Discussion:
SFTP
Add Reply
W Mainframe
2017-12-06 01:46:17 UTC
Reply
Permalink
Raw Message
Come on.....For years FTP works.... Now... Every OS needs to upgrade for SFTP?Only now, this protocol is unsure???? One more stupid option... bla. bla bla



Sent from Yahoo Mail for iPhone

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
Paul Gilmartin
2017-12-06 02:58:47 UTC
Reply
Permalink
Raw Message
Post by W Mainframe
Come on.....For years FTP works.... Now... Every OS needs to upgrade for SFTP?Only now, this protocol is unsure???? One more stupid option... bla. bla bla
Yup. Technology advances alike on the benign and the malicious sides. The former
must advance to keep up with the latter.
https://en.wikipedia.org/wiki/Red_Queen%27s_race

FTP grew in a heterogeneous envirnment and has accommodations for UNIX,
OS/360, VAX, TOPS-10, ... Somewhat dismayingly, sftp is inexorably UNIX-centric.
Co:Z tries to address this deficiency.
https://dovetail.com/products/sftp.html

FTPS (is this the same as FTP with AT/TLS?) requires elaborate setup and firewall
exceptions, partly because of the auxiliary data socket.

-- gil

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
Edward Gould
2017-12-06 06:40:06 UTC
Reply
Permalink
Raw Message
——————————SNIP------------------------------------------
FTP grew in a heterogeneous envirnment and has accommodations for UNIX,
OS/360, VAX, TOPS-10, ... Somewhat dismayingly, sftp is inexorably UNIX-centric.
Co:Z tries to address this deficiency.
https://dovetail.com/products/sftp.html
FTPS (is this the same as FTP with AT/TLS?) requires elaborate setup and firewall
exceptions, partly because of the auxiliary data socket.
— gil
Gil (or anyone):
Does dovetail ship their products in SMPE installable format or ? The company I am helping has a issue with *ANY* product that is not SMPE installable.
Before I go to bat for Dovetail, I would like to know that issue, if not I won’t bother. I tried to use their forum and it won’t let me ask the question as to cost for Gold for all 3 of their products. I refuse to spend my time for a simple question.

Ed
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
John McKown
2017-12-06 13:43:11 UTC
Reply
Permalink
Raw Message
Post by Edward Gould
Does dovetail ship their products in SMPE installable format or ? The
company I am helping has a issue with *ANY* product that is not SMPE
installable.
Before I go to bat for Dovetail, I would like to know that issue, if not I
won’t bother. I tried to use their forum and it won’t let me ask the
question as to cost for Gold for all 3 of their products. I refuse to spend
my time for a simple question.
​The standard distribution of Dovetailed Technologies' Co:Z products is
_not_ done with SMPE. Basically, it is a "shar" executable. "shar" is a
"SHell ARchive". I guess the closest equivalent would be a Windows "msi"
executable installation program. You download the file into a UNIX
directory; change it's permissions to include executable; then execute it
in a UNIX shell environment. The install script will basically ask you to
agree with the license; specify the UNIX directory to hold the UNIX files;
specify 2 PDS library names to hold the "source" data (JCL + config) and
the z/OS executable programs.

I don't have a maintenance contract, so I don't know how, of if, they
distribute PTFs or "patches". I use the free version. And basically I just
periodically look to see if a new version is available. If so, I install it.
Post by Edward Gould
Ed
--
I have a theory that it's impossible to prove anything, but I can't prove
it.

Maranatha! <><
John McKown

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
Kirk Wolf
2017-12-06 14:24:34 UTC
Reply
Permalink
Raw Message
Co:Z is not SMP/E. We have reasons for that. The requirement has only
come up once from our enterprise customers in ten years.

I'm sorry that you were unable to register for our Community Forum. We
have instructions on how to register on our support page. We had to make
it this way to keep the spammers out. In any case we are not going to
answer sales questions on the free Community technical forum. As
indicated on the support page, send an email to ***@dovetail.com if you
would like pricing information for the Enterprise License and Support
agreements.

