Microsoft word 2010, errors on startup

I got a virus on my laptop, ever since then I’ve been getting messages like the below when I start word:

“The add-in template is not valid”

This was happening 3 times every time I was going into word – super frustrating.  I was going into word, I could not delete the add-ins.  I was searching the registry – nothing.

No delete..

 

Then I found (after searching the C:\) this folder location.  Word tries to load all of the templates in this dir (great place for a virus to start?) when word starts.  I deleted the dodgy files from this dir, and now work starts fine! woot.

C:\Users\username\AppData\Roaming\Microsoft\Word\STARTUP

A techie post… finally I hear you say…

I agree, there have not been enough techie posts of late.  I’ve been on my soap box, talking about things that don’t really solve peoples oracle problems.  That is going to change (for this post anyway).

I’ve been using OATS (Oracle Application Testing Suite) quite a bit lately.  I also got my hands on a loan fusionIO card.  I had lots of fun with all of the technology, OVM, OEL, FusionIO, templates, JDE 9.1 etc etc.  But…  My white paper did not contain any real salient results.  Sure there was a lot of learning and we upgraded every piece of software around our OVM software.  We could not get the disks humming on the guests – damn.

Since then I’ve had some advice from an oracle guy, who will remain nameless in this forum:

• img files will be much, much slower than native access. Present the entire block device to the guest (bit hard when OVM can’t see the FusionIO disks)…
• LVM is also completely unsupported on OVM3 and will introduce massive bottlenecks, as it doesn't support block barriers. You should've just formatted the entire device with OCFS2 in local, vmstore mode if you want to use .img files. However, I would recommend again just presenting the entire device to the guest.  (Hmm he got me here)
• try HVM with PV Drivers and PCI passthrough to actually present the entire fusionIO card natively to a guest

These are all good things to try.  I hate chucking away 120 hours of hard work and not get a result, but that is the way things go sometimes.

Anyway, I seem to be drifting again.

My technical point was going to be OATS and tools release 9.1.  I kept getting errors with my 9.1 load testing using OATS.

Errors are listed in the above shot, “Failed to solve variable web.framesrc.RunAppHiddenIframe using path”  blah blah.  SOOOO frustrating!  The script would record fine in openscript, but then would not playback.

I got in contact with my VERY friendly and super helpful oracle pals who put me in contact with the right people the first time (you know who you are).  The very simple solution is (verbatim): 

“Going forward, prior to recording in your OpenScript screen please navigate to View >OpenScript Preferences > Record > HTTP and choose Record mode as ‘HTTP’ and click apply and then start recording.”

This will apply to all recording going forward, but it works…  Scripts record first time every time in JD Edwards 9.1.  Playback works and load testing works!

I also hope that I’m not bursting anyone’s bubble here, but I also got a scoop about JDE and OATS support in the future… “OATS 12.1 which is the next release has a separate protocol for “Oracle JD Edwards EnterpriseOne” load testing and has a custom made correlation library for JDE that we developed as part of the tool. This will help you to handle JDE specific correlation issues.”  That is very cool!

JD Edwards load testing as a service. Load testing as a service (LTAAS)

We’ve been able to license OATS (Oracle Application Testing Suite) from Oracle to enable us to be able to provide OATS based software (and services) as a service…  Amazing I hear you say – yes!

Proper load testing for JD Edwards has always been a very expensive and time consuming exercise which some clients just put into the too hard basket, but that no longer is the case.  Our offering can allow you access to our load testing licences for the volume that you need.  We can log into your site, get the OATS framework up and running (all done remotely).  For JD Edwards we can copy over our base scripts and testing routines (generic functionality, noise, generic search, menu navigation) and augment this with your specific requirements.

We recommend “day in the life” scenarios, where we replicate your major processes in OATS and then reproduce that load in the volumes that you require.

Note that this offering is not just limited to JD Edwards.  If you are after some OATS licenses for E-bus or any web based application – we can provide you our installation and configuration capabilities to execute your load testing requirements.

