Witango-Talk: Consultant posting

[MJPinckard]

We're looking for an expert in Witango server to help us
troubleshoot and fix stability problems with our interactive web site.
Must have experience in the following areas:
• Witango server configuration and admin
• Witango programming
• Mac OS X Server configuration and admin
• Witango/Oracle database connectivity

Needs to be a team player who can work with the
existing development team and fully document
all work products. This project will involve up to
80 hours of effort during January-February 2006.
Work site is located in Berkeley, CA.

Contact Rich Brown, [EMAIL PROTECTED], (510) 586-5896.


TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

Re: Witango-Talk: @sort not working

[MJPinckard]

I'd like to confirm that no... attempting to sort user scope using this 
code doesn't work for me either. But like everyone else... sorting the 
request scope does work, so I have a workaround for now.


Maggie

Robert Garcia wrote:


No.

Results:


User Unsorted: border="1">343331323467/>


User Sorted: border="1">343331323467/>






Request Unsorted: border="1">673433313234/>


Request Sorted: border="1">132333467343/>




Request: 
User: border="1">132333467343



--

Robert Garcia
President - BigHead Technology
VP Application Development - eventpix.com
13653 West Park Dr
Magalia, Ca 95954
ph: 530.645.4040 x222 fax: 530.645.4040
[EMAIL PROTECTED]  - 
[EMAIL PROTECTED] 

http://bighead.net/ - http://eventpix.com/

On Nov 16, 2005, at 6:40 PM, Customer Support wrote:


<@purge scope="user">
<@define user$test type="array" cols="1" rows="0">
<@assign name="test" scope="user" value='<@ARRAY 
VALUE="343;33;1;32;34;67;">'>

User Unsorted: @@user$test
<@sort scope="user" ARRAY="test" COLS="1 ASC">
User Sorted: @@user$test


<@PURGE scope="user">
<@define request$test type="array" cols="1" rows="0">
<@assign request$test '<@ARRAY VALUE="67;343;33;1;32;34;">'>
Request Unsorted: <@var request$test>
<@sort array=request$test cols="1 num asc">
Request Sorted: <@var request$test>
<@assign user$test "<@var request$test>">
<@define request$test type="array" cols="0" rows="0">

Request: <@var request$test>
User: <@var user$test>




TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf
 



TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

Re: Witango-Talk: @sort not working

[MJPinckard]

darn! must be OS X specific... definitely doesn't work on server version 
10.4.3 with Witango 5.5.009


even tried restarting the server just in case that might help.

sigh...

thanks anyway.

Maggie

Robert Garcia wrote:

I just threw it on a test taf. Even this works on my server, tried 
several variations. It worked. Windows 2003, 5.5.009


   <@assign name="test" value='<@ARRAY VALUE="1;32;34;67;343;33;2">' 
scope="user">


   @@user$test

   <@SORT ARRAY="user$test">

   @@user$test


--

Robert Garcia
President - BigHead Technology
VP Application Development - eventpix.com
13653 West Park Dr
Magalia, Ca 95954
ph: 530.645.4040 x222 fax: 530.645.4040
[EMAIL PROTECTED] <mailto:[EMAIL PROTECTED]> - 
[EMAIL PROTECTED] <mailto:[EMAIL PROTECTED]>

http://bighead.net/ - http://eventpix.com/

On Nov 16, 2005, at 5:02 PM, MJPinckard wrote:

   <@assign name="test" value='<@ARRAY VALUE="1;32;34;67;343;33">' 
scope="user">



   @@user$test


   <@SORT ARRAY="user$test" COLS="1 ASC">


   @@user$test




TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf
 



TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

Re: Witango-Talk: @sort not working

[MJPinckard]

Nah... that doesn't work either... I'd actually tried that in the 
original file giving me problems before I created the test file... in 
any event, I just scoped them in this test file and reloaded to the 
servers with the same result.


so new code (functional on 5.0.1.065, non-functional on 5.5.009) reads:
   <@assign name="test" value='<@ARRAY VALUE="1;32;34;67;343;33">' 
scope="user">


   @@user$test

   <@SORT ARRAY="user$test" COLS="1 ASC">

   @@user$test

still stuck.

Maggie


Robert Garcia wrote:


Scope your variables?

<@sort array=user$test cols="1 ASC">

--
Robert Garcia
President - BigHead Technology
VP Application Development - eventpix.com
13653 West Park Dr
Magalia, Ca 95954
ph: 530.645.4040 x222 fax: 530.645.4040
[EMAIL PROTECTED] - [EMAIL PROTECTED]
http://bighead.net/ - http://eventpix.com/

On Nov 16, 2005, at 4:46 PM, MJPinckard wrote:


I'm running Witango on OS X 10.4

On my dev server (running Witango 5.5.009) the @Sort tag doesn't  
work... no sorting of any kind takes place

see http://hes1.lbl.gov/hes/testsort.taf

on my production server (running Witango 5.0.1.065) it works.
see http://hes3.lbl.gov/hes/testsort.taf

first table is array as initialized, second is array after a sort  
action... same code for both servers. Here's the code for the  entire 
taf:
   <@assign name="test" value='<@ARRAY VALUE="1;32;34;67;343;33">'  
scope="user">


   @@user$test

   <@SORT ARRAY="test" COLS="1 ASC">

   @@user$test

It seems like it's clearly a bug in the 5.5 server.

I'm in the middle of development and of course under deadline  
pressure are there any easy fixes... I'd rather not be running  
development code on the production server... and of course rolling  
the dev server back to the older server version is a hassle, and  not 
really in my current schedule.


any comments/suggestions on some kind of quick fix?

cheers,

Maggie Pinckard
Principal Research Associate
Lawrence Berkeley National Laboratory
__ 
__

TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf




TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf



TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

Witango-Talk: @sort not working

[MJPinckard]

I'm running Witango on OS X 10.4

On my dev server (running Witango 5.5.009) the @Sort tag doesn't work... 
no sorting of any kind takes place

see http://hes1.lbl.gov/hes/testsort.taf

on my production server (running Witango 5.0.1.065) it works.
see http://hes3.lbl.gov/hes/testsort.taf

first table is array as initialized, second is array after a sort 
action... same code for both servers. Here's the code for the entire taf:
   <@assign name="test" value='<@ARRAY VALUE="1;32;34;67;343;33">' 
scope="user">


   @@user$test

   <@SORT ARRAY="test" COLS="1 ASC">

   @@user$test

It seems like it's clearly a bug in the 5.5 server.

I'm in the middle of development and of course under deadline 
pressure are there any easy fixes... I'd rather not be running 
development code on the production server... and of course rolling the 
dev server back to the older server version is a hassle, and not really 
in my current schedule.


any comments/suggestions on some kind of quick fix?

cheers,

Maggie Pinckard
Principal Research Associate
Lawrence Berkeley National Laboratory

TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

Re: Witango-Talk: OS X.4

[MJPinckard]

Well I have WAY more than 48 instances... my search returned 1420 
instances in 46 files... and as I said earlier, my servers have been 
bouncing like a yo-yo recently... fatal exceptions, generally associated 
with a DB call... so I'm really suspicious that this could be involved...


again... any real problem with just deleting the data dictionaries 
(actually I'd delete the column definitions within the data 
dictionaries, not the entire data dictionary) wholesale to provide a 
quick fix?


Maggie

Robert Garcia wrote:

Like I said, this is a major bug, and it has frustrated me very much  
that witango has not responded to me, or others about it. I verifed  
with Weiss, he sent in emails, and Roland also, but witango may have  
responded very recently, but not to me, and they hadn't to the others  
for some time.


Everyone using mac dev, should search for "!CST" in there tafs, and  
post the results. IMHO, this is a serious bug, it negates the  
usefulness of the studio, if you have to constantly open tafs in text  
editor to check for corruption.


I guess, if you don't pay specifically support, even if you pay for a  
license, support is random. I have asked this question, and have not  
had a response from witango. I have SEVERAL witango licenses, do what  
I can to support the community, provide free tools and assistance to  
the community, and have done alot of beta testing for witango,  
especially preparing for v5. What does one have to do to get a response?




TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

Re: Witango-Talk: OS X.4

[MJPinckard]

egads!!!

I just searched for "!CST" and it is all over the place in my tafs (and 
my server has been VERY unstable recently so this may be contributing to 
the delinquency of a server)... question, what should I do? Okay to 
delete all the Data Dictionary sections with this datatype, or do I 
need/have to attempt to correct each one?


My guess is that I can delete the offending sections/columns of the data 
dictionaries... since I had a problem with my editor for a while and was 
building update actions by hand in a text editor (wy too much 
fun...) and generally dispensed with this section in the hand built actions.


cheers,

Maggie

Robert Garcia wrote:


sorry, that is, search for "!CST".

--
Robert Garcia
President - BigHead Technology
VP Application Development - eventpix.com
13653 West Park Dr
Magalia, Ca 95954
ph: 530.645.4040 x222 fax: 530.645.4040
[EMAIL PROTECTED] - [EMAIL PROTECTED]
http://bighead.net/ - http://eventpix.com/

On Oct 19, 2005, at 4:23 PM, Robert Garcia wrote:


search your tafs fors !CST with a text editor.

see if you have any. I had some, with no error, and didn't realize  
it, then I started seeing the errors later.


--

Robert Garcia
President - BigHead Technology
VP Application Development - eventpix.com
13653 West Park Dr
Magalia, Ca 95954
ph: 530.645.4040 x222 fax: 530.645.4040
[EMAIL PROTECTED] - [EMAIL PROTECTED]
http://bighead.net/ - http://eventpix.com/

On Oct 19, 2005, at 3:22 PM, Bill Conlon wrote:


Thx much.  I found a used dual G4 1.25 GHz for the price of a  mini, 
so I'll probably stick with X.3.9.


BTW, I've never had the corruption issue while running X.2.8.

bill
On Wednesday, October 19, 2005, at 03:00  PM, Robert Garcia wrote:




I do, and there are others, like Mark Weiss and Roland.

There is a serious bug that I don't think is JUST 10.4, but seems  
to show up more in 10.4. Your insert and update actions can get  
corrupted. It is a major pain, it has been reported to Witango  
months ago. It has not been acknowledged, and none of (AFAIK)  have 
heard from witango on this.


See previous posts for more details on the bug.

--
Robert Garcia
President - BigHead Technology
VP Application Development - eventpix.com
13653 West Park Dr
Magalia, Ca 95954
ph: 530.645.4040 x222 fax: 530.645.4040
[EMAIL PROTECTED] - [EMAIL PROTECTED]
http://bighead.net/ - http://eventpix.com/

On Oct 19, 2005, at 12:32 PM, Bill Conlon wrote:





I think I'm ready to migrate from my circa 1998 G3 desktop.  Can't 
remember which feline is the current version, but does  anyone 
have experience running the Dev Studio on X.4?


thx

bill

___ 
_

TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf






 


TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf





_ 
___

TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf





__ 
__

TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf





TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf



TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

Witango-Talk: Witango in Canada?

[MJPinckard]

I'm considering relocating to the Vancouver BC Canada area. Anyone there 
using Witango?


Cheers,

Maggie Pinckard

TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

Re: Witango-Talk: duplicating a row in Oracle

[MJPinckard]

Well I can't seem to extract anything from @@resultset[0,1] - it's empty

I'd be delighted to use a stored procedure, but at this point, it is 
beyond my Oracle abilities to create... anyone have a good primer on 
making stored procedures (or code I can utilize)?


cheers,

Maggie

Peter Dobbs wrote:


You receive the column names in position 0 of the resultset.

i.e. @@local$resultset[0,1] is the column name of column 1.

Also, if you want to do something like this, make a stored procedure to do
it and call the stored procedure from a directDBMS action. No point of
bringing the info into Tango unless you are not comfortable with stored
procedures in which case doing it in tango will work just fine!


p.

-Original Message-
From: MJPinckard [mailto:[EMAIL PROTECTED] 
Sent: Wednesday, July 27, 2005 7:46 PM

To: Witango
Subject: Witango-Talk: duplicating a row in Oracle

I'm trying to add an option for users to clone their user session.

I have a table with ~500 columns (yeah I know, but that's how it's 
structured, legacy code...ugh).


I want to duplicate this row, changing only the primary key 
(session_id). Since it is not inconceivable that the column structure 
might change someday, I'd like to make the code as generic as possible 
(not coding each column into an update or insert action).