Kirk Wolf
Dovetailed Technologies
http://dovetail.com
Post by Paul Gilmartin
On Dec 5, 2017, at 9:00 PM, Paul Gilmartin <0000000433f07816-dmarc-
——————————SNIP------------------------------------------
FTP grew in a heterogeneous envirnment and has accommodations for UNIX,
OS/360, VAX, TOPS-10, ... Somewhat dismayingly, sftp is inexorably
UNIX-centric.
Co:Z tries to address this deficiency.
https://dovetail.com/products/sftp.html
FTPS (is this the same as FTP with AT/TLS?) requires elaborate setup and
firewall
exceptions, partly because of the auxiliary data socket.
— gil
Does dovetail ship their products in SMPE installable format or ? The
company I am helping has a issue with *ANY* product that is not SMPE
installable.
Before I go to bat for Dovetail, I would like to know that issue, if not I
won’t bother. I tried to use their forum and it won’t let me ask the
question as to cost for Gold for all 3 of their products. I refuse to spend
my time for a simple question.
Ed
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
Edward Gould
2017-12-07 10:55:04 UTC
Reply
Permalink
Raw Message
Post by Kirk Wolf
Co:Z is not SMP/E. We have reasons for that. The requirement has only
come up once from our enterprise customers in ten years.
I'm sorry that you were unable to register for our Community Forum. We
have instructions on how to register on our support page. We had to make
it this way to keep the spammers out. In any case we are not going to
answer sales questions on the free Community technical forum. As
would like pricing information for the Enterprise License and Support
agreements.
Kirk Wolf
Dovetailed Technologies
http://dovetail.com <http://dovetail.com/>
You guys really do need to simplify the registration process. *ANOTHER* vendor I deal with is equally a PITA for asking simple questions that are of non really technical issues.
I was asking the guy why they made asking questions so difficult and was it do to SPAM or just a lot of junk questions?
He asked me to wait for a minute and he brought the woman who actually looked at the questions asked and she said well there has never been a spam issue but some questions needed to be sent up to the support people and other questions that they didn’t know who/why the were being asked and emailed the back a info sheet, that seem to resolve all the misc stuff that came in, so I am not sure how much spam you guys got but it helped their sales as like me I get turned off by non answers.
I am reminded of a “simple” product like yours. I believe it was a 22.95 for a tape. They shipped you the source and you were expected to compile and link it. To actually run it you created a top that did a LISTC cat(what ever the catalog name was). Then it passed the file to this progam and it made a prettier IDCAMS list. The problem was that they didn’t write it very well and anytime IDCAMS changed the output *YOU* were expected to patch the program. I had a irate user and she wanted the reformatted output. I suggested that she got a program request and let the programmers deal with it. She was insulted I wouldn’t do it.

Ed



----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
John McKown
2017-12-06 14:40:50 UTC
Reply
Permalink
Raw Message
Not SMPE installable, however the packaging is brilliant!
1) download install file
2) run utility to "explode" install file.
Done. Both USS and z/OS components come out of the same utility.
​I always thought so too. The only enhancement that I could think of would
be if the JCL in SAMPJCL would be automatically customized with the DSNs
(SAMPJCL & LOADLIB) instead of making me do it after the installation. Yes,
I am lazy at times.​
--
I have a theory that it's impossible to prove anything, but I can't prove
it.

Maranatha! <><
John McKown

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
Kirk Wolf
2017-12-06 15:42:17 UTC
Reply
Permalink
Raw Message
That's a pretty good idea John.

As you mentioned, it's "shar"-ish. What is a little unique is that the
post-unload processing is done with a little REXX script. Does that make
up for not using SMPE? :-)


Kirk Wolf
Dovetailed Technologies
http://dovetail.com
Post by John McKown
Not SMPE installable, however the packaging is brilliant!
1) download install file
2) run utility to "explode" install file.
Done. Both USS and z/OS components come out of the same utility.
​I always thought so too. The only enhancement that I could think of would
be if the JCL in SAMPJCL would be automatically customized with the DSNs
(SAMPJCL & LOADLIB) instead of making me do it after the installation. Yes,
I am lazy at times.​
--
I have a theory that it's impossible to prove anything, but I can't prove
it.
Maranatha! <><
John McKown
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
John McKown
2017-12-06 16:12:54 UTC
Reply
Permalink
Raw Message
Post by Kirk Wolf
That's a pretty good idea John.
As you mentioned, it's "shar"-ish. What is a little unique is that the
post-unload processing is done with a little REXX script. Does that make
up for not using SMPE? :-)
​Not for the "check mark" gang.​
Post by Kirk Wolf
Kirk Wolf
Dovetailed Technologies
http://dovetail.com
Post by John McKown
​I always thought so too. The only enhancement that I could think of
would
Post by John McKown
be if the JCL in SAMPJCL would be automatically customized with the DSNs
(SAMPJCL & LOADLIB) instead of making me do it after the installation.
Yes,
Post by John McKown
I am lazy at times.​
--
I have a theory that it's impossible to prove anything, but I can't prove
it.

