3) the development tools
and ease of development for this environment. When the next release of Office
for VMS is developed it will be able to support both the VAX and the new Alpha
architectures. Development of WordPerfect 6.0 for this new OpenVMS environment
is now underway. Because of the tools and development needs shared by the
WordPerfect and the Office development groups, we expect to benefit from this
experience.

(12/18/93 Ferguson: Reply to my note)
--------------------------------------
I wrote, and got a very quick reply.
Not a lot of meat, but at least he listened. If there are enough VMS users
still left maybe they will listen more.
His reply attached....

From: IN%"el...@WordPerfect.com" 17-DEC-1993 14:01:21.63
To: IN%"FERG...@Eisner.DECUS.Org"
CC:
Subj: WordPerfect for Office 4.0 -Reply

Linwood,

Thanks for taking the time to drop me a note on this. We are continually
monitoring market conditions and the cost of delivering products on specific
platforms. At this time, we have not made the commitment to develop WP Office
4.0 for VMS. This decision has been based on the lack of a strong demand for
the product, DEC's own direction with regard to OA software for VMS, and the
amount of resources that would be required to provide an acceptable product
offering. As a long-time VMS user myself, I will be the first to hope that
DEC's new Alpha chip helps them to more clearly define the role of the VMS
operating system in this decade. Many of our long-time VMS customers are now
transitioning to micro-computers, workstations, and LAN environments.

- Eldon Greenwood
Product Marketing Director, WP Office

(12/19/93 Stair: Thank you Dale and Linwood)
---------------------------------------------
Thanks for the text, Dale. Also, I should point out that this statement
from the note you received:

"...WP Office 3.1 for VMS will be able to exchange mail and
scheduler requests with systems running WP Office 4.x."

does not agree with the documentation I've read, nor with what WPCorp support
has told me, namely that VMS 3.x can NOT communicate directly with 4.0, but

January, 1994 The DECUServe Journal Page: 6

needs to go through a DOS 3.x gateway. (The DOS 3.x is able to communicate
directly with a 4.0 gateway.) If VMS 3.x could communicate directly with DOS
4.0 I wouldn't particularly care how long (or even IF) WPCorp upgraded VMS to
4.0, and I'd convert my DOS users to 4.0 in a flash. (Well, not _that_
quickly -- it isn't a simple conversion.)

Linwood,
Can I quote from Eldon Greenwood's reply to you when I write to him? In
particular, I intend to point out to him that, if it is indeed the case that

"Many of [WPCorp's] long-time VMS customers are now transitioning
to micro-computers, workstations, and LAN environments."

then it is even more important now that WPCorp help us make that transition.
At our agency, for instance, we are adding Pathworks users, but our VMS users
will probably stay in the majority for a long time. During this time, I need
to provide for continual integration of my VMS users with my non-VMS users,
for however long I have VMS users. I also can not drop support of my VMS users
just because WPCorp has decided that VMS is not worth it. So if WPCorp drops
VMS, I will be looking for someone besides WPCorp to supply me with
enterprise-wide office automation capabilities such as mail and scheduling.
On the other hand, if WPCorp continues to support VMS, the I will continue
to look to WPCorp to provide me with products for ALL our users on ALL our
platforms.

(12/19/93 Ferguson: Sounds wrong (right))
------------------------------------------
This seems to match what we have heard. I have not been directly involved,
but was given a proposal from our consultant which involved separate 3.1 and
4.0 server PC's required to make it all work with our VAX. I didn't know the
technical reasons, but presumed there was one.

>Can I quote from Eldon Greenwood's reply to you when I write to him?
>In particular, I intend to point out to him that, if it is indeed the

I do not object, but I offer the thought that thinking all of these
inquiries are independent might be more effective.
Then again I suspect we are all tilting at windmills anyway.

(12/23/93 Whitmer: What WordPerfect Corporation listens to... )
----------------------------------------------------------------
I am in a heavy coding cycle (yes, WP 6.0) , so I haven't been logging in
here as often as I should. I apologize for being bad at remembering names.
That is why I program instead of doing marketing.
At DECUS, in the WordPerfect Q & A, someone stood and said something that
caught everyone's attention. It ended with something like: "I would be very
interested in moving to Office if it were supported on VMS, and I would
certainly buy it for my 30,000 PCs as well."

January, 1994 The DECUServe Journal Page: 7

Lacking sales figures which show that a VMS version could pay for itself
(as good as it was, the VMS Office 3.1 did not sell well enough to justify
itself), this is the only type of argument which I can think of which the
decision makers will pay attention to.
"We like WordPerfect Office, and there would be no question on which
package to choose if there were better VMS support. PC Lans are great for
small business, but as a big company, we need centralized access on VMS. It
is essential to our use of WordPerfect Office and purchase of 10^n copies of
WordPerfect Office across our enterprise."
Instead, many customers would rather have development efforts going into
DOS, MacIntosh, Windows NT, and other products instead of VMS. This includes
even many DECUS-symposium-attending representatives of large sites. It becomes
increasingly difficult to even make a case for doing _any_ further VMS
development. Without large sales figures in this environment, and apparently
lacking DEC committment to market VMS as a future solution, life is rough for
those of us who currently develop WordPerfect products for VMS.
Both the symposium, and a recent tour of potential VMS product sites didn't
do much to improve the assessment. At this point, I can only think of two
things which might change that. Lots of real sales of the existing products
for VMS or lots of extremely credible committments to buy a new version (on
various platforms) when it is developed for VMS.
If you gave input at the DECUS symposium, it wouldn't hurt to send E-mail
to Eldon anyway. E-mail is better than spoken support, but will always be very
inferior to hard sales of the existing version of the product.
I will now digress into discussion which, while related, may belong
somewhere else. It is a different product on a different platform, but it is
quite similar to the situation of WordPerfect products on VMS, which is why I
give it here:
I like OS/2, not because it is perfect, but because it is much better than
DOS and/or windows, and it is affordable and it works. I like OS/2 native
products so I can use the features and performance of a real operating system.
WordPerfect was greatly lauded in its efforts to produce a native OS/2
version of WordPerfect 5.2. I have the product on my system, and I use it
heavily, and it works beautifully eliminating many of the quirks of DOS and
windows, and it integrates into a nice system object model not unlike the Mac,
but very unlike Windows. I still use the OS/2 version in preference to the
6.0 windows version. I have never experienced a bug or problem in it.
But when it was released, the editor of the prominent OS/2 magazine, while
lauding the effort, basically stated that anyone with half a brain would not
purchase the 5.2 version, but wait for WordPerfect to develop version 6.0 for
OS/2.
It really sends a clear message to the developer of the product when people
are vocal and cry for a product, but then few buy it, while the others only
begin to cry for the next version. If having a version 6.0 is more important
than having a native OS/2 version in hand, then so be it. But the software
producers are unlikely to spend much time developing the native version again.
Either we love the hardware/OS and will stay there forever, forcing software
producers to come to us if we represent a large enough market share, or taking
the left-overs of non-main-stream software if we do not, or we love the
software, and will follow it wherever it takes us. As much as I love VMS, I
believe I generally love software better than a particular operating system or