Using directDBMS, I can easily extract all the existing data using the 
following code.


   select * from Input where session_id=someNumber

But this gives me data only (no column names) and in any event I can't 
figure out how to get the data into another row. I've tried a variety of 
different ways (using both INSERT and creating a row in another action 
then performing an UPDATE) but can't get anything to work.


Has anyone any suggestions about how I might accomplish this? I'm about 
to give in and manually code all 500 columns into an insert action... 
but surely there has got to be an easier way...


Mac OS X server
Witango 5.0.9 (I think)
Oracle db using Oracle OCI drivers.

Thanks for any assistance.

Cheers,

Maggie Pinckard
Principal Research Associate
Lawrence Berkeley National Laboratory

TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf


TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf
 



TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

Re: Witango-Talk: duplicating a row in Oracle

[MJPinckard]

alas... row zero seems to be empty... I repeated my search then printed 
out @@resultset[0,1], @@resultset[1,1]


it returned
, 26300

plus if I display @@resultset I get a one row array (data only).

Anyone know of a way to specifically request column names... I was 
thinking maybe trying to parse a DESCRIBE command, but haven't managed 
to get that working either.


Arrrggghhh.

Maggie

Alan Wolfe wrote:


Heya,

I do believe you're in luck.  In row 0 of an array that is the results
of an sql query, it has the column names.

so whereas @@request$resultset[1,1] would be the first column of data,
@@request$resultset[0,1] would be the NAME of the first column.

You should be able to loop through that and build an insert statement (:


On 7/27/05, MJPinckard <[EMAIL PROTECTED]> wrote:
 


I'm trying to add an option for users to clone their user session.

I have a table with ~500 columns (yeah I know, but that's how it's
structured, legacy code...ugh).

I want to duplicate this row, changing only the primary key
(session_id). Since it is not inconceivable that the column structure
might change someday, I'd like to make the code as generic as possible
(not coding each column into an update or insert action).

Using directDBMS, I can easily extract all the existing data using the
following code.

   select * from Input where session_id=someNumber

But this gives me data only (no column names) and in any event I can't
figure out how to get the data into another row. I've tried a variety of
different ways (using both INSERT and creating a row in another action
then performing an UPDATE) but can't get anything to work.

Has anyone any suggestions about how I might accomplish this? I'm about
to give in and manually code all 500 columns into an insert action...
but surely there has got to be an easier way...

Mac OS X server
Witango 5.0.9 (I think)
Oracle db using Oracle OCI drivers.

Thanks for any assistance.

Cheers,

Maggie Pinckard
Principal Research Associate
Lawrence Berkeley National Laboratory

TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

   



TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf
 



TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

Witango-Talk: duplicating a row in Oracle

[MJPinckard]

I'm trying to add an option for users to clone their user session.

I have a table with ~500 columns (yeah I know, but that's how it's 
structured, legacy code...ugh).


I want to duplicate this row, changing only the primary key 
(session_id). Since it is not inconceivable that the column structure 
might change someday, I'd like to make the code as generic as possible 
(not coding each column into an update or insert action).


Using directDBMS, I can easily extract all the existing data using the 
following code.


   select * from Input where session_id=someNumber

But this gives me data only (no column names) and in any event I can't 
figure out how to get the data into another row. I've tried a variety of 
different ways (using both INSERT and creating a row in another action 
then performing an UPDATE) but can't get anything to work.


Has anyone any suggestions about how I might accomplish this? I'm about 
to give in and manually code all 500 columns into an insert action... 
but surely there has got to be an easier way...


Mac OS X server
Witango 5.0.9 (I think)
Oracle db using Oracle OCI drivers.

Thanks for any assistance.

Cheers,

Maggie Pinckard
Principal Research Associate
Lawrence Berkeley National Laboratory

TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

Re: Witango-Talk: Mac OS X Timestamp issue: Found Problem.

[MJPinckard]

FWIW, I've been having the same problem Studio 5.5.09 (licensed) on Mac 
OS X 10.3.9


I see the same thing on the timestamp data dictionary, only variation on 
reported, is that if I edit the action, I get DataType="!CST" on every 
column, not just the date column... it's just that the date column is 
the only one to fall over if I don't manually correct it in BBEdit, so 
I've been leaving the others as is.


I had assumed it was just my machine/studio install, since I've also 
been experiencing datasource connection problems (that have stumped 
everyone, including Support). At this point my database connection 
issues have miraculously resolved themselves, I had assumed that the 
timestamp problem was a relict of that issue until this thread came up.


Sincerely,

Maggie Pinckard
Principal Research Associate
Lawrence Berkeley National Laboratory

Roland Dumas wrote:



On Jul 15, 2005, at 12:32 AM, Robert Garcia wrote:


It is definitely the studio.



Here is the DataDictionary of the insert on the bad taf:



phonelog
RVORefNumber


phonelog
PhoneLogged



The data types are clearly messed up, don't know what !CST means, but 
it should be varchar and timestamp. So this is the issue. Mark is 
using Lite version of studio, I am using licensed version. Anyone 
else seeing this problem, what version of the studio are you using?




here's a bad action for me:

orders
order_date


orders
order_billto_firstname


licensed version of the studio


TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf
 



TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

Witango-Talk: weird Oracle update behavior

[MJPinckard]

I'm seeing some strange behavior in connecting to our Oracle database. 
Basically what is happening is that I perform an update of a field and 
then immediately (within a few seconds) perform a search for that same 
record/field the value returned from Oracle is the pre-update value. If 
I wait a little longer (maybe 5-10 seconds) and perform the search again 
(without repeating the update) the value returned is the post update value.


My interpretation is that this looks like some kind of transaction or 
caching issue, although we are set to auto-commit after each update, 
which should address the transaction problem.


Do you have any ideas about what might be causing this problem, and how 
we could identify further the cause and resolve it.


I'm not certain how long this may have been occurring.

DB - HESPRD
Witango 5.01.065
Mac OS X 10.3.8

thanks,

Maggie Pinckard

TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

Re: Witango-Talk: A long time ago in a country far far away ...

[MJPinckard]

Well, I've still got a bunch of archived .qry files... might even have 
the original box with software somewhere around here... I know I had it, 
but it might have been purged when I changed offices this last July. Our 
site (Home Energy Saver - http://hes.lbl.gov) has been on Tango, now 
Witango, for many moons (since middle of 1996 I think).


Cheers,

Maggie Pinckard
Principle Research Associate
Lawrence Berkeley Laboratory

Steve Smith wrote:

The original version (1.0) was Mac, and supported connections only to 
Butler which was EveryWare's SQL database engine. The extension was 
.qry for 'query' files. Tango 1.5 was released Spring of 1996 and had 
both Mac and Windows with ODBC support. Tango for FileMaker was 
version 1.6 and then we had Tango 2.x. That was when Tango for Access 
was released as a separate product. I seem to recall Tango Merchant 
coming out around the same time.


The file extension was changed to .taf with the 3.x release I believe. 
Then came Tango 2000 which was Tango version 4.x but we never referred 
to it as version 4, it was always Tango 2000. That was the last 
EveryWare/Pervasive release. 

Witango started with a repackaged Witango 2000 and Witango 5 was their 
first official release.


I think I've got my facts straight. One little trivia piece was that 
the Bank of Montreal's mBanx, which was a major feather in our cap had 
it's original prototype written and running on a Mac (believe it was a 
PowerMac 7200). They used a 'special' version of the 1.5 release which 
stayed in production until 2000. There were a number of special tags 
which eventually became official tags in future releases. These tags 
were affectionately known as <@Rudy> tags (named after Rudy Wolfs). 


Hope this helps,

Steve Smith


Oakbridge Information Solutions

Office: (519) 624-4388

Fax:(519) 624-3353

Email:  [EMAIL PROTECTED] _

Web:_http://www.oakbridge.ca_



On Jun 3, 2005, at 1:55 PM, <[EMAIL PROTECTED] 
> <[EMAIL PROTECTED] 
> wrote:


What was the original file extension?  I got involved when Illinois 
State University bought a number of 2.0 seats in a package deal.  We 
ran it on a NT box with Access.

__
Bill Prigge aka [EMAIL PROTECTED] 
 phone: 310-570-4950

Director of Information Services
Execpro Information Services, LLC
cell: 310.488.5957

-Original Message-
*From:* David Shelley [mailto:[EMAIL PROTECTED]
*Sent:* Wednesday, June 01, 2005 9:15 PM
*To:* witango-talk@witango.com 
*Subject:* RE: Witango-Talk: A long time ago in a country far far
away ...

I can remember seeing Dan McKenzie and Rudy Wolfs demonstrate a
pre-release version at Dan’s cottage. They sure were excited
about it! Less than a year later I was working with them too.
Sure seems like a long time ago.

 


Dave Shelley

 


-Original Message-
*From:* Steve Smith [mailto:[EMAIL PROTECTED]
*Sent:* Wednesday, June 01, 2005 11:09 PM
*To:* witango-talk@witango.com 
*Subject:* Re: Witango-Talk: A long time ago in a country far far
away ...

 


I missed that by a few months, having joined EveryWare later in
1995 (November) when I became the first 'official' support person
for Tango. However I do remember meeting Mark and Greg Hemstreet
at a Toronto Mac Users Group meeting when they did a demo of
either a late ButlerLink/Web or an early Tango. I had been
playing around with Butler and the FirstClass connectivity that
had been available a year or two earlier. It's only problem was
that we could search Butler from FirstClass, we could create new
records from FirstClass, but we couldn't modify records from
FirstClass and they never did get that working. Tango came along
and it wasn't really needed as much.

 


Brings back a lot of great memories, including shrinkwrapping the
original release of Tango for FileMaker. I do have a set of
floppies around someplace with the Butler install that came with
Tango 1.0.

 


Steve Smith

 


Oakbridge Information Solutions

Office: (519) 624-4388

Fax:(519) 624-3353

Email:  [EMAIL PROTECTED] _

Web:_http://www.oakbridge.ca_

 

 


On May 31, 2005, at 8:08 PM, Phil Wade wrote:



EveryWare Development Corp, June 1995: ButlerLink/Web 1.0.2 was

released as part of the Butler DB suite.  It appeared in the Butler

suite of tools a few months earlier but we have no record of the

official release of 1.0.0.  The earliest record of ButlerLink we know

of was a demonstration of it at the first WebEdge in Austin on April

2-4, 1995 by Henry Lach and Mark Wickens.  It was billed as the "most

acc

Witango-Talk: Problems connecting to Oracle OCI

[MJPinckard]

I'm having difficulties connecting to my Oracle database with the Dev 
Studio.


It used to connect with no difficulties, but now won't. I don't believe 
there has been any changes on the Oracle side (my servers can still 
connect, it is only the dev studio that can't).


I have tried the following:

 1. Deleted and reinstalled the Oracle OCI 8.1.7.1 library
 2. Deleted and reinstalled the Dev Studio (5.5.08)
 3. Deleted the Witango Dev Studio preferences.
 4. Attempted to create a new datasource.
 5. Repaired permissions (Mac OS X version 10.3.9)
 6. restarted the computer multiple times


What can I do to get more information to diagnose and/or fix this 
problem? I need to connect to the database! Is there a transaction log 
somewhere that shows what it is trying to do?


Maggie Pinckard
[EMAIL PROTECTED]

TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

Re: Witango-Talk: Dev studio crash when connecting to Oracle

[MJPinckard]

further question... I've been able to continue working (got a deadline) 
by copying a search action, then manually editing the taf in BBEdit to 
point to the new columns in the new table... but... how necessary is the 
DataDictionary portion of the search action?

At this point, I've just ignored the old DataDictionary code (so it 
still has the old column information therein)... I'm concerned however 
that I'm likely to let myself in for later problems by leaving a 
mismatch between the DataDictionary and the SearchColumns... so, would 
it better to:

A. Delete the DataDictionary section altogether... it would presumably 
be automatically recreated later when/if I ever get the connection to 
the DB working again?

or
B. Manually edit the DataDictionary to conform to the columns in the new 
search?

FWIW, it does seem to be working on the server at present with the 
current mismatch, but this is of course not under load, and currently 
there isn't any data stored in the columns being referenced.

Maggie
MJPinckard wrote:
I'm having a problem connecting to Oracle (using Oracle OCI). If I  
attempt to open an existing configured datasource in the dev studio, 
it tells me that it can't reach the datasource, make sure  it is 
properly configured and online... this used to work (yesterday 
morning) so is properly configured, and I know the  database is online 
because I can access it through another tool - SQLGrinder. I doubt 
anything in the database has changed since then (I'm generally 
notified in advance of any changes).