Maranatha! <><
John McKown

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
Seymour J Metz
2017-12-06 17:36:39 UTC
Reply
Permalink
Raw Message
The point to requiring use of SMP is not to satisfy a checklist, but to provide certain capabilities, e.g., tracking changes, installing fixes without having to do a "level set", backing off bad fixes, suspending the application of fixes with errors or requiring customer action. If the REXX script does everything that SMP does but in a different way then IMHO it's acceptable.


--
Shmuel (Seymour J.) Metz
http://mason.gmu.edu/~smetz3

________________________________________
From: IBM Mainframe Discussion List <IBM-***@listserv.ua.edu> on behalf of Kirk Wolf <***@DOVETAIL.COM>
Sent: Wednesday, December 6, 2017 10:43 AM
To: IBM-***@listserv.ua.edu
Subject: Re: SFTP

That's a pretty good idea John.

As you mentioned, it's "shar"-ish. What is a little unique is that the
post-unload processing is done with a little REXX script. Does that make
up for not using SMPE? :-)


Kirk Wolf
Dovetailed Technologies
http://dovetail.com
Post by John McKown
Not SMPE installable, however the packaging is brilliant!
1) download install file
2) run utility to "explode" install file.
Done. Both USS and z/OS components come out of the same utility.
​I always thought so too. The only enhancement that I could think of would
be if the JCL in SAMPJCL would be automatically customized with the DSNs
(SAMPJCL & LOADLIB) instead of making me do it after the installation. Yes,
I am lazy at times.​
--
I have a theory that it's impossible to prove anything, but I can't prove
it.
Maranatha! <><
John McKown
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
John McKown
2017-12-06 17:45:08 UTC
Reply
Permalink
Raw Message
Post by Seymour J Metz
The point to requiring use of SMP is not to satisfy a checklist, but to
provide certain capabilities, e.g., tracking changes, installing fixes
without having to do a "level set", backing off bad fixes, suspending the
application of fixes with errors or requiring customer action. If the REXX
script does everything that SMP does but in a different way then IMHO it's
acceptable.
​OK, I misunderstood. I interpreted the "must be SMPE installed" as "we
won't even consider any alternate for installation and maintenance. No SMPE
means we will refused to get or use a product."

As to the "shar" script - it does a full install of the product (aka "level
set"). I don't know how Dovetailed Technologies does maintenance for a
company with a maintenance contract. Perhaps Kirk will say.
Post by Seymour J Metz
--
Shmuel (Seymour J.) Metz
http://mason.gmu.edu/~smetz3
--
I have a theory that it's impossible to prove anything, but I can't prove
it.

Maranatha! <><
John McKown

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
Andrew Rowley
2017-12-07 05:56:04 UTC
Reply
Permalink
Raw Message
Post by Seymour J Metz
The point to requiring use of SMP is not to satisfy a checklist, but to provide certain capabilities, e.g., tracking changes, installing fixes without having to do a "level set", backing off bad fixes, suspending the application of fixes with errors or requiring customer action.
The problem with applying fixes a la carte is that you end up with so
many combinations that virtually no-one is running the same combination,
let alone a comprehensively tested combination.

SMP/E does not work well with products that are written in modern
languages (and by modern, I mean languages with optimizers or even just
compilers that check parameters on calls to routines). SMP/E works best
for languages like assembler, where the programmer needs to track their
own dependencies, and products like the operating system where you have
a lot of components developed separately with loose relationships.

When developing a product in a higher level language, the compiler does
much of the work that SMP/E would do tracking interface changes.