January, 1994 The DECUServe Journal Page: 8

hardware. When people come to me and ask what hardware to buy (as though I
really knew hardware), I tell them to shop for their software that they really
like/can't live without and then go find the best hardware and OS value which
can run it (if it only runs on one platform, be certain you can't live without
it). If only corporate computing were so simple.
Those are the harsh realities, as I see them.
I very much want to see continued development on VMS, but more than that,
I want WordPerfect to continue to succeed and be profitable and produce the
best software. If DEC were committed to making VMS their preferred, premier,
marketed OS, and easy to develop/port/market software for, many would be
developing software for VMS who have stopped. Otherwise, development effort
spent porting and supporting additional platforms detracts to much from the
primary development of the software for other platforms which give a better
return on investment.
The real sure-fire way to get a 4.0 or AXP version of Office is to buy the
3.1 version like it were going out of style (literally!). Like an election,
every vote counts, but only the accumulation of many votes can succeed. The
voter may also see the odds and feel that he would be wasting his vote to hope
against hope. That is the voter's decision.
Disclaimer: I work for WordPerfect Corporation. I have a vested interest
in the success of WPCorp products (especially on VMS).

Batch ad-hoc report-writer for Oracle
-------------------------------------

The following article is an extract of the DECUServe Databases
conference topic 319. The discussion occurred between
August 8, 1993 and November 19, 1993.

By John Gorentz, Joe Crum, Tom Sammons, Tom McIntyre, Bill Wood

(09/09/93 Gorentz)
------------------
Does Oracle have a capability similar to the Ingres report writer, whereby
one can process an ad-hoc SQL query + format specification in batch mode and
produce an ASCII file having the resulting data?
We have a setup with some Ingres sites (two VMS and one Sun) whereby we
e-mail them a report specification. A "mail-listener" authenticates the
message, runs the report, and mails the data back. The Ingres report writer
also allows us to also put some identification information on the file, and
put in delimeters so we can easily parse it on our end. One nice thing about
version 6.4 of Ingres is that the reportwriter allows not just an SQL query to
be specified, but also other SQL statements that might create temporary files,
etc. We aren't using this capability now, but might want to in the future.
We want to set up a similar thing at a site using Oracle (on a Sun
workstation). The researcher there understands what we're trying to do, but
doesn't know much about Oracle. I'm having trouble communicating with the dba
he has hired to run Oracle for him. I'm not sure if it's a communication
problem or what, but when I explain what we're doing in Ingres terms, I'm not
getting any useful ideas of equivalent Oracle capabilities back.

January, 1994 The DECUServe Journal Page: 9

The researcher has faxed me the tables of contents from several Oracle
manuals, from which I'm picking some likely manuals to get, so I can study the
capabilities for myself. But in the meantime, I'm wondering if any Oracle
people here can tell me whether we're going to find what we need.
Here are some criteria:

It needs to run in "batch" mode.

It needs to accept ad-hoc SQL queries. (It would be good, for future
developments, if it could also process some arbitrary "set-up" statements in
SQL, too.)

We need to be able to control the format of the output. (Remember, the
queries are ad hoc.)

It should allow us to add, in addition to the data from the database,
additional information to be used for identification and to delimit the output
so we can parse it when it's received on our end.

It should use capabilities that are likely to be fairly stable from one
release of another. (I understand from comments made in this conference (or
was it elsewhere) that some of the Oracle interfaces are notorious for changing
from release to release.)

We should do this without writing any 3GL code.

Any ideas? Does the Oracle report writer have any capabilities like this?
The TOC for the SQL*ReportWriter Reference Manual (Version 1.1) seems to talk
mostly about interactive features. It has a lot of entries about "screens",
which is not at all what we want. The SQL*Plus User's Guide and Reference
(Version 3.0) has a chapter on "Formatting Query Results." This sounds
promising, but I'm notorious for reading my own wishful thinking into product
descriptions. Is it likely that this will give us what we need?

(08/09/93 Crum: Can be done, but...)
-------------------------------------
I don't know of any Oracle report writer that easily does what it sounds
like you are doing with Ingres, but it can be hacked by having SQL*Reportwriter
generate the SQL query, then sending the SQL file to Oracle via batch, and
routing the report back. But it requires a lot of coding to do.
Is the "mail listener" a native part of Ingres, and does it really use
e-mail to send the query and return the report?
Is it easy to use?