If I attempt to create a new datasource it crashes after submitting 
from the connect string window. This occurs no matter what type of 
datasource I attempt to create (JDBC, Oracle etc).

I have shutdown and restarted my computer in the hopes that that  
might clear something so it would work, this failed to correct the  
problem.

I have upgraded to 10.3.9 since the last time I created a new 
datasource the web servers aren't having a problem connecting it 
is just my Dev Studio.

Mac OS X 10.3.9 Dev Studio 5.5.08, Oracle OCI
Any suggestions on what to do... I need to resolve this quickly. I've 
put in a request to customer service, but if anyone else has an idea 
on things I can do to help research and resolve the problem, I'd 
appreciate the help.

thanks,
Maggie

TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

Re: Witango-Talk: Witango Studio 5.0 Unexpectedly Quits in OSX

[MJPinckard]

Would like scroll wheel to work myself in 5.5. on OS X.
Maggie
Wayne Irvine wrote:
Robert Garcia at [EMAIL PROTECTED] wrote:
 

That reminds me. I have sent up feature requests to witango, for a
LONG time, for 5.5 studio OS X to support scroll wheel on the mouse.
It is the only program I use, that I cannot quickly scroll through
text/code. I know it seems petty, but it really hinders my workflow.
I am always going to the mousewheel, forgetting it doesn't work, and
I have to do it manually.
Anybody else out there?
   

Yeah, me too. The scroll wheel thing is a pain and so are the issues with
pasting from other apps (targeting wrong window or sourcing from wrong
clipboard etc).
As for corrupt files and datasource changes I have become quite adept at
editing entire TAFs in BBEdit. But if I wanted that sort of hassle I would
have learnt PHP. ;)
I have a fair idea of the issues of allocating resources at WiTango towers
but I think it's time us OS X users got our fair share.
Wayne Irvine 

 Byte Services Pty Ltd
  http://www.byteserve.com.au/
 [EMAIL PROTECTED]
  Ph 02 9960 6099   Mob 0409 960 609   Fax 02 9960 6088

TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf
 


TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

Witango-Talk: Dev studio crash when connecting to Oracle

[MJPinckard]

I'm having a problem connecting to Oracle (using Oracle OCI). If I  
attempt to open an existing configured datasource in the dev studio, it 
tells me that it can't reach the datasource, make sure  it is properly 
configured and online... this used to work (yesterday morning) so is 
properly configured, and I know the  database is online because I can 
access it through another tool - SQLGrinder. I doubt anything in the 
database has changed since then (I'm generally notified in advance of 
any changes).

If I attempt to create a new datasource it crashes after submitting from 
the connect string window. This occurs no matter what type of datasource 
I attempt to create (JDBC, Oracle etc).

I have shutdown and restarted my computer in the hopes that that  might 
clear something so it would work, this failed to correct the  problem.

I have upgraded to 10.3.9 since the last time I created a new 
datasource the web servers aren't having a problem connecting it is 
just my Dev Studio.

Mac OS X 10.3.9 Dev Studio 5.5.08, Oracle OCI
Any suggestions on what to do... I need to resolve this quickly. I've 
put in a request to customer service, but if anyone else has an idea on 
things I can do to help research and resolve the problem, I'd appreciate 
the help.

thanks,
Maggie

TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

Witango-Talk: max seconds in SECSTOTIME tag

[MJPinckard]

What is the maximum value that can be entered as seconds in a 
<@SECSTOTIME> conversion?

I'm getting an "Invalid Parameter!" comment on a conversion I'm trying 
and believe it is due to a limit issue. The code is fine on smaller 
values but chokes on larger values... my data is not continuous, so it 
isn't easy to identify the choke point.

If I know the maximum, I can write an IF block to control the output 
(e.g. print "greater than XX hours" rather than "Invalid parameter!")

Thanks,
Maggie Pinckard
Principle Research Associate
Lawrence Berkeley National Laboratory

TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

Witango-Talk: compound CDelim value

[MJPinckard]

The default delimiter used for CDelim is a comma. I know it is possible 
to use another value (a "*" for example) instead of the comma. Is it 
possible to use a specific set of characters (instead of just one single 
character) as a delimeter in the cdelim paramter?

For example could I use the string "*|" instead of just "*" or "|"?
Alternatively is it possible to use a tab (and what would the syntax 
be... would it be "\t")?

Thanks,
Maggie

TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

Re: Witango-Talk: retrieving cookie values

[MJPinckard]

problem is... server1 isn't a witango server... it's just setting an 
apache cookie.

Maggie
Alan Wolfe wrote:
if you cant find a way to do this, when requesting the file on the second
server, you can always pass the cookie as an argument
ie like...
http://www.server2.com/blah.taf?somecookie=cookie$somecookie
- Original Message - 
From: "MJPinckard" <[EMAIL PROTECTED]>
To: "Witango" <[EMAIL PROTECTED]>
Sent: Wednesday, November 03, 2004 3:53 PM
Subject: Witango-Talk: retrieving cookie values

 

In my web site, users end up getting files from multiple servers.
I wish to check the value of a cookie set on server1 with a witango
process running on server2.
I've tried <@CGIPARAM NAME="HTTP_COOKIE" COOKIE="server1CookieName"> but
the only thing returned are the cookies set on server2.
What is the syntax for requesting (and assigning to a variable) the
value of a cookie set by another server?
thanks,
Maggie

TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf
   


TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf
 


TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

Re: Witango-Talk: retrieving cookie values

[MJPinckard]

server1 is hes0.lbl.gov
server2 is hes3.lbl.gov
will this work?
Maggie
Roland Dumas wrote:
These servers are using the same domain? If different domains, that's a
challenge. Then anyone could read a cookie written by another server.
On 11/3/04 3:53 PM, "MJPinckard" <[EMAIL PROTECTED]> wrote:
 

In my web site, users end up getting files from multiple servers.
I wish to check the value of a cookie set on server1 with a witango
process running on server2.
I've tried <@CGIPARAM NAME="HTTP_COOKIE" COOKIE="server1CookieName"> but
the only thing returned are the cookies set on server2.
What is the syntax for requesting (and assigning to a variable) the
value of a cookie set by another server?
thanks,
Maggie

TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf
   



TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf
 


TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

Witango-Talk: retrieving cookie values

[MJPinckard]

In my web site, users end up getting files from multiple servers.
I wish to check the value of a cookie set on server1 with a witango 
process running on server2.

I've tried <@CGIPARAM NAME="HTTP_COOKIE" COOKIE="server1CookieName"> but 
the only thing returned are the cookies set on server2.

What is the syntax for requesting (and assigning to a variable) the 
value of a cookie set by another server?

thanks,
Maggie

TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

Re: Witango-Talk: Returning all variables within a scope

[MJPinckard]

Fergal, et al,
I upgraded to 5.0.1.065 yesterday, and in my logs today it looks like 
the database connection isn't sticking. This is slowing my site down, 
since every request has to open a new db connection. I checked the 
witango.ini file and the settings between the old and new versions are 
the same (DATASOURCELIFE=30).

I'm seeing (I've replaced the db name and user name with X's, but they 
were correct before replacement):

Jul 21 11:26:16 HES3 WitangoServer[1572]: RUNTIMEINFO No 
existing connection to the data source found, creating a new connection. 
DSN: oracle.XX; User: XX;
Jul 21 11:26:26 HES3 WitangoServer[1572]: RUNTIMEINFO No 
existing connection to the data source found, creating a new connection. 
DSN: oracle.XX; User: XX;
Jul 21 11:26:26 HES3 WitangoServer[1572]: RUNTIMEINFO No 
existing connection to the data source found, creating a new connection. 
DSN: oracle.XX; User: XX;
Jul 21 11:27:09 HES3 WitangoServer[1572]: RUNTIMEINFO No 
existing connection to the data source found, creating a new connection. 
DSN: oracle.XX; User: XX;

This wasn't happening before the upgrade, any suggestions to correct.
Maggie
Fergal Donlon wrote:
Hi Maggie,
Firstly can I suggest you download the latest production version of the
Witango Server  - 5.0.1.065. There were number of issues fixed between .062
and .065 and some did cause crashes.
 

...snip...

TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

Re: Witango-Talk: Returning all variables within a scope

[MJPinckard]

Oh.. I already use that... I've got it set to turn debugging on for my 
machine (by IP address) and off for all others... but that doesn't help 
me in my situation. What I'm trying to do is whenever a user (no 
debugging) runs into a problem and activates an error trap, take a 
snapshot of all the variables for that user (request, user, etc.) and 
print that along with the location in the code/site where the error was 
encountered to a log file... then I can go back and review the log file 
for get details and hopefully track down this intermittent bug.

Alan Wolfe wrote:
this code will toggle debug on or off for a user, not sure if this is what
you were looking for or not.
<@ifempty expr="@@user$debugmode">
 <@assign user$debugmode value="forceon">
<@else>
 <@purge user$debugmode>

this is atleast one of the methods they used at the conference (:
- Original Message - 
From: "Driscoll, Kevin" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Friday, July 16, 2004 4:14 AM
Subject: RE: Witango-Talk: Returning all variables within a scope

Fergal,
Isn't there a variable that determines if debug is on or off?
At the Collaboree I thought someone showed that..
Kevin
-Original Message-
From: Fergal Donlon [mailto:[EMAIL PROTECTED]
Sent: Friday, July 16, 2004 12:51 AM
To: [EMAIL PROTECTED]
Subject: Re: Witango-Talk: Returning all variables within a scope
Hi Maggie,
Firstly can I suggest you download the latest production version of the
Witango Server  - 5.0.1.065. There were number of issues fixed between
.062
and .065 and some did cause crashes.
I think Alan has correctly shown how to display the name and values of
variables in a particular scope. There is also a component on the
component
zone of witango.com entitled 'Show all arguments and variables' which
also
demonstrates how to do this.
Can you look in the witangoevents.log file (located in your
configuration
directory) and see if anything is being written when the crash occurs.
If
this does not reveal anything, can you check your witango.log file. Note
also that you may want to increase the logging level (set in
witango.ini) so
that more information is returned. This will give you the ability to see
what was going on in the lead up to the crash without having debug
switched
on.
I hope this helps and if I can assist you further please do not hesitate
to
contact me.
Fergal
Fergal Donlon
Witango Technologies,
Suite 4, Level 1,
44 Miller Street,
North Sydney, 2060

- Original Message - 
From: "MJPinckard" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Friday, July 16, 2004 11:10 AM
Subject: Witango-Talk: Returning all variables within a scope

 

What would the syntax be to capture all the variables in a specified
scope (e.g in the manner that the debugging prints all "request" and
"user" variables to screen)?
I'm having an intermittent problem, and I'd like to log the user's
variables when the problem occurs, so I can identify the problem. My
thought is to capture variables whenever the error conditions are met
during the next few weeks (the testing period before go-live on this
code). My testers don't normally have debugging turned on, and if I
   

turn
 

debugging on and reload the screen, the intermittent problem clears,
   

so
 

I can't get details.
I'm looking for something along the lines of  <@PURGE
   

scope="request">,
 

only to print, not purge. I tried <@var scope="request"> but it didn't
work.
"Well it works for me" and "just reload, it'll go away" while
   

accurate,
 

is not satisfying to my users.
Thanks for your help.
Maggie Pinckard
=
Principal Research Associate
Enduse Forecasting
Lawrence Berkeley Laboratory
[EMAIL PROTECTED]
   


 

TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf
   


TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf
 


TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

Re: Witango-Talk: Returning all variables within a scope

[MJPinckard]

Unfortunately that seems to return just a list of names, and doesn't 
include the values. I'm looking for something like zipcode=95123 
HeatEnergy=558, etc... this just returns zipcode and HeatEnergy (as a 
single column array).

