Re: Version of BPCS client for 6.1 -- BPCS-L

From: Driss Hajjaji <DrissH@Eicon.com>
To: bpcs-l@midrange.com <bpcs-l@midrange.com>
Date: Tuesday, January 25, 2000 5:25 PM
Subject: Version of BPCS client for 6.1

>Hi,
>
>We are currently testing eBPCS 6.1 full client/server with the intent of
>going live in less than 8 weeks. We are doing a simple upgrade (no new
>functionality) from BPCS 6.0.02. However, we do have some problems getting
>the right version of the client and SSA helpline seems to have problem
>telling us (and providing us) with the latest version of the client (yet
>they seem to imply we do not have the latest version and this is the source
>of our problems).

Hello,

It appears the problem was first reported in early December, and the
Helpline could not duplicate the problem using instructions provided by the
caller and could find no matching prior calls on the problem as described.
Because you had been shipped the latest ORD BMRs for another call, and the
consultant requested you apply the BMRs and then re-test the issue, which at
the time was reported as being with ORD700. You were to call back if the
problem continued. Eicon only re-opened the call log 1/26/00. As it turned
out, this time the product involved was not ORD, but was instead shown to be
a panel in MRP. When Helpline today received a screen print from Eicon
showing the exact panel in MRP (this detail was not provided previously), it
was researched and resolved the next day.

Total time the SSA Helpline has dealt with this reported problem was just
over 2 days. You have a workaround (via the CUI), and a BMR will be done on
the VLF file in question to permanently repair it. I would say that is good
service, actually.

I suggest if you feel you are having problems dealing with the Helpline, or
the level of Helpline support you are receiving, you request to speak to a
Helpline manager so that they can look into the matter and resolve it.

And to clarify more stuff regarding this thread............

The version.cfg file is NOT the version of BPCS, but is the version of the
OPS Runtime software. The same version of presentation logic runs on both
6100 and 6004, and has nothing to do with BPCS application logic, which
according to the call log is what your problem is about .That is why it is
buried several directories down, and is not off the main BPCS Client Server
directory. To see the shipped software version of the PC Client software in
general, the main release shows up in the 'bpcs.ini' file, which is just off
the main /bpcscs directory.

Someone else posted to the list that they were Mixed mode and had this
directory and file as well and all was working - if so, the software is
never being invoked, because the OPS Runtime is only used with 2nd
Generation (Full Client Server) BPCS screens. So no matter what your
version.cfg file says in Mixed mode, upgrading it or changing it will make
no difference to the PC Client's behavior.

I guess I also have to take some exception to Dean's posting which basically
yells at software companies' support structures for asking about what
version of our software you have, despite the fact that he mentioned that I
am doing a good job :-) .

>Probably a combination of low unemployment versus an increased call level
for
>both, but that response should _NOT_ be what they try to pawn off on us for
>every single question we ask.
>The question from SSA is particularly onerous
>regarding client questions, as they ask you if you're on the latest version
>in one breath, yet cannot support IE5 on OGS in the next -- the last time
_I_
>checked, 5 _was_ the latest version of Internet Explorer and has been out
>longer than some of the latest client versions of BPCS.

It is a standard question to ask when trying to resolve a problem, and you
would not believe the number of times it DOES apply to a problem (speaking
as an ex-Helpline employee), so I hardly think it is a bad or lazy question
to be asking a user. In fact it is the _FIRST_ question which should
_ALWAYS_ be asked, so that proper testing can occur or so that proper
searching of call archives can be done to research the error.

In addition, we (SSA R&D) won't fix the program at your BMR level (if you
are not current), we fix the current BMR level (on that version of the
software). So logically, the first step is to test the reported problem at
current levels of software to see if it fails. If it works OK, and you test
on one cume back and it is broken as described, then it is natural to
imagine that a BMR in between must fix it. So, you may be requested to get
current on BMRs for the suspected program which is erroring. To me this is
just plain logic. If the problem can not be duplicated in current code, then
getting the customer to current code is a valid first step and resolves most
calls more quickly. If that doesn't fix it, then more research is required.
Sometimes the lack of duplication of complex issues in ERP software can have
to do with very exacting set up that can vary at hundreds of levels - and
yes, people (both clients and Helpliners) may not think of every possible
thing that might affect the program at first blush, and that is the cause of
failure to duplicate the issue - customer doesn't mention everything they
should or Helpline doesn't ask every question they could. I don't think it
is by any concerted effort of the giant software conglomerates that this
happens - it happens because every time you add a new function to a product
or make it more configurable - you are adding layers of complexity, and it
is harder to keep all that in your limited brain RAM at any one moment.

In addition, if you are reporting a error such as an RPG or CPF message and
it shows at a particular line of code, we can't match that to any code on
our system to see which line is failing unless you are at the same code
level (we can not possibly keep a copy of code for every BMR ever done on
every program in BPCS). So version and levels of code may be key to
resolving your issue if an error message is involved, whether at IBM or SSA.

As to the supposed problems cause by the combination of OGS and IE5...more
to rant on there. . . .

First of all, OGS (a web site) works just fine and dandy when viewed with
IE5. However the FTP site (which is a link off the OGS site) where BMR fixes
can be retrieved from, and which is a Unix server, is where the problem
starts. Netscape 4.7 will also cause it.

The problem occurs because your password is not carried over by the Web
Browser when you change directories at each click onto a new directory
level. If you type the entire path in your browser to begin with, you will
get there OK. Use this as a workaround (rather than clicking on each
directory from the beginning). I am told that there is NOTHING we (SSA) can
do about this, as it has nothing to do with our software on the site or our
firewall (this has been tested).

They believe it may have to do with the 'security enhancements' made on
those new browser versions - but it is the browser which is dropping your
login name and password. I was told by the people who manage the FTP and
other internal systems, that they have seen this happen if you go to OTHER
Unix FTP servers, not just the one at SSA. If you get one of those free FTP
clients available on the internet and you will have no problems at all using
the FTP site. If anyone thinks they know a way for SSA to fix this, let me
know off line by directly replying to my e-mail address. We are open to
suggestions, but at the moment from our research, we do not believe the
problem is in any software that we are using on our FTP site server.

Thanks

Genyphyr Novak
SSA



+---
| This is the BPCS Users Mailing List!
| To submit a new message, send your mail to BPCS-L@midrange.com.
| To subscribe to this list send email to BPCS-L-SUB@midrange.com.
| To unsubscribe from this list send email to BPCS-L-UNSUB@midrange.com.
| Questions should be directed to the list owner: dasmussen@aol.com
+---