As a developer you want to minimize public exposure of internal
interfaces, to simplify future changes and development. If you allow
components to be maintained individually, every interface becomes
quasi-public, in that you don't know how it might be called in the
future. Any change requires careful research of upstream and downstream
effects and writing PRE/IF/COREQs to match. On the other hand, if you do
a whole product replacement the need to keep track of requisites
essentially goes away, it is handled as part of the build. To get an
idea of the problems you could run into, have a play with the
refactoring tools available in modern IDEs and try to figure out how you
could package the changes in SMP/E, other than product replacement.

Individual fixes also prohibit a lot of the optimization that the
compiler can perform. The compiler can look at the program and inline
routines, throw away code that has no effect etc. To do that it has to
be able to examine the program as a whole. If you reserve the right to
replace individual modules then you limit the optimization the compiler
can do.

That's not to say that vendors can't provide a single fix to a specific
problem - they probably can. In many cases they would be able to check
out a specific version from source-control and provide a full
replacement with a single fix. How difficult that is would depend how
different the specific version was to the regular version containing the
fix. (Of course, if offered it might be a premium service at a
corresponding price.)

It's the same as a site's application code - I doubt there are any sites
that package their applications using SMP/E and migrate fixes from
development to production relying on PREREQs and COREQs to manage
dependencies.
--
Andrew Rowley
Black Hill Software

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
Andrew Rowley
2017-12-07 05:57:24 UTC
Reply
Permalink
Raw Message
Post by Edward Gould
Does dovetail ship their products in SMPE installable format or ? The company I am helping has a issue with *ANY* product that is not SMPE installable.
I hope you are encouraging them to modernize their thinking.
--
Andrew Rowley
Black Hill Software

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
Edward Gould
2017-12-07 10:34:18 UTC
Reply
Permalink
Raw Message
Not SMPE installable, however the packaging is brilliant!
1) download install file
2) run utility to "explode" install file.
Done. Both USS and z/OS components come out of the same utility.
That will never be allowed here and I agree. In the past I have seen such thing in other products and it is easily forgotten and I can just see it happening here as there is reasonably high turnover here and after two turnover the staff no longer remembers it. Oh well there goes some money for the dovetail people.

Ed
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
Grant Taylor
2017-12-06 04:23:01 UTC
Reply
Permalink
Raw Message
Post by W Mainframe
Come on.....For years FTP works....
Horse and buggy works too. As does a washboard and a wood burning
kitchen stove. Yet we don't see those in /common/ use anymore.

I'm fairly certain that Buss & Tag cables worked just fine too. As did
ESCON. But technology moved on and there were newer / faster / better
(?) solutions that are now used. I.e. FICON.

Or how about old 24-bit or 31-bit addressing?

New and arguably better things come along. Time change.
Post by W Mainframe
Now... Every OS needs to upgrade for SFTP?
I don't know about Mainframe OSs, but I can say that every open system
OS that I've used does have support for SFTP, both client and server.
(and likely has had it for years.)

I've been told that Mainframe OSs (z/OS in particular) also have had
support for SFTP for years. I've been lead to believe that many
Mainframe users have been reluctant to change what they are doing
because what they are doing works and always has. - I get the desire
to not rock the boat or change things unnecessarily. - I view the
security risks of FTP (not FTPS) to be motivation to migrate away from
FTP. - I also think that SFTP has some security options, as in control
of what the logged in user can do on the remote system, that make it
more compelling than FTP(S).

I've also been told that sftp is supposed to be a drop in replacement
for ftp. At least from the open systems side, just about any script
that interacted with the ftp command line interface can also do the
exact same thing with the sftp command line interface, unmodified. -
(As I understand it.) Somewhat for this very reason.
Post by W Mainframe
Only now, this protocol is unsure???? One more stupid option... bla. bla bla
Nope. There's nothing unsure or unknown about FTP. In fact, it's how
well known it is that's part of the problem. Particularly the fact that
FTP (not FTPS) is unencrypted. Thus it is a security nightmare.

Think about it like old encryption methods for the RACF database, DES
(3DES?) verses newer methods that are far more secure.

Any (former) IBMers will likely know of the ISEC issues (CIRATS anyone?)
that FTP (not FTPS) presents.