Thanks anyway, but still looking
Maggie
Alan Wolfe wrote:
<@varnames scope="user">
that will give user scope, you can do any scope.
<@argnames> returns a list of the post and search arguments (:
if you cant figure out your problem, if you post the symptoms, it might be
obvious what the issue is.
If it's not obvious, there are several witango veterans that do witango
consulting that could help you out (Ben Johansen I know for one).
- Original Message - 
From: "MJPinckard" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Thursday, July 15, 2004 6:10 PM
Subject: Witango-Talk: Returning all variables within a scope

 

What would the syntax be to capture all the variables in a specified
scope (e.g in the manner that the debugging prints all "request" and
"user" variables to screen)?
I'm having an intermittent problem, and I'd like to log the user's
variables when the problem occurs, so I can identify the problem. My
thought is to capture variables whenever the error conditions are met
during the next few weeks (the testing period before go-live on this
code). My testers don't normally have debugging turned on, and if I turn
debugging on and reload the screen, the intermittent problem clears, so
I can't get details.
I'm looking for something along the lines of  <@PURGE scope="request">,
only to print, not purge. I tried <@var scope="request"> but it didn't
work.
"Well it works for me" and "just reload, it'll go away" while accurate,
is not satisfying to my users.
Thanks for your help.
Maggie Pinckard
=
Principal Research Associate
Enduse Forecasting
Lawrence Berkeley Laboratory
[EMAIL PROTECTED]

TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf
   


TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf
 


TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

Witango-Talk: @URL request crashing Witango plugin

[MJPinckard]

I incorporate content from another server, and have run into a severe 
problem.

The other server was taken down for maintenance, and whenever my site 
attempts to contact it, I get a hard crash of my Witango server 
(Client/Server Plug-in Error) and have to manually relaunch the Witango 
server to get it to work again (even for non-related requests).

My original thought was to test the value of the @URL and if it was 
empty (e.g. the server was unavailable), eliminate that piece of 
content. But with it crashing at every failed request attempt , I can't 
even test and build an error trap!

What's going on, and how can I work around this problem? Eliminating the 
content from the other server (and hence the @URL call) is not an 
option... I've tried restarting the machine in the hopes that it was a 
temporary settings problem, but no luck there either.

Everything works fine as long as the other server is up, but I obviously 
can't guarantee 100% up time on someone else's machine.

Mac OS X 10.3.4
Witango 5.0.1.062
Maggie Pinckard
=
Principal Research Associate
Enduse Forecasting
Lawrence Berkeley Laboratory
[EMAIL PROTECTED]

TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

Witango-Talk: Returning all variables within a scope

[MJPinckard]

What would the syntax be to capture all the variables in a specified 
scope (e.g in the manner that the debugging prints all "request" and 
"user" variables to screen)?

I'm having an intermittent problem, and I'd like to log the user's 
variables when the problem occurs, so I can identify the problem. My 
thought is to capture variables whenever the error conditions are met 
during the next few weeks (the testing period before go-live on this 
code). My testers don't normally have debugging turned on, and if I turn 
debugging on and reload the screen, the intermittent problem clears, so 
I can't get details.

I'm looking for something along the lines of  <@PURGE scope="request">, 
only to print, not purge. I tried <@var scope="request"> but it didn't 
work.

"Well it works for me" and "just reload, it'll go away" while accurate, 
is not satisfying to my users.

Thanks for your help.
Maggie Pinckard
=
Principal Research Associate
Enduse Forecasting
Lawrence Berkeley Laboratory
[EMAIL PROTECTED]

TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

Witango-Talk: can't get results of external action

[MJPinckard]

I've been banging my head against this for hours, and hope someone can help.
I'm calling a shell script using an external action. I cannot seem to 
get the output of the shell script to return to Witango so it can be 
evaluated in the resultset. Server debugging always shows the following:
*[External Action]* 	*[1112]* 	check_for_mask
*[Query]* 	*[1113]* 	/TOU/masks/checkMask2.sh; Parameters: 
tariffID=430.txt;
*[External Action]* 	*[1113]* 	Forking process
*[External Action]* 	*[1125]* 	Waiting on the child process
*[External Action]* 	*[1226]* 	Error: No child processes
*[External Action]* 	*[1226]* 	Waiting done
*[ActionResults]* 	*[1231]* 	
*[External Action]* 	*[1234]* 	Complete
*[Changed Vars]* 	*[1234]* 	request$maskPresent=;

I know that the "430.txt" file is present and readable by Witango, and 
when I test it on the command line the script returns the name of the 
file. Permissions are set correctly for Witango to execute the script. I 
have several other scripts running (although they don't need to return 
results), and this script is set up identically to the functional ones, 
so I'm reasonably certain it is being called. The variable maskPresent 
is being set to @@resultset.

Here is the actual script.
#! /bin/sh
echo `ls | grep $tariffID`
How can I get the results to come back to Witango? Server is OS X 
(10.3.4) Server running Witango 5.0.1.062 Chimera (Mac OS X)

thanks for any help you can give.
Maggie Pinckard

TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

Witango-Talk: <@CGIPARAM CLIENT_IP> Resolution

[MJPinckard]

Thanks Russell your tip solved the problem. Turning off performance 
caching solved the problem.

Maggie

[EMAIL PROTECTED] wrote:

maggie,

MJPinckard <[EMAIL PROTECTED]> wrote on 08/05/2004 06:04:03 AM:

> Since I upgraded to Witango 5 from Tango 3.5, I can no longer get
> accurate client IPs. The value returned using <@CGIPARAM CLIENT_IP> is
> always localhost. I'm running Witango and Apache on OS 10.3
are you running osx server?  the server has a "perfomance cache" which 
defaults to on, and i think that that could cause this problem.  that 
can be turned off using server admin.

of course, if you are on osx client, i have no idea (sorry).

cheers,

russell

> The problem seems to be caused by the fact that Apache is redirecting
> all Witango requests to the Witango server, therefore all requests are
> indeed coming from localhost.
> I need to either get this (and other CGIPARAM tags) to work as it used
> to or figure out a way to pass the client IP through as a hidden form
> argument (using javascript).
> Any suggestions on how to do either of these two thing reliably?

> Maggie


TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

Witango-Talk: Parsing XML from SOAP request

[MJPinckard]

I'm attempting to use a SOAP service with Witango. After getting a 
response from the soap server, how do I parse the returned XML so that I 
can access the variables sent back. Here is a sample of returned XML



http://schemas.xmlsoap.org/soap/envelope/";
xmlns:xsd="http://www.w3.org/2001/XMLSchema";
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance";
xmlns:SOAP-ENC="http://schemas.xmlsoap.org/soap/encoding/";
xmlns:ns4="urn:onTAP"
SOAP-ENV:encodingStyle="http://schemas.xmlsoap.org/soap/encoding/";>




Arizona Public Service Co
126
Arizona
UNITED STATES
http://tariffs.lbl.gov/utility/utility.php?util_id=126
803


I've tried using <@DOM> but it doesn't work (my guess is because it is a 
soap-env rather than xml). I've attempted to use <@replace> to replace 
the soap envelope information with  but that didn't work either...

Any ideas, recommendations?

thanks,

Maggie Pinckard

--
* Margaret Pinckard
* [EMAIL PROTECTED]
* Principal Research Associate
* End-use Forecasting Group
* E.O. Lawrence Berkeley National Laboratory

TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

Witango-Talk: <@CGIPARAM CLIENT_IP>

[MJPinckard]

Since I upgraded to Witango 5 from Tango 3.5, I can no longer get 
accurate client IPs. The value returned using <@CGIPARAM CLIENT_IP> is 
always localhost. I'm running Witango and Apache on OS 10.3

The problem seems to be caused by the fact that Apache is redirecting 
all Witango requests to the Witango server, therefore all requests are 
indeed coming from localhost.

I need to either get this (and other CGIPARAM tags) to work as it used 
to or figure out a way to pass the client IP through as a hidden form 
argument (using javascript).

Any suggestions on how to do either of these two thing reliably?

Maggie

--
* Margaret Pinckard
* [EMAIL PROTECTED]
* Principal Research Associate
* End-use Forecasting Group
* E.O. Lawrence Berkeley National Laboratory

TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

Witango-Talk: Server 5.5 Release?

[MJPinckard]

I have a project that needs to take advantage of one of the new features
of the 5.5 version.
Does anyone know when the server version is likely to be released? 
I know you probably can't be exact, I'm looking for a rough estimate
(e.g. end of May, end of the summer, end of the year), so that I can
make a judgment call about delaying the task until the server is
released or spending effort to make a coding work-around for this
problem so that I can make my deadline.

Thanks,

Maggie Pinckard

Dan Stein wrote:

Actually I posted this several days ago. But I thought we would see
something from the powers to be also.
I am happy to report I no longer need to request new temp keys/ Yeah!

on 4/2/04 14:59, Robert Shubert at [EMAIL PROTECTED] wrote:

 

It seems no one has made any noise on the list yet...

The Witango 5.5 studio was released on 3/31. There is an announcement on
the www.witango.com homepage, and a press release here:
http://www.witango.com/newsroom/PR31032004.thtml

The release version is 5.5.08 and installers are available from
www.witango.com as well as reseller's sites (should be)
You should consider visiting your favorite reseller to purchase an
upgrade. I like the features and stability I've seen so far. If you
qualify for a free upgrade, Witango Tech will be contacting you
directly.
Note that this only pertains to the Studio. Servers which support the
new 5.5 tags will arrive in the future. If you are interested in using
the new J2EE support, you should enquire at [EMAIL PROTECTED] about the
J2EE run-time libraries.
Enjoy!

Robert Shubert
www.witangostore.com

TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf
   

 

--
* Margaret Pinckard
* [EMAIL PROTECTED]
* Principal Research Associate
* End-use Forecasting Group
* E.O. Lawrence Berkeley National Laboratory

TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

Witango-Talk: purgecache?

[MJPinckard]

I have a taf I use to purge the cached files, so that when I make a code 
change, I can implement it without stopping and restarting the witango 
server.

it consists of a single metatag... <@purgecache>

I get the following error when I try to call the taf

[Results Action][5]Purge_cached_files
[Error]   [5]-27 Invalid Password detected. 
Insufficient security privileges
[request$ Vars] [10]variableTimeout=30;
[Meta Stack] [10](0001): <@PURGECACHE>

OS X 10.3, witango server 5.0.1.062

What do I do to get this working? I really can't bounce the server every 
time I make a code change.

Maggie

--
* Margaret Pinckard
* [EMAIL PROTECTED]
* Principal Research Associate
* End-use Forecasting Group
* E.O. Lawrence Berkeley National Laboratory

TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

Re: Witango-Talk: Crashes on OS X

[MJPinckard]

Just chiming in to say that Timbuktu causes no problem for me. I have it
installed, and typically Timbuktu into the server, and leave the window to
the server open all day (e.g same connection for hours on end), 
periodically
looking over the server logs for problems...

My server is a dual G5 10.3.x  current version of Witango.

I do see crashes, but they are always preceded by a failed Oracle OCI 
connection.
These crashes occur with the same frequency whether I TB2 into the box 
or not
(e.g. over the weekend).

Maggie Pinckard

Chuck Leavens wrote:

Pardon me if you have already looked at this BUT-

Do you have Timbuktu loaded on these machines?  My machines are quite 
stable UNLESS-

1) I TB2 into the box.

2) Database records are empty. (database there but no records)

Either is a recipe to have witangod crash for me.

So I don't Timbuktu and always populate the database with a few dummy 
records to start.

HTH

I have 2 witango licenses on two machines that are essentially 
identical.

Server 1 never crashed for a long time
Server 2 started crashing (witango crash) about once a day. Those 
crashes abated for about a month and returned. the crash log always 
indicates it's an odbc issue, though I can't figure out anything 
more than that.