(08/09/93 Gorentz: e-mail for SQL queries and resulting data)
--------------------------------------------------------------
>Is the "mail listener" a native part of Ingres, and does it really use
>e-mail to send the query and return the report?

No, it's not part of Ingres, but yes, it uses e-mail. On one VMS system
we use PMDF deliver to help out; on another a self-resubmitting batch job does

January, 1994 The DECUServe Journal Page: 10

it. (Yuck.) On the Sun Sparcstation, cron does some scheduling. (I've had
to have help with that part and the writing of the script -- I'm not very
familiar with things Unix.) We use uudecode and uuencode to encode and decode
the queries and data to make sure they're mailable. The query mechanism is
smart enough to figure out how to break down the query into small chunks such
that no single e-mail message containing data is too big.
Easy to use? It's pretty simple, but at the levels I'm describing, the
messages are all sent and received by machine. And we take measures to make
sure that Joe Hacker can't send a big query to one of what we call the "SQL
Servers" to create temporary tables using up all available disk space and/or
sending back tons of data to clog up the network.
If you're really interested, I can post a description of the whole thing.
It's really just a working prototype, but it's intended so that Joe User
anywhere on the Internet can e-mail a query asking for climate data from one
or more separate databases at different sites, and receive the requested data
back via e-mail. The intellectually and technically interesting part (to me)
is an automated method of merging and aggregating the data from these
independently designed, independently managed databases into a canonical form.
It's being done for something called the Long Term Ecological Research
Network, which consists of 17 sites in the U.S. (including Puerto Rico) and
Antarctica, and is funded by the National Science Foundation. The sites
involved in the prototype are in Alaska, Colorado, Wisconsin, Kansas, and
Michigan. The Oracle site is at Kansas State University. No, you can't use
it to get weather data from Antarctica -- at least not yet.
And we aren't ready to make it public yet, anyhow. At some time, though
we hope to invite people to try it out. It's been moribund for a few months,
but there has been a flurry of activity in the past week or so to reactivate it.
If you'd like to know more, I'll post more. But I probably won't be able
to until late next week. (And there's a slight possibility that budget cuts
will leave it high and dry. I'll probably find out next week.)

(08/09/93 Gorentz: This chapter sounds promising)
--------------------------------------------------
>I don't know of any Oracle report writer that easily does what it sounds like
>you are doing with Ingres, but it can be hacked by having

The stuff in chapter 4 in the SQL*Plus User's Guide and Reference sounds
almost like a report writer. I'm having this manual sent to me. Do you know
if there is any relationship between what is described there and the
SQL*Reportwriter?

(08/10/93 Sammons)
------------------
>The query mechanism is smart enough to figure out how to break down the query
>into small chunks such that no single e-mail message containing data is too
>big.

I'd like to see this part, posted, John...
We're (finally) getting into data warehousing, and this would/could be
something that we might need as time goes on...
--- Tom_who_doesn't_particularly_care_to_reinvent_the_wheel

January, 1994 The DECUServe Journal Page: 11

(08/10/93 Crum: Depends on report complexity... )
--------------------------------------------------
>The stuff in chapter 4 in the SQL*Plus User's Guide and Reference sounds
>almost like a report writer. I'm having this manual sent to me.

SQL*Plus is just Oracle's implementation of SQL. It does include some
rudimentary formatting capability, and you can generate simple reports with it.
I have used it for lots of reports, and I've never used SQL*Reportwriter,
though some folks I work with have used it. SQL*Reportwriter is a separate
product (read "more money") which is unrelated to SQL*Plus except for, I
suspect, a common ancestry.
Disclaimer: I don't have any relationship to Oracle, except as a griping
customer. 8^}

(08/17/93 Gorentz: Uncomplexity)
---------------------------------
>SQL*Plus is just Oracle's implementation of SQL. It does include some
>rudimentary formatting capability, and you can generate simple reports with it.

Rudimentary may be all I need. E.g. for numeric data, we were putting
each number on a separate line.
I hope to see the manuals soon.

(08/17/93 Gorentz: Glad to help, but this part is pretty mundane)
------------------------------------------------------------------
>I'd like to see this part, posted, John...

I may be digging this out and blowing the dust off in a few days (I hope.)
However, the reason the query mechanism is smart enough to do this is that we
are dealing only with stupid data. Er, maybe I should say instead that we are
working only with some simple, two-dimensional tables that each have some sort
of date-time variable(s) as the key. Each request specifies the variables to
be retrieved, and the range of date-times. The data are (mostly) collected at
regular intervals. There is no rocket science here. With these constraints,
it's fairly simple to estimate the maximum amount of data that will be returned
in response to a request.
There *are* some aspects of this project that are mildly interesting, having
to do with the variety of types of date-time intervals involved. But the part
about breaking the query into chunks is so far pretty mundane. I hope I didn't
mislead you.
On a somewhat related issue:
At the Atlanta DECUS I learned that PMDF has (or is going to have?) some
mechanism for dealing with multi-part messages. I don't mean where a single
e-mail message contains >1 smaller entities, but where some large piece of
data is sent in multiple e-mail messages. I'm going to look into this before
doing *anything* more with this part of the project.

January, 1994 The DECUServe Journal Page: 12

(08/17/93 McIntyre: You might want to look at the KEY mail server)
-------------------------------------------------------------------
When I set up the mail service for KEY, I put in a REPORT verb that
generates a separate mail message. A single request can have several occurences
of the REPORT verb. There is also an address change capability so that a
single request can mail reports to several different addresses.
You can get documentation on the KEY mail server by sending mail to
decusp...@eisner.decus.org with "document" as the subject.