Do you want to go into a crowded room and announce "My (Social
Security|Credit Card|mainframe password) Number is <pause so people can
get something to write with or record> one, two, three, four, five...."
I didn't think so.
--
Grant. . . .
unix || die


P.S. z14 ... with it's pervasive encryption ... why use an unencrypted
file transfer protocol?
Elardus Engelbrecht
2017-12-06 14:04:44 UTC
Reply
Permalink
Raw Message
Post by W Mainframe
Come on.....For years FTP works.... Now...
Yes, it is working good, but for security reasons, you need to use ____something else___ so unauthorized transfers can not take place and your data is not exposed somewhere.
Post by W Mainframe
Every OS needs to upgrade for SFTP?
Of course. Unless you are in a system setup where source and destination machines are locked up properly inside a [closed/fire-walled] network.

Oh, that is not good enough. Say, you transfer something from a secure / locked up system to a [unprotected] PC. Now some person inserts a USB stick into it and steal your data after that transfer.
Post by W Mainframe
Only now, this protocol is unsure????
What protocol is 'unsure'? FTP? SFTP? or FTPS?

Rather, please post what you are needing to resolve your problem.

Of course, secure transfers come with a price, time, setup and testing.

Groete / Greetings
Elardus Engelbrecht

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
John McKown
2017-12-06 14:13:20 UTC
Reply
Permalink
Raw Message
On Wed, Dec 6, 2017 at 8:06 AM, Elardus Engelbrecht <
Post by Elardus Engelbrecht
Come on.....For years FTP works.... Now...
Yes, it is working good, but for security reasons, you need to use
____something else___ so unauthorized transfers can not take place and your
data is not exposed somewhere.
Every OS needs to upgrade for SFTP?
Of course. Unless you are in a system setup where source and destination
machines are locked up properly inside a [closed/fire-walled] network.
Oh, that is not good enough. Say, you transfer something from a secure /
locked up system to a [unprotected] PC. Now some person inserts a USB stick
into it and steal your data after that transfer.
Only now, this protocol is unsure????
What protocol is 'unsure'? FTP? SFTP? or FTPS?
I think he meant "insecure" instead of "unsure". But should have said
"vulnerable".

FTP is "clear text" and so insecure. FTPS is supposed "secure" because it
uses TLS (was SSL). But TLS has vulnerabilities under some conditions (
https://www.gracefulsecurity.com/tls-ssl-vulnerabilities/). ​ SFTP is
secure because it uses the SSH encryption protocol. But that has
vulnerabilities too.
https://www.cvedetails.com/vulnerability-list/vendor_id-120/product_id-317/SSH-Ssh2.html
Post by Elardus Engelbrecht
Rather, please post what you are needing to resolve your problem.
Of course, secure transfers come with a price, time, setup and testing.
Groete / Greetings
Elardus Engelbrecht
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
--
I have a theory that it's impossible to prove anything, but I can't prove
it.

Maranatha! <><
John McKown

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
Jesse 1 Robinson
2017-12-06 19:35:57 UTC
Reply
Permalink
Raw Message
SMP(E) was a solution to the overwhelming problem of managing MVS and the intricate complexity of its (necessarily) modular construction. The bigger the component, the more SMPE has to offer. OTOH the smaller the component, the less clear advantage it offers. Historically I see two deviations from canonical SMPE.

1. ISVs who have adopted SMPE for their product(s) have sometimes adopted a practice where sysmods are 'optional'. That is, here's an enhancement; install it or not according to whether you want that function. Eliminating or routinely bypassing REQs and PREs undermines the fundamentals of SMPE. I suspect that such tactics were intended to mollify customers who wanted SMPE support without understanding the details.

2. Vendors who have ported products to z/OS from another platform, usually Unix. The poster child is Java, where most every PTF is in fact a total product replacement. I understand that most such products were never designed to be modular. If a product cannot be updated in pieces, then SMPE has questionable value other than a snapshot view its maintenance status. If the product provides its own maintenance doc, then demanding SMPE delivery may not be well motivated.