Server 1 just started crashing. It's about once ever few days. Crash 
log is different than server 2. I don't know what these things mean 
or what I can do about them, but if someone can read the gibberish, 
I'd appreciate some clues as to what I might do (in either case). 
Last incident from server 1:

**

Date/Time:  2004-02-06 01:49:38 -0800
OS Version: 10.2.8 (Build 6R73)
Host:   new
Command:witangod
PID:4058
Exception:  EXC_BAD_ACCESS (0x0001)
Codes:  KERN_INVALID_ADDRESS (0x0001) at 0x028d7fff
Thread 0:
 #0   0x90042588 in semaphore_timedwait_signal_trap
 #1   0x9003e7b4 in _pthread_cond_wait
 #2   0x00152b6c in 0x152b6c
 #3   0x0014e16c in 0x14e16c
 #4   0x3b64 in 0x3b64
 #5   0x270c in 0x270c
 #6   0x258c in 0x258c
Thread 1:
 #0   0x9002568c in select
 #1   0x001498a4 in 0x1498a4
 #2   0xb2a8 in 0xb2a8
 #3   0xb1ac in 0xb1ac
 #4   0x0014e09c in 0x14e09c
 #5   0x90020c28 in _pthread_body
Thread 2:
 #0   0x90042588 in semaphore_timedwait_signal_trap
 #1   0x9003e7b4 in _pthread_cond_wait
 #2   0x0014e6cc in 0x14e6cc
 #3   0x55c0 in 0x55c0
 #4   0x4db8 in 0x4db8
 #5   0x0001a7d4 in 0x1a7d4
 #6   0x0014e09c in 0x14e09c
 #7   0x90020c28 in _pthread_body
Thread 3:
 #0   0x90042588 in semaphore_timedwait_signal_trap
 #1   0x9003e7b4 in _pthread_cond_wait
 #2   0x0014e6cc in 0x14e6cc
 #3   0x55c0 in 0x55c0
 #4   0x4db8 in 0x4db8
 #5   0x0001a7d4 in 0x1a7d4
 #6   0x0014e09c in 0x14e09c
 #7   0x90020c28 in _pthread_body
Thread 4:
 #0   0x9002568c in select
 #1   0x0012a288 in 0x12a288
 #2   0x0012a558 in 0x12a558
 #3   0x0012ce20 in 0x12ce20
 #4   0x0012cadc in 0x12cadc
 #5   0x0012c428 in 0x12c428
 #6   0x0012be3c in 0x12be3c
 #7   0x4988 in 0x4988
 #8   0x483c in 0x483c
 #9   0x0001a944 in 0x1a944
 #10  0x0014e09c in 0x14e09c
 #11  0x90020c28 in _pthread_body
Thread 5:
 #0   0x90042588 in semaphore_timedwait_signal_trap
 #1   0x9003e7b4 in _pthread_cond_wait
 #2   0x0014e6cc in 0x14e6cc
 #3   0x55c0 in 0x55c0
 #4   0x4db8 in 0x4db8
 #5   0x0001a7d4 in 0x1a7d4
 #6   0x0014e09c in 0x14e09c
 #7   0x90020c28 in _pthread_body
Thread 6:
 #0   0x90042588 in semaphore_timedwait_signal_trap
 #1   0x9003e7b4 in _pthread_cond_wait
 #2   0x0014e6cc in 0x14e6cc
 #3   0x55c0 in 0x55c0
 #4   0x4db8 in 0x4db8
 #5   0x0001a7d4 in 0x1a7d4
 #6   0x0014e09c in 0x14e09c
 #7   0x90020c28 in _pthread_body
Thread 7:
 #0   0x90042588 in semaphore_timedwait_signal_trap
 #1   0x9003e7b4 in _pthread_cond_wait
 #2   0x0014e6cc in 0x14e6cc
 #3   0x55c0 in 0x55c0
 #4   0x4db8 in 0x4db8
 #5   0x0001a7d4 in 0x1a7d4
 #6   0x0014e09c in 0x14e09c
 #7   0x90020c28 in _pthread_body
Thread 8:
 #0   0x9002568c in select
 #1   0x0012a288 in 0x12a288
 #2   0x0012a558 in 0x12a558
 #3   0x0012ce20 in 0x12ce20
 #4   0x0012cadc in 0x12cadc
 #5   0x0012c428 in 0x12c428
 #6   0x0012be3c in 0x12be3c
 #7   0x4988 in 0x4988
 #8   0x483c in 0x483c
 #9   0x0001a944 in 0x1a944
 #10  0x0014e09c in 0x14e09c
 #11  0x90020c28 in _pthread_body
Thread 9:
 #0   0x9002568c in select
 #1   0x0012a288 in 0x12a288
 #2   0x0012a558 in 0x12a558
 #3   0x0012ce20 in 0x12ce20
 #4   0x0012cadc in 0x12cadc
 #5   0x0012c428 in 0x12c428
 #6   0x0012be3c in 0x12be3c
 #7   0x4988 in 0x4988
 #8   0x483c in 0x483c
 #9   0x0001a944 in 0x1a944
 #10  0x0014e09c in 0x14e09c
 #11  0x90020c28 in _pthread_body
Thread 10:
 #0   0x90042588 in semaphore_timedwait_signal_trap
 #1   0x9003e7b4 in _pthread_cond_wait
 #2   0x0014e6cc in 0x14e6cc
 #3   0x55c0 in 0x55c0
 #4   0x4db8 in 0x4db8
 #5   0x0001a7d4 in 0x1a7d4
 #6   0x0014e09c in 0x14e09c
 #7   0x90020c28 in _pthread_body
Thread 11 Crashed:
 #0   0x00068b00 in 0x68b00
 #1   0x0004

Re: Witango-Talk: Is it safe to use IFRAME now

[MJPinckard]

Works fine in Mozilla 1.5.1, but the menus weird out in Safari... the 
first selection loads correctly, then it stops working and nothing loads.

Don't know if this is a problem with the Iframes or your menu code... my 
guess is the Javascript. I was having the same problem with my dropdown 
menus (http://hes.lbl.gov/ enter a zipcode in the zipcode box and press 
Go to get to the Witango part of the site), where the first one worked 
and subsequent selections didn't. I've heard that Safari is suppose to 
adhere to strict code guidelines, and when I cleaned up my code the 
problem went away, which supports that theory. Unfortunately, I can't 
remember exactly what was wrong and what I did to fix it sorry. I do 
remember that once I located the problem, fixing it was trivial...

In addition, not all the labels get the "over" highlight color change. 
It seems like the first three items get the highlight, so longer menus 
don't highlight after item three.

Maggie

Charles Brown wrote:

Seems to work fine on Netscape 7.1/Windows 2000.  The main frame is 
initially blank, but fills in nicely with the menu selections.

Dan Stein wrote:

Works fine in Safari for Mac and Camino on Mac.

I know there are at least two others on Mac Netscape and Opera but I don't
use either.


on 1/12/04 16:19, Ben Johansen at [EMAIL PROTECTED] wrote:

 

So are all the other browsers finally catching up to IE ;-)

If some of you with off brand browsers could check out this link and let
me know
http://68.213.121.50

Ben Johansen - http://www.pcforge.com
Authorized Witango & MDaemon Reseller
Available for Witango Developement



TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf
   

 

 
TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf 


--
* * * * * * * * * * * * * * * * * * * * 
* Margaret Pinckard
* [EMAIL PROTECTED]
* Principal Research Associate
* End-use Forecasting Group
* E.O. Lawrence Berkeley National Laboratory


TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

Witango-Talk: Witango Server Freezing overnight

[MJPinckard]

I'm running Witango 5 with Apache under Mac OS X 10.3.

I having problems with the Witango server freezing, which returns a 
plugin error to Apache. Restarting the Witango application restores service.

Nothing is printing to Witangoevents.log. The last item printed in the 
witango.log is generally a notification that the Oracle datasource has 
expired.

When this freeze occurs, it is difficult to get the application to quit 
and free up it's port so I can restart Witango.

Anyone seen this before or have any ideas what is going on?

Maggie

--
* * * * * * * * * * * * * * * * * * * * 
* Margaret Pinckard
* [EMAIL PROTECTED]
* Principal Research Associate
* End-use Forecasting Group
* E.O. Lawrence Berkeley National Laboratory


TO UNSUBSCRIBE: Go to http://www.witango.com/maillist.taf

Witango-Talk: CGIparam CLIENT_IP shows localhost

[MJPinckard]

I have a block of code that turns debugging on for my computer, and off 
for everyone else.

Since I converted from Tango 3.5.2 to Witango 5.0.x running on Mac OSX 
this code doesn't work.

<@IF EXPR="<@CGIPARAM CLIENT_IP> = 'myIPaddress'">
   <@ASSIGN NAME="debugMode"  VALUE="forceOn" scope="User">
<@ELSE>
   <@ASSIGN NAME="debugMode"  VALUE="forceOff" scope="User">

When I look at the Witango.log all requests are coming from localhost, 
even though the apache log shows the correct use IP.

How do I get Witango to get the correct IP?

Maggie

--
* * * * * * * * * * * * * * * * * * * * 
* Margaret Pinckard
* [EMAIL PROTECTED]
* Principal Research Associate
* End-use Forecasting Group
* E.O. Lawrence Berkeley National Laboratory


TO UNSUBSCRIBE: Go to http://www.witango.com/maillist.taf

Witango-Talk: Write Action problem

[MJPinckard]

I'm trying to write a file to a mounted drive using the write action.

If I mount it using appletalk file protocol (afp) it fails with the 
following error
-304 Unable to obtain a write lock on the file

If I mount the external drive using smb I get this error.
-303 Can not open the specified file.
I've checked, and authenticated users (e.g. witango) have full 
permissions to the directory in question.

Any ideas what's wrong?

Mac OS X 10.3, Apache, Witango 5.0.x

thanking ye much.

Maggie

--
* * * * * * * * * * * * * * * * * * * * 
* Margaret Pinckard
* [EMAIL PROTECTED]
* Principal Research Associate
* End-use Forecasting Group
* E.O. Lawrence Berkeley National Laboratory


TO UNSUBSCRIBE: Go to http://www.witango.com/maillist.taf

Re: Witango-Talk: Invalid Response (resolution)

[MJPinckard]

I think I've found the problem. For memory management purposes (under 
T3.5) I was using local scope for my variables, and when I finished with 
a batch, would purge local scope.

In Witango 5 the next search action after a purge of local scope was 
failing. I found this because I tried changing the local scope to 
request scope (hoping that the problem might be related to the change in 
scope terminology), which didn't fix the problem but did make it obvious 
what the problem was. When I remove the <@purge scope="request"> I'm 
back in business with no crashes. I hope the purge scope isn't necessary 
and the built in memory management can cope with the variables hanging 
around (I have hundreds of variables)... I suppose I can purge each 
individually if need be... I have a batch of code doing that with no 
problems.

Hopefully all will proceed swimmingly from here...

Maggie

MJPinckard wrote:

I just tested that by changing the branch in the previous taf, so that 
rather than pointing back to the suspect taf where it has been dying, 
it points to a new taf. At the same point in execution (e.g. X actions 
after the branch) in the new taf the server crashed. I don't think it 
is the code of any particular taf, it is the process. How can I 
monitor what is going on in the process?

Maggie

Jeff Bohmer wrote:

To me, it just means your server crashed ;)

But now at least you're getting the proper messages in 
witangoevents.log and that's critical.

At this point, if I were in your shoes, I'd rebuild the TAF. 
Hopefully, it isn't huge.  I know you've tried replacing actions, but 
that may not replace everything involved with the code being 
executed.  Since the crash always happens with this TAF, it follows 
that replacing the entire TAF will either fix the problem or finally 
rule out the TAF as the culprit.

- Jeff

--
* * * * * * * * * * * * * * * * * * * * 
* Margaret Pinckard
* [EMAIL PROTECTED]
* Principal Research Associate
* End-use Forecasting Group
* E.O. Lawrence Berkeley National Laboratory


TO UNSUBSCRIBE: Go to http://www.witango.com/maillist.taf

Re: Witango-Talk: Invalid Response

[MJPinckard]

I just tested that by changing the branch in the previous taf, so that 
rather than pointing back to the suspect taf where it has been dying, it 
points to a new taf. At the same point in execution (e.g. X actions 
after the branch) in the new taf the server crashed. I don't think it is 
the code of any particular taf, it is the process. How can I monitor 
what is going on in the process?

Maggie

Jeff Bohmer wrote:

To me, it just means your server crashed ;)

But now at least you're getting the proper messages in 
witangoevents.log and that's critical.

At this point, if I were in your shoes, I'd rebuild the TAF. 
Hopefully, it isn't huge.  I know you've tried replacing actions, but 
that may not replace everything involved with the code being 
executed.  Since the crash always happens with this TAF, it follows 
that replacing the entire TAF will either fix the problem or finally 
rule out the TAF as the culprit.

- Jeff



ah ha!  Caught a fatal

[  838] 2003-12-05 16:06:52 RUNTIMEFATALCaught fatal signal 
10 (SIGBUS); thread id = 25566208; code: 1; address: 000196d0; value: 
; errno: 0; status: 0;

know what it means?

Maggie

Jeff Bohmer wrote:


Perhaps it could be a time out or overload of the db connection? I 
still get no indication of a problem in the witangoevent.log... the 
last recorded event was the initial startup of the server (the last 
time I manually restarted it after changing the witango.ini file). 
It never shows the restarts(?) that I think are happening based on 
the change in PID following one of my errors.


That seems strange.  If witangoevents.log does not show anything for 
the app server starting up, it may indicate a permissions problem on 
the configuration directory.  Stop the app server and do this in 
Terminal (assuming default install location):

sudo chown -R witango /Applications/Witango/Server/configuration

sudo chmod 755 /Applications/Witango/Server/configuration

Then start it and try again.  If this doesn't fix the problem, 
please send the output of the ps command you're using to check for 
witangod's PID before/after you run the TAF.