(11/19/93 Gorentz: Date-time grumbles)
---------------------------------------
From working Oracle 6.0, I've gathered that it doesn't have the rich
repertoire of date-time functions and formats that Ingres has. In particular,
it's a nuisance that I have to express dates in the format DD-MON-YY. For
example, as in the clause, "where time_id > 15-MAR-93". Note the two-digit
year.
It could become more than a nuisance. In the entire "distributed" database
I'm working with, we already have over a hundred years worth of climate data.
The 19th century data we have thus far happen to reside in an Ingres database
rather than in Oracle. But another century is coming up shortly. Are the
Oracle people not thinking ahead?
In the meantime, it's a nuisance that I have to express SQL queries one way
for Ingres and another, inferior way for Oracle. Or is it possible that I'm
missing out on something? I don't have all the Oracle manuals, but the
SQLPLUS manual is quite explicit in stating that the format is DD-MON-YY, and
that I have to use conversion functions to express dates in any other format.
I wish there were an environment variable I could set, or something of the kind.

(11/19/93 Wood: There appear to be options in V7.0.)
-----------------------------------------------------
In Oracle 7.0 there are several options to handle other data processing.
There is a function TO_DATE which allows the expression of dates in a different
format (although this does not allow compatibility with Ingress). There is
also an initialization parameter NLS_DATE_FORMAT which appears to allow one to
change the date expression for the instance from DD-MON-YY to something else
like DD-MON-SYYYY (where S is set to "-" for dates which are B.C.)

DECnmail (Nmail) R 9.3-4
------------------------

The following article is an extract of the DECUServe
DEC_Software conference topic 626. This discussion occurred
between November 5, 1993 and December 18, 1993.

By Larry Kilgallen, Don Vickers, Terry Kennedy, John Osudar, Harrison Spain,
John McMahon, Richard Marino, Sharon Frey

January, 1994 The DECUServe Journal Page: 13

(11/05/93 Kilgallen)
--------------------
>DECnmail (Nmail) R 9.3-4

This software is installed on ISVnet, and it is really neat.
I suppose for some it may seem to duplicate capabilities of third party
products, but Nmail has impressed me as requiring minimal change in usage
patterns.

(11/05/93 Vickers: Don't leave your node (via VMSmail) without it)
-------------------------------------------------------------------
NMAIL (or DECnmail) is a great tool for anyone who uses VMSmail in a
networking environment. It provides a form of store and forward processing.
It does the actual message sending in batch so the sender is freed from having
to wait or worry about whether the links between systems are up.
The best approach is to use the (largely undocumented) SET TRANSPORT
command in mail:

MAIL> SET TRANS NM%

will cause NMAIL to be used for all of your outgoing VMSmail. This makes the
use of the tool virtually transparent to the user.
It's great that this great tool is FINALLY available for real customers.

(11/06/93 Kennedy)
------------------
> NMAIL (or DECnmail) is a great tool for anyone who uses VMSmail in a
> networking environment. It provides a form of store and forward

Having once tried to mail a long message from inside DEC to outside (via
DECWRL::"internet-address", I can see why this is needed inside Digital - it
took *forever* to do it with vanilla VMS mail.
On the other hand, how does this compare with commercial (and free) products
like PMDF and MX, which let you do either SMTP-over-DECnet or DECnet Mail-11 in
batch mode?

(11/06/93 Osudar: (yawn...) another missed boat)
-------------------------------------------------
Great! And it's only about six years too late for many of us (who relied
heavily on DECnet back in the late 1980's, but now rely mostly on TCP/IP for
networking and SMTP for electronic mail). After I saw NMAIL in the HELP files
on a VAXFAM system at a DECUS symposium in 1986 or 1987 (and was told that
NMAIL was indeed terrific, but definitely not available to the great unwashed
customer public), I cobbled up my own version of a queued DECnet mailer, which
we used for several years. I think the last serious user switched to SMTP
around 1991.
Keep in mind that my site was once a major DECnet user on what's allegedly
the second largest DECnet in the world, HEPnet/ESnet. Of course, we also used
to be a major Digital customer (our VAX managers' group represented about 20

January, 1994 The DECUServe Journal Page: 14

internal organizations that depended upon VAXes and VMS for their computing);
now we could hold a Digital users' meeting in a small closet with room to spare.
Sorry, but this NMAIL thing reopened some old wounds...

(11/07/93 Spain)
----------------
We like NMAIL since it does *not* add the NM% to the address! :-) The only
oddity we have found is that address node::username becomes
node::node::username :-).

(12/07/93 McMahon: Ancient History)
------------------------------------
I have to echo John's sediments here, having also fought this war and lost.
From 1986 to 1990 I worked on the NASA side of the World DECnet, and somewhere
during that time John and I first crossed paths :-) We needed this tool badly
back then, and I suspect that we could have sold a copy to every major system on
SPAN, HEPnet and it's european counterparts. Every symposium I went to I asked
for this tool, and every symposium I was told that they would research it and
let us know.
In 1990, I left DECnet behind for the most part. My legacy to this story is
that the MultiNet SMTP Symbiont supported a similar syntax for queueing DECnet
mail (SMTP%foo::bar), I managed to convince them to make it a documented
feature. :-)

((12/07/93 Marino: Can DECnMail do this?)
------------------------------------------
Re DECnmail; I talked to DEC here at DECUS and they say DECnmail won't do
what I want to do, but I'm inclined to be a doubter, so I'm curious if anyone
else think DECnmail could do this (or what are other alternatives):
I have an office in Madison with a VAX; we plan to install a small VAX in
an office in the UK. Currently the people in the UK dialup into our VAX
directly or using a dialup telnet service in the UK. It would be ideal if
they could just use VMSMAIL on their new VAX to send us mail but we won't have
a permanent connection between the systems.
Ideally DECnmail will queue the mail sent to the unreachable node and then
we could periodically and automatically dial from node to node and make an
DDCMP connection and all the mail between the two sites would transfer. DEC
says DECnmail can't create the link, they suggest VAXModem (another assets
package) be used but they don't know if it can create a DECNET link over modems.
BTW, I talked to PDMP and they say their software can do it. I guess I'll
post this question also in the Third Party Area.