This is an inexpensive way to properly load test your JD Edwards upgrade, tools release or ESU before you put it into production.  We’ll keep all of your scripts on file, in case you want to retest at any time.

This service is also not just limited to Load and Stress testing, scripts can be written to unit test and provide you with automated methods of unit testing between tools releases, ESU’s and upgrades.

If you’d like to learn more about the service, please do not hesitate to get in contact with the team at www.myriad-it.com and see if we can assist.

JD Edwards application release 9.1 is available!

You might want to wait for the official collaborate notification, or just log into https://edelivery.oracle.com and download it!  We are working on a demo site at the moment, so watch this space…

Which of today’s UBEs ran slower than usual?

Wouldn’t it be nice to have a list of jobs that ran today, but ran slower / faster than they do on average.   Perhaps know how much slower and how many there were.  Over time, this should give you some good statistical information on how JDE performance is tracking.

The invention of F986114 and F986114A are integral to being able to get this information out of the system.  A fairly basic oracle query over these tables can give you information on averages, increases and or decreases.

 

SELECT JCPID, JCVERS, JCJOBNBR, INNERCOUNT, INNERAVERAGE, INNERMIN, INNERMAX, (86400*(JCETDTIM-JCSTDTIM)) AS RUNTIME, TRUNC((86400*(JCETDTIM-JCSTDTIM))/INNERAVERAGE*100,2) AS "PERCENT INCREASE"
FROM SVM812.F986114,
(SELECT JCPID as INNERPID, JCVERS as INNERVERS, count(1) as INNERCOUNT, Avg(86400*(JCETDTIM-JCSTDTIM)) as INNERAVERAGE, min(86400*(JCETDTIM-JCSTDTIM)) AS INNERMIN, max(86400*(JCETDTIM-JCSTDTIM)) AS INNERMAX
from svm812.f986114
group by jcpid, JCVERS)
WHERE TO_CHAR(JCETDTIM, 'DDMMYYYY') = TO_CHAR(sysdate, 'DDMMYYYY')
AND JCPID = INNERPID and JCVERS=INNERVERS
order by 9 desc
;

JCVERS    JCJOBNBR    INNERCOUNT    INNERAVERAGE    INNERMIN    INNERMAX    RUNTIME   %INCREASE R55093021P    AALL001       351200    488    2.80737704918033    0    22    14    498.68
R55093021P    AALL001       351122    488    2.80737704918033    0    22    13    463.06
R5604572OZ    VHWT001       351154    3794    3.57564575645756    2    51    11    307.63
R55093021P    AALL001       351278    488    2.80737704918033    0    22    7    249.34
R5503B413C    AALL002       351008    1243    16.8833467417538    1    191    39    230.99
R551010Z      AALL002S      351260    5053    5.20878685929151    2    18    11    211.18
R12855        AALL001       351217    60    107.466666666667    10    315    223    207.5

The first two lines show that R55093021P has run 498% longer than average, although when we look at this – it’s taken 14 seconds, when it generally averages 2.8 for the 488 times it’s run! Wow that is very cool statistical information.

But are we faster or slower, or just an average day?