- Jeff



Maggie

Bill Conlon wrote:

Have you turned on debugging (level 4).

I wouldn't be surprised if there is a divide check or some other 
arithmetic exception, particularly if the calculation is done 
using JavaScript.

No it doesn't crash... it is delighted to send me the same error 
repeatedly... and always fails at the same location in the 
process...

Maggie

Jon van der Raadt wrote:


Is the server crashed at this point or does it come back with 
the next  taf?  There are a few other options for running the 
server that give a  bit more information someone else will know 
them better...  I believe I  was using ./witangod -uo

Jon
On Dec 4, 2003, at 7:44 PM, MJPinckard wrote:
 

Actually when this occurs, I get no debugging, even though it 
is turned on... so no debugging is available to peruse.

witangoevents.log shows nothing after the last time the server 
started  up...

any other ideas on where to look?

Maggie

Jon van der Raadt wrote:

   

Do you have debug turned on -- sometimes that will help 
determine the   problem.  You can also look in the 
witangoevents.log It will record  a  crash if that is 
happening...

On Dec 4, 2003, at 7:26 PM, MJPinckard wrote:

 

I'm getting a Client/Plug-in error message and hope someone 
knows  what's going on. Here's the message.

 Client/Plug-in Error

 The response received from the application server is invalid.

It always happens at the same location in my application 
process,  but  doesn't seem to be tied to any specific code 
(e.g. if I remove  the  search action causing the error, the 
next search action fails,  if I  duplicate a earlier 
successful search action right before the  failing  action 
then it fails on the duplicated action).

It seems as if it is something like a maxaction limitation 
but I've   got maxaction set to 0, which if I understand 
correctly is unlimited   (I tested setting maxaction to 4000, 
way more than I need, but that   didn't change anything). The 
failure is happening quickly which  rules out time out 
problems, and it is not deep in branch actions which  rules 
that out.

If I look on the console, I see the SQL query being parsed, 
then a   line including system information and the License 
number.
Mac OX 10.3 running Witango 5.0.x (current version) under 
Apache. I   don't see a response from the database.

Anyone know what this is and how to fix or can steer me 
towards more   diagnostics?

Thanks,

Maggie

* * * * * * * * * * * * * * * * * * * *
* Margaret Pinckard
* [EMAIL PROTECTED]
* Principal Research Associate
* End-use Forecasting Group
* E.O. Lawrence Berkeley National Laboratory
_ 
__ _
TO UNSUBSCRIBE: Go to http://www.witango.com/mai

Witango-Talk: SOAP client

[MJPinckard]

I'm trying to spec out how to go about accessing/using a SOAP client 
from Witango. We want to add a process to use utility tariff information 
from another project here at the Lab to calculate energy bills estimated 
by my site (which is all in Witango). Someone else will be building the 
SOAP server and getting it to query his database of tariffs. I simply 
need to be able to access a SOAP client to parse a query, and receive 
the results and incorporate them in code returned to users.

The key-value pairs sent back and forth will not change, just the 
"values" in the value part of each pair (which will be supplied from 
Witango calculated variables).

At this point I'm trying to get enough information to put together a 
budget proposal (relative effort or money to buy existing code), and am 
interested in hearing suggestions about how I might go about this. I 
don't particularly care how it happens (via AppleScript (I'm on Mac OSX) 
or, by hard-coding a request in Witango which would send it by IP 
address) but I need to pass values from Witango through SOAP and get 
results back. I don't know anything about SOAP, but am the only one who 
speaks Witango (and it's my site), so I'll be tying whatever we do in...

Any suggestions would help.

Maggie

--
* * * * * * * * * * * * * * * * * * * * 
* Margaret Pinckard
* [EMAIL PROTECTED]
* Principal Research Associate
* End-use Forecasting Group
* E.O. Lawrence Berkeley National Laboratory


TO UNSUBSCRIBE: Go to http://www.witango.com/maillist.taf

Re: Witango-Talk: Invalid Response

[MJPinckard]

ah ha!  Caught a fatal

[  838] 2003-12-05 16:06:52 RUNTIMEFATALCaught fatal signal 10 
(SIGBUS); thread id = 25566208; code: 1; address: 000196d0; value: 
; errno: 0; status: 0;

know what it means?

Maggie

Jeff Bohmer wrote:


Perhaps it could be a time out or overload of the db connection? I 
still get no indication of a problem in the witangoevent.log... the 
last recorded event was the initial startup of the server (the last 
time I manually restarted it after changing the witango.ini file). It 
never shows the restarts(?) that I think are happening based on the 
change in PID following one of my errors.


That seems strange.  If witangoevents.log does not show anything for 
the app server starting up, it may indicate a permissions problem on 
the configuration directory.  Stop the app server and do this in 
Terminal (assuming default install location):

sudo chown -R witango /Applications/Witango/Server/configuration

sudo chmod 755 /Applications/Witango/Server/configuration

Then start it and try again.  If this doesn't fix the problem, please 
send the output of the ps command you're using to check for witangod's 
PID before/after you run the TAF.

- Jeff




Maggie

Bill Conlon wrote:

Have you turned on debugging (level 4).

I wouldn't be surprised if there is a divide check or some other 
arithmetic exception, particularly if the calculation is done using 
JavaScript.


No it doesn't crash... it is delighted to send me the same error 
repeatedly... and always fails at the same location in the process...

Maggie

Jon van der Raadt wrote:

 

Is the server crashed at this point or does it come back with the 
next  taf?  There are a few other options for running the server 
that give a  bit more information someone else will know them 
better...  I believe I  was using ./witangod -uo

Jon
On Dec 4, 2003, at 7:44 PM, MJPinckard wrote:
   

Actually when this occurs, I get no debugging, even though it is 
turned on... so no debugging is available to peruse.

witangoevents.log shows nothing after the last time the server 
started  up...

any other ideas on where to look?

Maggie

Jon van der Raadt wrote:

 

Do you have debug turned on -- sometimes that will help 
determine the   problem.  You can also look in the 
witangoevents.log It will record  a  crash if that is happening...

On Dec 4, 2003, at 7:26 PM, MJPinckard wrote:

   

I'm getting a Client/Plug-in error message and hope someone 
knows  what's going on. Here's the message.

 Client/Plug-in Error

 The response received from the application server is invalid.

It always happens at the same location in my application 
process,  but  doesn't seem to be tied to any specific code 
(e.g. if I remove  the  search action causing the error, the 
next search action fails,  if I  duplicate a earlier successful 
search action right before the  failing  action then it fails 
on the duplicated action).

It seems as if it is something like a maxaction limitation but 
I've   got maxaction set to 0, which if I understand correctly 
is unlimited   (I tested setting maxaction to 4000, way more 
than I need, but that   didn't change anything). The failure is 
happening quickly which  rules  out time out problems, and it 
is not deep in branch actions  which  rules that out.

If I look on the console, I see the SQL query being parsed, 
then a   line including system information and the License number.
Mac OX 10.3 running Witango 5.0.x (current version) under 
Apache. I   don't see a response from the database.

Anyone know what this is and how to fix or can steer me towards 
more   diagnostics?

Thanks,

Maggie

* * * * * * * * * * * * * * * * * * * *
* Margaret Pinckard
* [EMAIL PROTECTED]
* Principal Research Associate
* End-use Forecasting Group
* E.O. Lawrence Berkeley National Laboratory
_ 
__ _
TO UNSUBSCRIBE: Go to http://www.witango.com/maillist.taf

 
__ 
__
TO UNSUBSCRIBE: Go to http://www.witango.com/maillist.taf
   


--
* * * * * * * * * * * * * * * * * * * * * Margaret Pinckard
* [EMAIL PROTECTED]
* Principal Research Associate
* End-use Forecasting Group
* E.O. Lawrence Berkeley National Laboratory
___ 
_
TO UNSUBSCRIBE: Go to http://www.witango.com/maillist.taf

 
---
Jon van der Raadt
TeamOnly Network

412, Alastair Ross Technology Centre
3553 - 31 Street NW
Calgary, Alberta Canada  T2L 2K7
http://www.teamonly.com

  Office: (403) 286-5586
 Fax: (403) 286-5576
 Cell: (403) 863-6304
Email: [EMAIL PROTECTED]
 

TO UNSUBSCRIBE: Go to http://www.witango.com/maillist.taf
   
--
* * * * * * * * * * * * * * * * * * * * * Margaret Pinckard
* [EMAIL PROTECTED

Re: Witango-Talk: Invalid Response

[MJPinckard]

I have, it doesn't look any different at this point than does level 3.

All I'm doing is grabbing a bunch of columns from db then assigning them 
to variables in order to later run a bunch of @Calcs in following 
Results actions (no display, it is all under the hood at this point).

But, it dies here on any search action (regardless of which tables in 
db, which columns retrieved or which criteria used for search) I tested 
all the previous to eliminate the problem being in the code... I really 
think it is in the server or my set up, not the code. Known working 
searches (e.g. the process went through them successfully in order to 
reach the failure point)  fail when moved to this particular location in 
the calculation process.

I just tried putting a branch to another taf right in front of the 
offending action it branched, and then on the next action (a Search) it 
died just like before.

I know the particular search criteria provide a result (I've taken the 
SQL into SQLGrinder and queried the db directly and received a response).

Perhaps it could be a time out or overload of the db connection? I still 
get no indication of a problem in the witangoevent.log... the last 
recorded event was the initial startup of the server (the last time I 
manually restarted it after changing the witango.ini file). It never 
shows the restarts(?) that I think are happening based on the change in 
PID following one of my errors.

Maggie

Bill Conlon wrote:

Have you turned on debugging (level 4).

I wouldn't be surprised if there is a divide check or some other 
arithmetic exception, particularly if the calculation is done using 
JavaScript.

 

No it doesn't crash... it is delighted to send me the same error 
repeatedly... and always fails at the same location in the process...

Maggie

Jon van der Raadt wrote:

   

Is the server crashed at this point or does it come back with the 
next  taf?  There are a few other options for running the server that 
give a  bit more information someone else will know them better...  I 
believe I  was using ./witangod -uo

Jon
On Dec 4, 2003, at 7:44 PM, MJPinckard wrote:
 

Actually when this occurs, I get no debugging, even though it is  
turned on... so no debugging is available to peruse.

witangoevents.log shows nothing after the last time the server 
started  up...

any other ideas on where to look?

Maggie

Jon van der Raadt wrote:

   

Do you have debug turned on -- sometimes that will help determine 
the   problem.  You can also look in the witangoevents.log It will 
record  a  crash if that is happening...

On Dec 4, 2003, at 7:26 PM, MJPinckard wrote:

 

I'm getting a Client/Plug-in error message and hope someone knows   
what's going on. Here's the message.

 Client/Plug-in Error

 The response received from the application server is invalid.

It always happens at the same location in my application process,  
but  doesn't seem to be tied to any specific code (e.g. if I 
remove  the  search action causing the error, the next search 
action fails,  if I  duplicate a earlier successful search action 
right before the  failing  action then it fails on the duplicated 
action).

It seems as if it is something like a maxaction limitation but 
I've   got maxaction set to 0, which if I understand correctly is 
unlimited   (I tested setting maxaction to 4000, way more than I 
need, but that   didn't change anything). The failure is happening 
quickly which  rules  out time out problems, and it is not deep in 
branch actions  which  rules that out.

If I look on the console, I see the SQL query being parsed, then 
a   line including system information and the License number.
Mac OX 10.3 running Witango 5.0.x (current version) under Apache. 
I   don't see a response from the database.

Anyone know what this is and how to fix or can steer me towards 
more   diagnostics?

Thanks,

Maggie

* * * * * * * * * * * * * * * * * * * *
* Margaret Pinckard
* [EMAIL PROTECTED]
* Principal Research Associate
* End-use Forecasting Group
* E.O. Lawrence Berkeley National Laboratory
_ 
__ _
TO UNSUBSCRIBE: Go to http://www.witango.com/maillist.taf

   

__ 
__
TO UNSUBSCRIBE: Go to http://www.witango.com/maillist.taf
 

--
* * * * * * * * * * * * * * * * * * * * * Margaret Pinckard
* [EMAIL PROTECTED]
* Principal Research Associate
* End-use Forecasting Group
* E.O. Lawrence Berkeley National Laboratory
___ 
_
TO UNSUBSCRIBE: Go to http://www.witango.com/maillist.taf

   

---
Jon van der Raadt
TeamOnly Network

412, Alastair Ross Technology Centre
3553 - 31 Street NW
Calgary, Alberta Canada  T2L 2K7
http://www.teamonly.com

  Office: (403) 286-5586
 Fax: (403

Re: Witango-Talk: Invalid Response

[MJPinckard]

Nope, running the converter didn't work.