(12/07/93 Kennedy: Pointer to answer in 3rd_party)
---------------------------------------------------
Please see my answer over in 3rd-party. It's free and it will do exactly
what you need.

January, 1994 The DECUServe Journal Page: 15

(12/16/93 Frey: :-) )
----------------------
>I have to echo John's sediments here, having also fought this war and lost.
^^^^^^^^^

Are you saying that John has a dirty mind?

(12/16/93 Osudar: silliness :-) )
----------------------------------
No, he's just saying that all the stuff floating around in there has
settled to the bottom... :-)
(Besides, dirty minds are prohibited by both DECUServe canons and U.S.
government regulations... ;-)

(12/18/93 Vickers: _JUST_ for the record - NMAIL will do what .6 wishes)
-------------------------------------------------------------------------
NMAIL will indeed operate in an environment where the link is only
available for very short periods of time. It cannot create the link, as you
know, but it will use it when it is available. It has timers and retry
mechanisms which can be adjusted to have it funnel the messages during the
period of availability. One mechanism would be to have it operate in a batch
queue which is started in conjunction with the opening of the link.
VAXmodem will indeed dial and establish DECnet links very easily. I used
both it and NMAIL extensively when I was a remote employee at Digital.
As many have pointed out here and in other places, there are many products
which are cheaper and better supported than the Digital 'offerings'. You
should obtain the software that seems best for your needs, of course. There
MAY be good reasons to buy these ASSETS packages but my guess is that there
are fewer of them every day. Digital is proving less and less interested in
being a software company and seems to be getting less and less able to
understand what customers need. Just listen to anything Mr. Palmer says.

PostScript printing on HP Laserjet III
--------------------------------------

This article is an extract of the DECUServe Office_Automation
conference topic 135. The discussion occurred between
September 26, 1991 and December 2, 1992.

By Forrest Towne, Lyndon Bartels, Jim Scanga, Bob Graham, Bill Norton

(09/26/91 Towne)
----------------
Has anyone had experience with getting an HP Laserjet III with the
PostScript cartridge installed working. I have the HP connected using reverse
LAT. I need to print to it from VMS and from PCs. I am running VMS 5.4 and
Pathworks 4.0. Must of the PCs are running MS-DOS 4.01, we are in the middle

January, 1994 The DECUServe Journal Page: 16

of upgrading to MS-DOS 5.0. I have DECprint 4.0 installed on the VAX. I have
able to copy the CPS$IVP_POST.DAT test file successfully. However when I
print it, I get a known PostScript data error on the printer front panel using
the same file. I have also setup a normal queue using the LAT symbiont, with
the same results. I have turned off the burst, flag, and trailer pages. A17
help would be appreciated.

(09/27/91 Bartels: Yes, it is possible !!!)
--------------------------------------------
Yes, I got it to work and it works great!
A couple things. First the LTA terminal/device/port whatever it's called
has to be set up just so. (So says Colorado.) Second there has to be some
modifications made to the device control library that DECprint printing
services uses. This is of course assuming the HP postscript cartridge to the
III. It's not too tough. And what's neat you can tell the print headers to say
"HP Laserjet III" instead of "DEClaser 2150". Also after you get this working,
spend the $$$ and get the DECprint printing services manual(s). I can't
remember how much they cost, but they have some neat stuff in them. And, if
you have the greenbacks just "burning a hole in your pocket" I can suggest a
couple PostScript books too. Give me a call and I can fill you in more.

(09/28/91 Towne: Still without PostScript)
-------------------------------------------
> First the LTA terminal/device/port whatever it's
> called has to be set up just so. (So says Colorado.)

I used the TSM com files in SYS$EXAMPLES.CPS, is ther something else I
need to do.

> Second ther has to be some modifications made to the device
> control library that DECprint printing services uses.

Any idea what the changes are or where I can look them up.

> This is of course assuming the HP postscript cartridge to the III.

Exactly.

> Also after you get this working, spend the $$$ and get
> the DECprint printing services manual(s).

I have the DECprint Printing Services manual.
I guess I'm still lost as how to proceed. At least I know it is possible.
If possible could you show me how your LAT terminal port is setup, full queue
display of generic and execution queues, form definition, and changes made to
the device control library. I hope that's not to much to ask. I really
appreciate the help. :=)

January, 1994 The DECUServe Journal Page: 18

(09/30/91 Bartels: Give these "modifications" a try.)
------------------------------------------------------
There are four (4) modules in the CPS$DEVCTL library that take care of this
problem. First, I suggest you should make a copy of the library and work with
the copy. Then in CPS$startup, where it asks for the device control library,
put in the new one. The modules I modified are as follows.

LPS$$Printertype,
LPS$$LNXX_Initpsdevice,
LPS$$Loaddict,
LPS$$SystemPag19

First LPS$$Printertype. Add the following lines at the beginning.

statusdict begin
/product (DEClaser 2150) def
end

This should allow DECprint to believe the printer is a DEClaser 2150.
NEXT LPS$$LNXX_Initpsdevice

statusdict/checkconfiguration known not
{ statusdict begin
/checkconfiguration { true } def
end
}
if
statusdict begin
/product (DEClaser 2150) def
/returnstatus {cleartomark} def
end

This sets various variables to the proper definitions to continue "fooling"
DECprint

Next LPS$LOADDICT

About the second line of code you find, near the end "/lps$pn ..."
The next line you'll find "{pop}ifelse/lps$input-tray ..."

Between these two make sure there is
"(Hewlett-Packard LaserJet) def}"

This should be at the beginning of the third line of code. I don't know
what this does, nor do I know if this step is really needed, but...

