OneWorld ASPs - WTS

CNCPawn

Member
Our organization has been looking at moving our OneWorld installation to an ASP, namely WTS (www.wts.com). If we were to make that choice we would be their largest client. We use all of the Xe modules except manufacturing. We have about 150 users with plans to grow to about 800 in the next couple of years. All users except developers connect throught Citrix.

One of WTS's big marketing claims to our management is that they haven't had anyone involuntarily disconnected from OneWorld in 3 years. That issue, which seems to be caused by a “memory violation”, has plagued us for almost 2 years. After hearing that claim, I tried to do some more research on the issue and found that, according to the Quest user groups and this forum, it is a fairly common issue, especially with the P4210 & P4310.

If there are any WTS customers out there, I would really like to hear your comments on this.

Has WTS found a resolution that they (JDE included) could be sharing with the rest of us?

FYI – JDE owns a fairly large portion of WTS and WTS’s home office is in one of the JDE buildings in Denver.

Configuration: AS/400 Enterprise Server, W2K Logic Server, W2K Deployment Server, 6 Citrix Metaframe XPe Servers
 
I think one of the most important issues on remote location access to
OneWorld is the SLA of your network service provider. In my opinion, keepin=
g
up OneWorld is fairly easy. It has become very stable with Xe (working with
it since B733.0), the one gotcha for me has always and continues to be, the
connection to the remote offices..A ASP would not help in that instance
since you are still at the mercy of the network provider.
One question on the ASP..how do they actually host it. Are they just
running a bunch of terminal servers as well? If so, you are not going to
have any higher reliability than if your company did it themselves. The use=
r
override issues will still occur, etc. If the ASP is actually running the
servers and you access them with FAT clients, the pipe you would have to be
very large and quite expensive..In my opinion, I don't really see ASPs
proving that much value (at least in the JDE application)...
 
The obvious answer here is yes, as your configuration isn't all that
unusual (I'm currently at a client site with AS/400 and Citrix, and we
don't get any memory violations. I don't know how often you get these
memory violations, but as a JD Edwards customer, you should have the
ability to say "fix it". There are numerous clients using your same
architecture that aren't "plagued" as you say.

My first step would be to get the MTR's (hardware and software) for your
specific release of OW, SP level, Update Level from GSS (open a call with
GSS and get them to get you the MTR's). Ensure that all of your
supporting software meets the MTR, BUT, important but here, does not
exceed the MTR to the area of being unsupported. Of crucial importance
in your situation...Client Access, PTF's on the AS/400, Citrix release
levels, OS service pack levels on the Citrix boxes. My point here is
that I've walked into too many client sites where the latest and greatest
SP or release is applied because the thought process is "I meet the
MTR...heck, I superiorly exceed the MTR"...not good in the case of OW.

If everything is kosher with the software levels and setup, next check
and verify those BSFN mappings to the logic or ES. I can't tell you how
many times a client was having problems with Citrix usage and it was
because 1 or 2 BSFN mappings were not made.

Check OneWorld itself. Are you having User Override issues? Is the user
doing something not supported (yes, there have been times when I've found
that a user is not following the correct process, such as hitting ENTER
or clicking on cancel at the wrong time, thus causing memory errors or
application failure), what shows up in the logs (both OW and OS), and do
your fat clients have the same issues?

I'm sure you've been through all of this, but again, I've walked into
clients locations who swore up and down that MTR's were met or OCM's were
verified and it was not the case. Once you've done all of your leg
work...then tell JD Edwards you want a solution. Other clients are using
your same platform without being plagued by memory violations...so it CAN
work. You've just not stumbled over the solution(s) yet.

Of course, that depends on the definition of "plagued"...I've been called
to a site where the problem was described as "we are having package build
errors"...and the "errors" were a single warning during the build
process...not an error mind you...all objects built successfully and the
package could be deployed with no client issues...but their definition of
'error' was that single warning. :)

Jim
On Thu, 27 Feb 2003 07:41:01 -0800 (PST) CNCPawn <[email protected]>
writes:

One of WTS's big marketing claims to our

If there are any WTS customers out there, I would really

Has WTS found a resolution that

FYI – JDE owns a fairly large portion of WTS and WTS’s home office

Configuration: AS/400

________________________________________________________________
Sign Up for Juno Platinum Internet Access Today
Only $9.95 per month!
Visit www.juno.com
 
Back
Top