Maggie

MJPinckard wrote:

I can try, but it doesn't seem called for since I've tested the 
following:
   1. rebuild the offending action entirely in Witango 5 Dev studio
   2. substitute a known working action for the offending action
In both cases, it still crashes at that position in the taf.

I've even tried looking at the offending taf in BBEdit, to see if 
there is garbage in the XML, but nothing shows up.

Maggie

Jeff Bohmer wrote:

Another thought: if this taf has been converted from T3/MacOS, there 
may be a problem with high ASCII characters.  MacOS and OS X use 
different character sets (MacRoman and Latin-1, respectively).  If 
there is a MacRoman high ASCII character in this TAF it could cause 
the app server to crash when it hits it.

There is a utility that comes with the Witango Dev Studio called 
"MacROman to Latin-1 Converter".  Try running that on the TAF in 
question.  Be sure to only run it through the converter once! (And 
make your own backup of the TAF first).  Then, try the "converted" 
TAF and see if the app server still crashes.

- Jeff


Can you rebuild the TAF from scratch?  There may be a problem with 
the XML in this particular TAF that causes the server to die.

- Jeff



Checking the PID, the server is crashing and restarting... e.g. PID 
changes after getting the error message.

1. Caching was turned off... I turned it on, but no change in problem.
2. Yes other TAFs hit the db and run. I've tried replacing the 
offending Search with a Search from the successful TAF, but it 
still dies at the same location.

Maggie

Jeff Bohmer wrote:

The last line in that log excerpt seems to indicate the app server 
starting up.  This could happen if the app server crashes and 
restarts itself.  Check witangoevents.log again for any crash 
information (it should be in a line with [FATAL]).  Also, pay 
attention to the process ID for the witango app server.  If it 
crashes, that PID will change.  Use ps to check on the app 
server's PID, like this:

[berthoud:~] vl_admin% ps axc | grep witango
  428  ??  Ss50:40.01 witangod
(428 is that process' ID)

If you are getting app server crashes, check the following first:

- Is caching turned on in witango.ini?  I've found that the app 
server on OS X tends to crash a lot if caching is turned off.

- Do you have other TAFs that hit your database and run without 
problems?

- Jeff

Here is a selection from the witango log...

04/12/200317:12:51127.0.0.1
[EMAIL PROTECTED]  2549606415354
[Results Action] Start_calcs
04/12/200317:12:51127.0.0.1
[EMAIL PROTECTED]  2549606415354
[Search Action] Assign_cw_kwh
04/12/200317:12:51127.0.0.1
[EMAIL PROTECTED]  2549606415354
[Query] SELECT A1.CWENRG FROM HESADM.APPLIANCES A1 WHERE 
(A1.YR=1993)
04/12/200317:12:54Version: 5.0.1.062 Chimera (Mac OS X), 
License number: PB1JQ---, Process ID: 2227

the Result Action (Start_calcs) works, the Assign_cw_kwh Search 
action is where it is failing... but it doesn't seem to be 
related to the code, since I can substitute any other search 
action (including one right before Start_calcs that works) and it 
still fails at that location. In all instances, immediately after 
getting the Client/Plug-in Error the final item in the log is the 
last line, which re-iterates the License number.

Is there something other than maxactions which limits the number 
of actions that can be executed? My site is performing a building 
energy calculation (not the typical ecommerce transaction) and 
there are a lot of steps between user pages. This site is 
currently running on a beige G3 Mac running OS 8.6 and Tango 
3.5.2... it doesn't seem like it would overload a dual processor 
1.8 Ghz G5 running OS X and Witango 5 and since this problem 
seems independent of the actual code being executed, I'm not sure 
what to do.

Maggie

Customer Support wrote:

Maggie,
Set LOGGINGLEVEL=3 in the witango.ini file.  This will log the 
debug  like information to the witango log.  Stop and start the 
server with  the following terminal command:

sudo SystemStarter restart WITANGO5

It will ask for an admin password.  Once the command has 
finished you  will see some output like this:

Welcome to Macintosh.
Stopping Witango server
Waiting for WITANGO5
Starting Witango Application Server
[ 4106] 2003-12-05 13:49:58 START  INFO Starting Witango 
Application Server Process
Startup complete.
Hangup

This indicates that the server has restarted.

Now run the taf again and have a look in the witango.log file. 
You should be able to see what the error is that is causing the 
server to  return an incorrect response to the web plugin.  What 
is the error  message?

Witango Support

On 05/12/2003, at 1:44 PM, MJPinckard wrote:

Actually when this occurs, I get no debugg

Re: Witango-Talk: Invalid Response

[MJPinckard]

I can try, but it doesn't seem called for since I've tested the following:
   1. rebuild the offending action entirely in Witango 5 Dev studio
   2. substitute a known working action for the offending action
In both cases, it still crashes at that position in the taf.
I've even tried looking at the offending taf in BBEdit, to see if there 
is garbage in the XML, but nothing shows up.

Maggie

Jeff Bohmer wrote:

Another thought: if this taf has been converted from T3/MacOS, there 
may be a problem with high ASCII characters.  MacOS and OS X use 
different character sets (MacRoman and Latin-1, respectively).  If 
there is a MacRoman high ASCII character in this TAF it could cause 
the app server to crash when it hits it.

There is a utility that comes with the Witango Dev Studio called 
"MacROman to Latin-1 Converter".  Try running that on the TAF in 
question.  Be sure to only run it through the converter once! (And 
make your own backup of the TAF first).  Then, try the "converted" TAF 
and see if the app server still crashes.

- Jeff


Can you rebuild the TAF from scratch?  There may be a problem with 
the XML in this particular TAF that causes the server to die.

- Jeff



Checking the PID, the server is crashing and restarting... e.g. PID 
changes after getting the error message.

1. Caching was turned off... I turned it on, but no change in problem.
2. Yes other TAFs hit the db and run. I've tried replacing the 
offending Search with a Search from the successful TAF, but it still 
dies at the same location.

Maggie

Jeff Bohmer wrote:

The last line in that log excerpt seems to indicate the app server 
starting up.  This could happen if the app server crashes and 
restarts itself.  Check witangoevents.log again for any crash 
information (it should be in a line with [FATAL]).  Also, pay 
attention to the process ID for the witango app server.  If it 
crashes, that PID will change.  Use ps to check on the app server's 
PID, like this:

[berthoud:~] vl_admin% ps axc | grep witango
  428  ??  Ss50:40.01 witangod
(428 is that process' ID)

If you are getting app server crashes, check the following first:

- Is caching turned on in witango.ini?  I've found that the app 
server on OS X tends to crash a lot if caching is turned off.

- Do you have other TAFs that hit your database and run without 
problems?

- Jeff

Here is a selection from the witango log...

04/12/200317:12:51127.0.0.1
[EMAIL PROTECTED]  2549606415354
[Results Action] Start_calcs
04/12/200317:12:51127.0.0.1
[EMAIL PROTECTED]  2549606415354[Search 
Action] Assign_cw_kwh
04/12/200317:12:51127.0.0.1
[EMAIL PROTECTED]  2549606415354[Query] 
SELECT A1.CWENRG FROM HESADM.APPLIANCES A1 WHERE (A1.YR=1993)
04/12/200317:12:54Version: 5.0.1.062 Chimera (Mac OS X), 
License number: PB1JQ---, Process ID: 2227

the Result Action (Start_calcs) works, the Assign_cw_kwh Search 
action is where it is failing... but it doesn't seem to be related 
to the code, since I can substitute any other search action 
(including one right before Start_calcs that works) and it still 
fails at that location. In all instances, immediately after 
getting the Client/Plug-in Error the final item in the log is the 
last line, which re-iterates the License number.

Is there something other than maxactions which limits the number 
of actions that can be executed? My site is performing a building 
energy calculation (not the typical ecommerce transaction) and 
there are a lot of steps between user pages. This site is 
currently running on a beige G3 Mac running OS 8.6 and Tango 
3.5.2... it doesn't seem like it would overload a dual processor 
1.8 Ghz G5 running OS X and Witango 5 and since this problem 
seems independent of the actual code being executed, I'm not sure 
what to do.

Maggie

Customer Support wrote:

Maggie,
Set LOGGINGLEVEL=3 in the witango.ini file.  This will log the 
debug  like information to the witango log.  Stop and start the 
server with  the following terminal command:

sudo SystemStarter restart WITANGO5

It will ask for an admin password.  Once the command has finished 
you  will see some output like this:

Welcome to Macintosh.
Stopping Witango server
Waiting for WITANGO5
Starting Witango Application Server
[ 4106] 2003-12-05 13:49:58 START  INFO Starting Witango 
Application Server Process
Startup complete.
Hangup

This indicates that the server has restarted.

Now run the taf again and have a look in the witango.log file. 
You should be able to see what the error is that is causing the 
server to  return an incorrect response to the web plugin.  What 
is the error  message?

Witango Support

On 05/12/2003, at 1:44 PM, MJPinckard wrote:

Actually when this occurs, I get no debugging, even though it is 
turned on... so no debugging is available to peruse.

witan

Re: Witango-Talk: Invalid Response

[MJPinckard]

Is there a limit of the number of tafs a process can branch through (all 
without returns, branches are just used for routing)? My site is made of 
smaller files that are run in sequence for the final process. In this 
case we've gone through four tafs before we get to the error point. 
(Start -> AppCalc -> MiscCalc -> AppCalc) and there are about five more 
tafs to go before we finish the calculation. Any reason to believe this 
is different in Witango 5 from the way it worked in Tango 3.5.2?

Maggie

Jeff Bohmer wrote:

The last line in that log excerpt seems to indicate the app server 
starting up.  This could happen if the app server crashes and restarts 
itself.  Check witangoevents.log again for any crash information (it 
should be in a line with [FATAL]).  Also, pay attention to the process 
ID for the witango app server.  If it crashes, that PID will change.  
Use ps to check on the app server's PID, like this:

[berthoud:~] vl_admin% ps axc | grep witango
  428  ??  Ss50:40.01 witangod
(428 is that process' ID)

If you are getting app server crashes, check the following first:

- Is caching turned on in witango.ini?  I've found that the app server 
on OS X tends to crash a lot if caching is turned off.

- Do you have other TAFs that hit your database and run without problems?

- Jeff



Here is a selection from the witango log...

04/12/200317:12:51127.0.0.1[EMAIL PROTECTED]   
2549606415354[Results Action] Start_calcs
04/12/200317:12:51127.0.0.1[EMAIL PROTECTED]   
2549606415354[Search Action] Assign_cw_kwh
04/12/200317:12:51127.0.0.1[EMAIL PROTECTED]   
2549606415354[Query] SELECT A1.CWENRG FROM 
HESADM.APPLIANCES A1 WHERE (A1.YR=1993)
04/12/200317:12:54Version: 5.0.1.062 Chimera (Mac OS X), 
License number: PB1JQ---, Process ID: 2227

the Result Action (Start_calcs) works, the Assign_cw_kwh Search 
action is where it is failing... but it doesn't seem to be related to 
the code, since I can substitute any other search action (including 
one right before Start_calcs that works) and it still fails at that 
location. In all instances, immediately after getting the 
Client/Plug-in Error the final item in the log is the last line, 
which re-iterates the License number.

Is there something other than maxactions which limits the number of 
actions that can be executed? My site is performing a building energy 
calculation (not the typical ecommerce transaction) and there are a 
lot of steps between user pages. This site is currently running on a 
beige G3 Mac running OS 8.6 and Tango 3.5.2... it doesn't seem like 
it would overload a dual processor 1.8 Ghz G5 running OS X and 
Witango 5 and since this problem seems independent of the actual 
code being executed, I'm not sure what to do.

Maggie

Customer Support wrote:

Maggie,
Set LOGGINGLEVEL=3 in the witango.ini file.  This will log the 
debug  like information to the witango log.  Stop and start the 
server with  the following terminal command:

sudo SystemStarter restart WITANGO5

It will ask for an admin password.  Once the command has finished 
you  will see some output like this:

Welcome to Macintosh.
Stopping Witango server
Waiting for WITANGO5
Starting Witango Application Server
[ 4106] 2003-12-05 13:49:58 START  INFO Starting Witango 
Application Server Process
Startup complete.
Hangup

This indicates that the server has restarted.