Finally, LPS$$Systempag20

Near the end there is "/systemid" followed by a bunch of stuff including
"Digital Equipment Corporation" ( I think )
I replaced "(Digital Equipment Corporation)" with our company name
enclosed in parenthesis.

January, 1994 The DECUServe Journal Page: 19

Near the end there is "/symbiontid" followed by a bunch of hen-scratching.
This is what I have.

/symbiontid {
helvbold-12 setfont
558 72 position
(Hewlett-Packard LaserJet)
rightshow
helvbold-18 setfont
558 50 position
rightshow
} bdef

All these do, as far as I can see, is make your flag sheets look different.
These two areas are near the bottom of the sheet. I think that's all I did.
I know those were the only four modules I modified. A bit about Postscript if
you don't know.
1: It IS case sensitive.
2: A character string is delineated by parens. Not quotation marks.
(So the quotation marks I entered are my quotation marks for
this message and should not be part of the text.)

I hope these work for you. About the lta. If I remember correctly, there
has to be a "/secure" qualifier in the creation or something. But I'm sure
since you followed the example thing, it's probably correct.
Give me a call if you need more info.

(09/30/91 Towne: Closer, but still need help)
----------------------------------------------
> This should allow DECprint to believe the printer is a DEClaser 2150.

Things seem to be better. I still get "PS ERROR 18", which I believe is
caused by invalid PostScript command. However I am now getting a trailer. The
trailer appears to be intact. There are two messages.

30-SEP-1991 12:36 %LPS-W-UNDEF, undefined: Name not known - offend command is
dopa22
30-SEP-1991 12:36 %LPS-E-FLUSHING, Rest of Job (ro EOJ) will be ignored

I did some research. I extracted all the modules from HPIII$DEVCTL (my
copy of CPS$DEVCTL), I then searched them for "dopage", only the modu23
LPS$$LN03R_INITPSDEVICE contains a "dopage" command. I assume that the syst24
still thinks I'm using a LN03R rather than a DEClaser 2150.
Any ideas on where I might have messed up.

(10/01/91 Bartels: Try, try, again.)
-------------------------------------
Actually, no, the symbiont does NOT think you're running a LN03. It thinks
you have a DEClaser 2150. The "offending command" error means that it cannot
find the "dopage" command. not because it is a bad comman25

January, 1994 The DECUServe Journal Page: 20

In sys$examples there, I think there is a CPS subdirectory or at least some
CPS related files. One of them has a bunch of stuff in for "work arounds."
These are usually definitions and stuff for various postscript commands. A
postscript definition starts with a "/" and ends with a "def" type of thing.
For example. "/Weird_string (This is a weird string.) def" Look for something
like that in there. Or look in that LN05R module you found the "dopag27
command. You might find something there as well.

As for anything else, give me a call. I also suggest you get the following
books:
PostScript Language Reference Manual. Second Edition.
Understanding POSTSCRIPT programming.

They're pretty good for looking things up and helping you find out stuff.
My understanding of Postscript is minimal, but I'll help. Give me a call if
you have more problems.

(10/10/91 Scanga: How about a MAC?)
------------------------------------
I didn't quite follow the replies here but that's not going to stop me,
also I'm not quite sure how things tie together. With that said...
I have an HP Laserjet IIISi for evaluation with a postscript driver
installed. I had no trouble creating a queue on the VAX and having my PCs
print postscript via Pathworks 4.0 to this printer. Non-postscript also prints
fine from the VAX. The VAX queue has been created with a third party print
symbiont product called ScriptServer.
The problem lies with my MACs. How do I get these to print to this printer?
I created a queue within pathworks for the MAC software, but when I try to
print I get the following error:
The chosen printer has an incompatible version of the laserprep file and
cannot be changed. You must manually reset the printer or change your version
of the laserwriter software to be compatible.
DEC tells me that they don't support printing to the HP Laserjet from
pathworks for MACs. Anyone out there doing this? I'm guessing I need some
special laser prep file for my MAC.

(10/11/91 Bartels: I don't know about Scriptserver, but DECprint works.)
-------------------------------------------------------------------------
Although I don't have a IIIsi, I can print to my III from my Macintosh's.
We run Pathworks for MAC V1.0 for the Macintosh side, and a modified DECprint
printing services for the VMS side of the equation. DECprint suggests setting
up two (or more) generic queues feeding the one terminal queue. I simply set
up a Printserver whatchamacallit for the Macs that pointed the Postscripting
Generic queue and everything seems happy. I, personally, tried it on the Macs
and it does work. If you're not using DECprint, I'm stumpified.

January, 1994 The DECUServe Journal Page: 21

(10/11/91 Graham: need MSAP$PROCSETn modules)
----------------------------------------------
>The chosen printer has an incompatible version of the laserprep file and
>cannot be changed. You must manually reset the printer or change your version
>of the laserwriter software to be compatible.

Not having ScriptServer here, I can't be sure, but I suspect that your
problem is that the device control library that you have for the ScriptServer
symbiont doesn't have the modules that the Pathworks print receiver is looking
for.
When you start a Pathworks for MAC queue, a detached printer receiver
process is created to "receive" the incoming printer jobs from the MACs.
Essentially, this receiver process is emulating a network attached LaserWriter.
One of the things that a MAC does before sending a print job is to ask the
LaserWriter whether the printer has the same version prep file loaded as the
driver expects. The receiver process decides how to answer this by looking at
the device control library assosicated with the queue, and checking for modules
named MSAP$PROCSETn, where "n" is currently 1 thru 5. MSAP$PROCSET5 is
laserprep v7, MSAP$PROCSET4 is v6.0, etc.
The modules are inserted into the CPS$DEVCTL.TLB library of DECprint v4.0
when Pathworks for MAC is installed. I doubt if the Pathworks installation
inserted them into whatever device control library your ScriptServer symbiont
uses.
I don't know if you can just extract the modules from the Pathworks kit and
insert them into the ScripServer symbiont or not. Aside from the legal aspects,
they are "optimized" for the DECprint environment and might not work for
ScriptServer.