.
.
J.O.Skip Robinson
Southern California Edison Company
Electric Dragon Team Paddler
SHARE MVS Program Co-Manager
323-715-0595 Mobile
626-543-6132 Office ⇐=== NEW
***@sce.com

-----Original Message-----
From: IBM Mainframe Discussion List [mailto:IBM-***@LISTSERV.UA.EDU] On Behalf Of John McKown
Sent: Wednesday, December 06, 2017 9:46 AM
To: IBM-***@LISTSERV.UA.EDU
Subject: (External):Re: SFTP
Post by Seymour J Metz
The point to requiring use of SMP is not to satisfy a checklist, but
to provide certain capabilities, e.g., tracking changes, installing
fixes without having to do a "level set", backing off bad fixes,
suspending the application of fixes with errors or requiring customer
action. If the REXX script does everything that SMP does but in a
different way then IMHO it's acceptable.
​OK, I misunderstood. I interpreted the "must be SMPE installed" as "we won't even consider any alternate for installation and maintenance. No SMPE means we will refused to get or use a product."

As to the "shar" script - it does a full install of the product (aka "level set"). I don't know how Dovetailed Technologies does maintenance for a company with a maintenance contract. Perhaps Kirk will say.
Post by Seymour J Metz
--
Shmuel (Seymour J.) Metz
http://mason.gmu.edu/~smetz3
--
I have a theory that it's impossible to prove anything, but I can't prove it.

Maranatha! <><
John McKown


----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
Seymour J Metz
2017-12-07 20:50:02 UTC
Reply
Permalink
Raw Message
SMP precedes MVS.

Any rule enforced by people who do not understand or care about its purpose will lead to undesirable consequences. Remember "x% of the code lines must habe a comment"?


--
Shmuel (Seymour J.) Metz
http://mason.gmu.edu/~smetz3

________________________________________
From: IBM Mainframe Discussion List <IBM-***@listserv.ua.edu> on behalf of Jesse 1 Robinson <***@SCE.COM>
Sent: Wednesday, December 6, 2017 2:37 PM
To: IBM-***@listserv.ua.edu
Subject: Re: SFTP

SMP(E) was a solution to the overwhelming problem of managing MVS and the intricate complexity of its (necessarily) modular construction. The bigger the component, the more SMPE has to offer. OTOH the smaller the component, the less clear advantage it offers. Historically I see two deviations from canonical SMPE.

1. ISVs who have adopted SMPE for their product(s) have sometimes adopted a practice where sysmods are 'optional'. That is, here's an enhancement; install it or not according to whether you want that function. Eliminating or routinely bypassing REQs and PREs undermines the fundamentals of SMPE. I suspect that such tactics were intended to mollify customers who wanted SMPE support without understanding the details.

2. Vendors who have ported products to z/OS from another platform, usually Unix. The poster child is Java, where most every PTF is in fact a total product replacement. I understand that most such products were never designed to be modular. If a product cannot be updated in pieces, then SMPE has questionable value other than a snapshot view its maintenance status. If the product provides its own maintenance doc, then demanding SMPE delivery may not be well motivated.

.
.
J.O.Skip Robinson
Southern California Edison Company
Electric Dragon Team Paddler
SHARE MVS Program Co-Manager
323-715-0595 Mobile
626-543-6132 Office ⇐=== NEW
***@sce.com

-----Original Message-----
From: IBM Mainframe Discussion List [mailto:IBM-***@LISTSERV.UA.EDU] On Behalf Of John McKown
Sent: Wednesday, December 06, 2017 9:46 AM
To: IBM-***@LISTSERV.UA.EDU
Subject: (External):Re: SFTP
Post by Seymour J Metz
The point to requiring use of SMP is not to satisfy a checklist, but
to provide certain capabilities, e.g., tracking changes, installing
fixes without having to do a "level set", backing off bad fixes,
suspending the application of fixes with errors or requiring customer
action. If the REXX script does everything that SMP does but in a
different way then IMHO it's acceptable.
​OK, I misunderstood. I interpreted the "must be SMPE installed" as "we won't even consider any alternate for installation and maintenance. No SMPE means we will refused to get or use a product."

As to the "shar" script - it does a full install of the product (aka "level set"). I don't know how Dovetailed Technologies does maintenance for a company with a maintenance contract. Perhaps Kirk will say.
Post by Seymour J Metz
--
Shmuel (Seymour J.) Metz
http://mason.gmu.edu/~smetz3
--
I have a theory that it's impossible to prove anything, but I can't prove it.