SELECT count(1) as slower
FROM SVM812.F986114,
(SELECT JCPID as INNERPID, JCVERS as INNERVERS, count(1) as INNERCOUNT, Avg(86400*(JCETDTIM-JCSTDTIM)) as INNERAVERAGE, min(86400*(JCETDTIM-JCSTDTIM)) AS INNERMIN, max(86400*(JCETDTIM-JCSTDTIM)) AS INNERMAX
from svm812.f986114
group by jcpid, JCVERS)
WHERE TO_CHAR(JCETDTIM, 'DDMMYYYY') = TO_CHAR(sysdate, 'DDMMYYYY')
AND JCPID = INNERPID and JCVERS=INNERVERS
AND TRUNC((86400*(JCETDTIM-JCSTDTIM))/INNERAVERAGE*100,2) > 100
;
  SELECT count(1) as faster
FROM SVM812.F986114,
(SELECT JCPID as INNERPID, JCVERS as INNERVERS, count(1) as INNERCOUNT, Avg(86400*(JCETDTIM-JCSTDTIM)) as INNERAVERAGE, min(86400*(JCETDTIM-JCSTDTIM)) AS INNERMIN, max(86400*(JCETDTIM-JCSTDTIM)) AS INNERMAX
from svm812.f986114
group by jcpid, JCVERS)
WHERE TO_CHAR(JCETDTIM, 'DDMMYYYY') = TO_CHAR(sysdate, 'DDMMYYYY')
AND JCPID = INNERPID and JCVERS=INNERVERS
AND TRUNC((86400*(JCETDTIM-JCSTDTIM))/INNERAVERAGE*100,2) <= 100

The combination of the above two queries (I’m sure there is a better way) shows how many jobs are slower than average (69 in my example) and how many are faster (140).  So on average, today is a faster day!

Application release delta, tools release delta, what functionality was released in tools release XXX

Ever wanted to know when the apparel module was introduced?  Wanted to know what release specifically iPads were released?  What about when Design Time and Runtime Support for Embedded Oracle Business Intelligence (OBI) Reports was released?  Well – have I got the tool for you [said the Bishop to the Nun]. 

Goto this URL (be warned you need a oracle support login).

https://apex.oracle.com/pls/apex/f?p=24153:100:0

Agree to the disclaimer and POW – you have the power.

You can compare tools releases and application releases.  You can determine what functional and technical enhancements were put into what release.  This is really terrific information.  Note that you can also create reports in WORD or XML (me thinks that this might have been created with BI Publisher [maybe??]).

This is great when you want to tell a client the DELTA in technology and functionality between their release and the current release.  This is going to make you look like a rock star next time you are bought to a client and you need to sell them the next release.

100 day upgrade, JDE accelerated upgrade

Oracle have started delivering some outstanding content to give clients the tools and framework to allow them to upgrade to the latest edition of JD Edwards in less than 100 working days.

Myriad was able to sponsor and co-present the first 100 day upgrade workshop to be held in Australia.  What is the 100 day upgrade I hear you ask.  The 100 day upgrade is a methodology and framework for simplifying the upgrade and breaking the upgrade down so that it can be achieved in 100 working days.

The upgrade is broken down into 5 main challenges:

Data

Code

Testing

Training

Go live

These challenges are found within the greater program:

Each of these challenges is further broken down into various elements and how partner solutions can facilitate the process of rectifying these challenges.  The workshop dealt with a fictitious customer that was upgrading from XE to 9.02 and had it's fair share of upgrade challenges (as all clients do).

The workshop was able to answer many complicated questions regarding upgrades.  The workshop was instrumental in showing how oracle products like OATS and UPK can help you get the best out of your upgrade, help you simplify and automate your tasks and help you succeed.

The Myriad Technology Upgrade is based upon the 100 day upgrade framework.  We have a "cookie cut" project plan which we are able to tailor for each of our clients so that might be able to determine how much and how long the project is going to take.

An essential element of the 100 day upgrade is identifying processes within JDE.  Documenting these processes or flows and then making sure that testing, training, archiving etc make reference to these processes and use these processes. 

Careful identification of the processes will also allow you a governance framework, in that you can say NO to process change if it's different from the original process flow.  This governance is going to ensure that the project is delivered on time and on budget.

Myriad has specific IP, tools and tricks that can help you in ALL 5 areas identified above.  We can quickly reconcile your data, we can quickly document your processes, we can quickly assist in identifying modifications.  We can create a project plan based upon this best practice 100 day upgrade framework and help you achieve all of the benefits from being on the latest and greatest software platform.

Please familiarise yourself with the outline content available at:

www.upgradejde.com

If you want to know more about the process or the tips and tricks that Myriad employ within this framework, please get in contact with Shannon from Myriad.