(10/24/91 Towne: HPIII PostScript problems persist)
----------------------------------------------------

(10/28/91 Scanga: It's working)
--------------------------------
>I doubt if the Pathworks installation inserted them into whatever device
>control library your ScriptServer symbiont uses.

The above is exactly right. When we changed to the laserprep file for
Version 7 we didn't update the modules in the scriptserver device control
library.

>I don't know if you can just extract the modules from the Pathworks kit
>and insert them into the ScripServer symbiont or not.

We did the above and this works fine, except for some cryptic error message
on the cover page that doesn't seem to effect the output.
Thanks for your help.

January, 1994 The DECUServe Journal Page: 22

(12/02/92 Norton: HP LaserJet 4 & CPS$SMB)
-------------------------------------------
We tried the mods suggested in 135.3, and are succesfully running LaserJet
4's under CPS$SMB. We only modified LPS$$PRINTERTYPE and
LPS$$LAXX_INITPSDEVICE, and it seems to work. I don't know what happens when
the printer complains about something during a job. Likely CPS$SMB doesn't
know how to handle HP backtalk.
As for official DEC support of HP printers, a CSC DECprint support person
said about Jan 1, '93, is the scheduled release of something called DECPrint
Supervisor, or DCPS, that supports several 3rd-party printers, including some
HP's.

High Speed modems for disaster recovery?
----------------------------------------

The following article is an extract of the DECUServe
Telecommunications conference topic 92. The conversation
occurred between December 16, 1993 and December 17, 1993.

By Rick Bowen, Keith Chadwick, Brian Tillman, Matt Holdrege, Linwood Ferguson

(12/16/93 Bowen)
----------------
With recent improvements in modem technology I was wondering if anyone is
using dial-up high speed modems for passing ethernet traffic. We are starting
to take a look if the technology is sufficient to provide emergency backup to
dedicated circuits.
Any experiences would be helpfull.

(12/16/93 Chadwick: Ethernets can be bridged via ISDN)
-------------------------------------------------------
We have been experimenting with Ethernet to ISDN "half bridges" (the ones
that we are using are manufactured by Combinet).

========================Ethernet================
|
+----------+ +----------+
| Combinet |-------ISDN-------| Combinet |
+----------+ +----------+
|
========================Ethernet================

We are in the process of putting together a pilot project to move them
into production.

January, 1994 The DECUServe Journal Page: 23

(12/16/93 Tillman: Our LUG meeting just had a presentation about this)
-----------------------------------------------------------------------
Our LUG meeting last Tuesday dealt with that very thing. There are some
great high-speed stat muxes out there that can be configured to run main
traffic over a 56K bridge, but revert to dial-up lines (multiple lines, too,
aggregating them to obtain enough bandwidth) in case of a failure. The
information I received can from a company named Data Race, 11550 IH 10 West,
San Antonio, TX 78230, Tel 210-558-1900, FAX 210-558-1929. One of the big
selling point of muxes like this are that you get phone connectivity for "free".
That is, phone traffic can be piggybacked over your data pathway and you don't
have to pay the phone company separately for the phone traffic if you call the
same people with whom you're networking.

(12/16/93 Holdrege: V.FAST)
----------------------------
We use Motorola Codex V.FAST modems with Synchronous Data Compression. We
use these if we have to bring up a new office faster than we can get a 56K
circuit installed. If you use routers, this works great. If you use bridges,
you must make sure that you keep the broadcast down to a minimum.

(12/16/93 Ferguson: Significant speed differ in same standard different brands)
--------------------------------------------------------------------------------
Very good article comparing various modem speeds for compressed transfer
on a variety of lines in last weeks (or week before) Network World.
Sorry, mines gone (I meant to save it but somehow it dissappeared).

(12/17/93 Bowen: V.FAST with 3Com?)
------------------------------------
>a 56K circuit installed. If you use routers, this works great. If you
>use bridges, you must make sure that you keep the broadcast down to a

Have you tried these in conjunction with the dial up option on your
NetBuilders?
We would use them for emergency backup to T1 circuit so the alternate path
would filter out a lot of normal traffic. Probably cut out entire protocols
and implement call of service.

(12/17/93 Holdrege: not yet)
-----------------------------
We haven't had a chance to test dial restoral with the V.fast modems yet.
It works with Adtran switched 56K units though.

January, 1994 The DECUServe Journal Page: 24

crontab file format
-------------------

The following article is an extract of the DECUServe Unix_OS
conference topic 221. This discussion occurred between
November 13, 1993 and November 15, 1993.

By John Gorentz, Terry Kennedy, Matt Holdrege

(11/13/93 Gorentz)
------------------
Does anyone know where I can read about the proper syntax and format of a
crontab file?
I am mostly Unix illiterate.
I am using a Sun Sparc-something running SunOS 4.1.1 at another site, far
away. In other words, I don't have any manuals. I need to set up a script
that will be executed by cron. I have copies of a script and crontab file,
etc., that are performing a similar function at yet another site (also far
away). But I need to modify the crontab file. I can guess at the meaning of
some of the pieces of information in it, but I'd rather not do this by
guesswork.
I've got the man pages for cron and crontab, but they don't tell what to
put in a crontab file. Would this information normally be available in a file
somewhere on the system? The man pages don't give me any clues.
Is this information likely to be SunOS specific, or is there some generic
Unix book I could find it in?