Maranatha! <><
John McKown


----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
Paul Gilmartin
2017-12-06 23:07:29 UTC
Reply
Permalink
Raw Message
We always do a full-replace. For a <20MB install that is 95% Unix files,
this makes sense since we have a automated process that does builds and I
wouldn't want to introduce human errors by crafting partial replacements
each time.
Post by Seymour J Metz
The point to requiring use of SMP is not to satisfy a checklist, but to
provide certain capabilities, e.g., tracking changes, installing fixes
without having to do a "level set", ...
I'm trying to envision a service contract written to prohibit "level set"s.
This would require that each PTF:
o Have no PRErequisites
o Repair (only) the defect the customer reported.
o Introduce no additional defects.
o Introduce no other behavioral changes. In particular, it must repair
no defects other than the one the customer reported.

That would be perceived as paradise by Ed G.

--gil

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
Jesse 1 Robinson
2017-12-07 00:39:42 UTC
Reply
Permalink
Raw Message
This would seem to be a case 'use SMPE just cuz'. I don't see much real value. Except maybe for SMPE's ability to track all relevant data sets via DDDEF. You still have to find the CSI on your own, but once there it's pretty easy to locate the target files.

Unless customers are beating down your door, I wouldn't invest the resources in SMPE packaging.

.
.
J.O.Skip Robinson
Southern California Edison Company
Electric Dragon Team Paddler
SHARE MVS Program Co-Manager
323-715-0595 Mobile
626-543-6132 Office ⇐=== NEW
***@sce.com


-----Original Message-----
From: IBM Mainframe Discussion List [mailto:IBM-***@LISTSERV.UA.EDU] On Behalf Of Kirk Wolf
Sent: Wednesday, December 06, 2017 12:46 PM
To: IBM-***@LISTSERV.UA.EDU
Subject: (External):Re: SFTP

We always do a full-replace. For a <20MB install that is 95% Unix files, this makes sense since we have a automated process that does builds and I wouldn't want to introduce human errors by crafting partial replacements each time.
I suppose that we could wrap the whole thing in SMPE like Java if there were any real demand, but we have other requirements that I am not sure how to address with SMPE.


Kirk Wolf
Dovetailed Technologies
http://dovetail.com
Post by John McKown
Post by Seymour J Metz
The point to requiring use of SMP is not to satisfy a checklist, but
to provide certain capabilities, e.g., tracking changes, installing
fixes without having to do a "level set", backing off bad fixes,
suspending the application of fixes with errors or requiring
customer action. If the
REXX
Post by Seymour J Metz
script does everything that SMP does but in a different way then
IMHO
it's
Post by Seymour J Metz
acceptable.
​OK, I misunderstood. I interpreted the "must be SMPE installed" as
"we won't even consider any alternate for installation and
maintenance. No SMPE means we will refused to get or use a product."
As to the "shar" script - it does a full install of the product (aka
"level set"). I don't know how Dovetailed Technologies does
maintenance for a company with a maintenance contract. Perhaps Kirk will say.
Post by Seymour J Metz
--
Shmuel (Seymour J.) Metz
http://mason.gmu.edu/~smetz3
--
I have a theory that it's impossible to prove anything, but I can't
prove it.
Maranatha! <><
John McKown
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions, send
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions, send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN


----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
Edward Gould
2017-12-07 11:02:50 UTC
Reply
Permalink
Raw Message
Post by Jesse 1 Robinson
This would seem to be a case 'use SMPE just cuz'. I don't see much real value. Except maybe for SMPE's ability to track all relevant data sets via DDDEF. You still have to find the CSI on your own, but once there it's pretty easy to locate the target files.
Unless customers are beating down your door, I wouldn't invest the resources in SMPE packaging.
.
SKIP,

The place I help out at got burned with easy types of installs. They had to back out a system because one product was overlooked and it took too much time to fix the issue. After that the SMPE rule came down.
Heck at a former place they had to deinstall a couple of CPU’s because a product used the CPU serial # (it was given to us incorrectly).
So a simple issue balloons into major issues.

Ed
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
Loading...