(11/14/93 Kennedy: Manual section 5 (usually), copy attached)
--------------------------------------------------------------
My other Unixes have a crontab entry for both the program (section 1) and
the file format (section 5). My SunOS 4.1.1 test system died a horrible death
Saturday (hard disk positioner failure), so I can't verify it's the same on
SunOS. However, here's the manual page from a PD cron clone:

[0:2] spcuna:~> apropos cron
cron (8) - daemon to execute scheduled commands (Vixie Cron)
crontab (1) - maintain crontab files for individual users
crontab (5) - tables for driving cron
[0:3] spcuna:~> man 5 crontab

CRONTAB(5) CRONTAB(5)

NAME
crontab - tables for driving cron

DESCRIPTION
A crontab file contains instructions to the cron(8) daemon
of the general form: ``run this command at this time on
this date''. Each user has their own crontab, and com-

January, 1994 The DECUServe Journal Page: 25

mands in any given crontab will be executed as the user
who owns the crontab. Uucp and News will usually have
their own crontabs, eliminating the need for explicitly
running msu as part of a cron command.

Blank lines and leading spaces and tabs are ignored.
Lines whose first non-space character is a pound-sign (#)
are comments, and are ignored. Note that comments are not
allowed on the same line as cron commands, since they will
be taken to be part of the command. Similarly, comments
are not allowed on the same line as environment variable
settings.

An active line in a crontab will be either an environment
setting or a cron command. An environment setting is of
the form,

name = value

where the spaces around the equal-sign (=) are optional,
and any subsequent non-leading spaces in value will be
part of the value assigned to name. The value string may
be placed in quotes (single or double, but matching) to
preserve leading or trailing blanks.

Several environment variables are set up automatically by
the cron(8) daemon from the /etc/passwd line of the
crontab's owner: USER, HOME, and SHELL. HOME and SHELL
may be overridden by settings in the crontab; USER may
not.

(Note: for UUCP, always set SHELL=/bin/sh, or cron(8) will
cheerfully try to execute your commands using
/usr/lib/uucp/uucico.)

(Another note: the USER variable is sometimes called LOG-
NAME or worse on System V... on these systems, LOGNAME
will be set rather than USER.)

In addition to USER, HOME, and SHELL, cron(8) will look at
MAILTO if it has any reason to send mail as a result of
running commands in ``this'' crontab. If MAILTO is
defined (and non-empty), mail is sent to the user so
named. If MAILTO is defined but empty (MAILTO=""), no
mail will be sent. Otherwise mail is sent to the owner of
the crontab. This option is useful if you decide on
/bin/mail instead of /usr/lib/sendmail as your mailer when

15 January 1990 1

CRONTAB(5) CRONTAB(5)

you install cron -- /bin/mail doesn't do aliasing, and
UUCP usually doesn't read its mail.

January, 1994 The DECUServe Journal Page: 26

The format of a cron command is very much the V7 standard,
with a number of upward-compatible extensions. Each line
has five time and date fields, followed by a command.
Commands are executed by cron(8) when the minute, hour,
and month of year fields match the current time, and when
at least one of the two day fields (day of month, or day
of week) match the current time (see ``Note'' below).
cron(8) examines cron entries once every minute. The time
and date fields are:

field allowed values
----- --------------
minute 0-59
hour 0-23
day of month 0-31
month 0-12 (or names, see below)
day of week 0-7 (0 or 7 is Sun, or use names)

A field may be an asterisk (*), which always matches the
current time.

Ranges of numbers are allowed. Ranges are two numbers
separated with a hyphen. The specified range is inclu-
sive. For example, 8-11 for an ``hours'' entry specifies
execution at hours 8, 9, 10 and 11.

Lists are allowed. A list is a set of numbers (or ranges)
separated by commas. Examples: ``1,2,5,9'', ``0-4,8-12''.

Step values can be used in conjunction with ranges. Fol-
lowing a range with ``/<number>'' specifies skips of the
number's value through the range. For example, ``0-23/2''
can be used in the hours field to specify command execu-
tion every other hour (the alternative in the V7 standard
is ``0,2,4,6,8,10,12,14,16,18,20,22'').

Names can also be used for the ``month'' and ``day of
week'' fields. Use the first three letters of the partic-
ular day or month (case doesn't matter). Ranges or lists
of names are not allowed.

The ``sixth'' field (the rest of the line) specifies the
command to be run. The entire command portion of the
line, up to a newline or % character, will be executed by
the user's login shell or by the shell specified in the
SHELL variable of the cronfile. Percent-signs (%) in the
command, unless escaped with backslash (\), will be
changed into newline characters, and all data after the
first % will be sent to the command as standard input.

Note: The day of a command's execution can be specified by

January, 1994 The DECUServe Journal Page: 27

15 January 1990 2

CRONTAB(5) CRONTAB(5)

two fields -- day of month, and day of week. If both
fields are restricted (ie, aren't *), the command will be
run when either field matches the current time. For exam-
ple,
``30 4 1,15 * 5'' would cause a command to be run at 4:30
am on the 1st and 15th of each month, plus every Friday.

EXAMPLE CRON FILE
# use /bin/sh to run commands, no matter what /etc/passwd says
SHELL=/bin/sh
# mail any output to `paul', no matter whose crontab this is
MAILTO=paul
#
# run five minutes after midnight, every day
5 0 * * * $HOME/bin/daily.job >> $HOME/tmp/out 2>&1
# run at 2:15pm on the first of every month -- output mailed to paul
15 14 1 * * $HOME/bin/monthly
# run at 10 pm on weekdays, annoy Joe
0 22 * * 1-5 mail -s "It's 10pm" joe%Joe,%%Where are your kids?%
23 0-23/2 * * * echo "run 23 minutes after midn, 2am, 4am ..., everyday"
5 4 * * sun echo "run at 5 after 4 every sunday"