Now run the taf again and have a look in the witango.log file.  You 
should be able to see what the error is that is causing the server 
to  return an incorrect response to the web plugin.  What is the 
error  message?

Witango Support

On 05/12/2003, at 1:44 PM, MJPinckard wrote:

Actually when this occurs, I get no debugging, even though it is 
turned on... so no debugging is available to peruse.

witangoevents.log shows nothing after the last time the server 
started  up...

any other ideas on where to look?

Maggie

Jon van der Raadt wrote:

Do you have debug turned on -- sometimes that will help determine 
the   problem.  You can also look in the witangoevents.log It will 
record  a  crash if that is happening...

On Dec 4, 2003, at 7:26 PM, MJPinckard wrote:

I'm getting a Client/Plug-in error message and hope someone 
knows  what's going on. Here's the message.

  Client/Plug-in Error

  The response received from the application server is invalid.

It always happens at the same location in my application 
process,  but  doesn't seem to be tied to any specific code (e.g. 
if I remove  the  search action causing the error, the next 
search action fails,  if I  duplicate a earlier successful search 
action right before the  failing  action then it fails on the 
duplicated action).

It seems as if it is something like a maxaction limitation but 
I've   got maxaction set to 0, which if I understand correctly is 
unlimited  

Re: Witango-Talk: Invalid Response

[MJPinckard]

Checking the PID, the server is crashing and restarting... e.g. PID 
changes after getting the error message.

1. Caching was turned off... I turned it on, but no change in problem.
2. Yes other TAFs hit the db and run. I've tried replacing the offending 
Search with a Search from the successful TAF, but it still dies at the 
same location.

Maggie

Jeff Bohmer wrote:

The last line in that log excerpt seems to indicate the app server 
starting up.  This could happen if the app server crashes and restarts 
itself.  Check witangoevents.log again for any crash information (it 
should be in a line with [FATAL]).  Also, pay attention to the process 
ID for the witango app server.  If it crashes, that PID will change.  
Use ps to check on the app server's PID, like this:

[berthoud:~] vl_admin% ps axc | grep witango
  428  ??  Ss50:40.01 witangod
(428 is that process' ID)

If you are getting app server crashes, check the following first:

- Is caching turned on in witango.ini?  I've found that the app server 
on OS X tends to crash a lot if caching is turned off.

- Do you have other TAFs that hit your database and run without problems?

- Jeff



Here is a selection from the witango log...

04/12/200317:12:51127.0.0.1[EMAIL PROTECTED]   
2549606415354[Results Action] Start_calcs
04/12/200317:12:51127.0.0.1[EMAIL PROTECTED]   
2549606415354[Search Action] Assign_cw_kwh
04/12/200317:12:51127.0.0.1[EMAIL PROTECTED]   
2549606415354[Query] SELECT A1.CWENRG FROM 
HESADM.APPLIANCES A1 WHERE (A1.YR=1993)
04/12/200317:12:54Version: 5.0.1.062 Chimera (Mac OS X), 
License number: PB1JQ---, Process ID: 2227

the Result Action (Start_calcs) works, the Assign_cw_kwh Search 
action is where it is failing... but it doesn't seem to be related to 
the code, since I can substitute any other search action (including 
one right before Start_calcs that works) and it still fails at that 
location. In all instances, immediately after getting the 
Client/Plug-in Error the final item in the log is the last line, 
which re-iterates the License number.

Is there something other than maxactions which limits the number of 
actions that can be executed? My site is performing a building energy 
calculation (not the typical ecommerce transaction) and there are a 
lot of steps between user pages. This site is currently running on a 
beige G3 Mac running OS 8.6 and Tango 3.5.2... it doesn't seem like 
it would overload a dual processor 1.8 Ghz G5 running OS X and 
Witango 5 and since this problem seems independent of the actual 
code being executed, I'm not sure what to do.

Maggie

Customer Support wrote:

Maggie,
Set LOGGINGLEVEL=3 in the witango.ini file.  This will log the 
debug  like information to the witango log.  Stop and start the 
server with  the following terminal command:

sudo SystemStarter restart WITANGO5

It will ask for an admin password.  Once the command has finished 
you  will see some output like this:

Welcome to Macintosh.
Stopping Witango server
Waiting for WITANGO5
Starting Witango Application Server
[ 4106] 2003-12-05 13:49:58 START  INFO Starting Witango 
Application Server Process
Startup complete.
Hangup

This indicates that the server has restarted.

Now run the taf again and have a look in the witango.log file.  You 
should be able to see what the error is that is causing the server 
to  return an incorrect response to the web plugin.  What is the 
error  message?

Witango Support

On 05/12/2003, at 1:44 PM, MJPinckard wrote:

Actually when this occurs, I get no debugging, even though it is 
turned on... so no debugging is available to peruse.

witangoevents.log shows nothing after the last time the server 
started  up...

any other ideas on where to look?

Maggie

Jon van der Raadt wrote:

Do you have debug turned on -- sometimes that will help determine 
the   problem.  You can also look in the witangoevents.log It will 
record  a  crash if that is happening...

On Dec 4, 2003, at 7:26 PM, MJPinckard wrote:

I'm getting a Client/Plug-in error message and hope someone 
knows  what's going on. Here's the message.

  Client/Plug-in Error

  The response received from the application server is invalid.

It always happens at the same location in my application 
process,  but  doesn't seem to be tied to any specific code (e.g. 
if I remove  the  search action causing the error, the next 
search action fails,  if I  duplicate a earlier successful search 
action right before the  failing  action then it fails on the 
duplicated action).

It seems as if it is something like a maxaction limitation but 
I've   got maxaction set to 0, which if I understand correctly is 
unlimited   (I tested setting maxaction to 4000, way more than I 
need, but that   didn't change anything). The failure is 
happening quickly which  rules  out time out p

Re: Witango-Talk: Invalid Response

[MJPinckard]

No it doesn't crash... it is delighted to send me the same error 
repeatedly... and always fails at the same location in the process...

Maggie

Jon van der Raadt wrote:

Is the server crashed at this point or does it come back with the 
next  taf?  There are a few other options for running the server that 
give a  bit more information someone else will know them better...  I 
believe I  was using ./witangod -uo

Jon
On Dec 4, 2003, at 7:44 PM, MJPinckard wrote:
Actually when this occurs, I get no debugging, even though it is  
turned on... so no debugging is available to peruse.

witangoevents.log shows nothing after the last time the server 
started  up...

any other ideas on where to look?

Maggie

Jon van der Raadt wrote:

Do you have debug turned on -- sometimes that will help determine 
the   problem.  You can also look in the witangoevents.log It will 
record  a  crash if that is happening...

On Dec 4, 2003, at 7:26 PM, MJPinckard wrote:

I'm getting a Client/Plug-in error message and hope someone knows   
what's going on. Here's the message.

  Client/Plug-in Error

  The response received from the application server is invalid.

It always happens at the same location in my application process,  
but  doesn't seem to be tied to any specific code (e.g. if I 
remove  the  search action causing the error, the next search 
action fails,  if I  duplicate a earlier successful search action 
right before the  failing  action then it fails on the duplicated 
action).

It seems as if it is something like a maxaction limitation but 
I've   got maxaction set to 0, which if I understand correctly is 
unlimited   (I tested setting maxaction to 4000, way more than I 
need, but that   didn't change anything). The failure is happening 
quickly which  rules  out time out problems, and it is not deep in 
branch actions  which  rules that out.

If I look on the console, I see the SQL query being parsed, then 
a   line including system information and the License number.
Mac OX 10.3 running Witango 5.0.x (current version) under Apache. 
I   don't see a response from the database.

Anyone know what this is and how to fix or can steer me towards 
more   diagnostics?

Thanks,

Maggie

* * * * * * * * * * * * * * * * * * * *
* Margaret Pinckard
* [EMAIL PROTECTED]
* Principal Research Associate
* End-use Forecasting Group
* E.O. Lawrence Berkeley National Laboratory
_ 
__ _
TO UNSUBSCRIBE: Go to http://www.witango.com/maillist.taf


__ 
__
TO UNSUBSCRIBE: Go to http://www.witango.com/maillist.taf


--
* * * * * * * * * * * * * * * * * * * * * Margaret Pinckard
* [EMAIL PROTECTED]
* Principal Research Associate
* End-use Forecasting Group
* E.O. Lawrence Berkeley National Laboratory
___ 
_
TO UNSUBSCRIBE: Go to http://www.witango.com/maillist.taf


---
Jon van der Raadt
TeamOnly Network

412, Alastair Ross Technology Centre
3553 - 31 Street NW
Calgary, Alberta Canada  T2L 2K7
http://www.teamonly.com

   Office: (403) 286-5586
  Fax: (403) 286-5576
  Cell: (403) 863-6304
Email: [EMAIL PROTECTED]

TO UNSUBSCRIBE: Go to http://www.witango.com/maillist.taf


--
* * * * * * * * * * * * * * * * * * * * 
* Margaret Pinckard
* [EMAIL PROTECTED]
* Principal Research Associate
* End-use Forecasting Group
* E.O. Lawrence Berkeley National Laboratory


TO UNSUBSCRIBE: Go to http://www.witango.com/maillist.taf

Re: Witango-Talk: Invalid Response

[MJPinckard]

Here is a selection from the witango log...

04/12/200317:12:51127.0.0.1[EMAIL PROTECTED]
2549606415354[Results Action] Start_calcs
04/12/200317:12:51127.0.0.1[EMAIL PROTECTED]
2549606415354[Search Action] Assign_cw_kwh
04/12/200317:12:51127.0.0.1[EMAIL PROTECTED]
2549606415354[Query] SELECT A1.CWENRG FROM 
HESADM.APPLIANCES A1 WHERE (A1.YR=1993)
04/12/200317:12:54Version: 5.0.1.062 Chimera (Mac OS X), License 
number: PB1JQ---, Process ID: 2227

the Result Action (Start_calcs) works, the Assign_cw_kwh Search action 
is where it is failing... but it doesn't seem to be related to the code, 
since I can substitute any other search action (including one right 
before Start_calcs that works) and it still fails at that location. In 
all instances, immediately after getting the Client/Plug-in Error the 
final item in the log is the last line, which re-iterates the License 
number.

Is there something other than maxactions which limits the number of 
actions that can be executed? My site is performing a building energy 
calculation (not the typical ecommerce transaction) and there are a lot 
of steps between user pages. This site is currently running on a beige 
G3 Mac running OS 8.6 and Tango 3.5.2... it doesn't seem like it would 
overload a dual processor 1.8 Ghz G5 running OS X and Witango 5 and 
since this problem seems independent of the actual code being executed, 
I'm not sure what to do.

Maggie

Customer Support wrote:

Maggie,
Set LOGGINGLEVEL=3 in the witango.ini file.  This will log the debug  
like information to the witango log.  Stop and start the server with  
the following terminal command:

sudo SystemStarter restart WITANGO5

It will ask for an admin password.  Once the command has finished you  
will see some output like this:

Welcome to Macintosh.
Stopping Witango server
Waiting for WITANGO5
Starting Witango Application Server
[ 4106] 2003-12-05 13:49:58 START  INFO Starting Witango  
Application Server Process
Startup complete.
Hangup

This indicates that the server has restarted.

Now run the taf again and have a look in the witango.log file.  You  
should be able to see what the error is that is causing the server to  
return an incorrect response to the web plugin.  What is the error  
message?

Witango Support

On 05/12/2003, at 1:44 PM, MJPinckard wrote:

Actually when this occurs, I get no debugging, even though it is  
turned on... so no debugging is available to peruse.

witangoevents.log shows nothing after the last time the server 
started  up...

any other ideas on where to look?

Maggie

Jon van der Raadt wrote:

Do you have debug turned on -- sometimes that will help determine 
the   problem.  You can also look in the witangoevents.log It will 
record  a  crash if that is happening...

On Dec 4, 2003, at 7:26 PM, MJPinckard wrote:

I'm getting a Client/Plug-in error message and hope someone knows   
what's going on. Here's the message.

  Client/Plug-in Error

  The response received from the application server is invalid.

It always happens at the same location in my application process,  
but  doesn't seem to be tied to any specific code (e.g. if I 
remove  the  search action causing the error, the next search 
action fails,  if I  duplicate a earlier successful search action 
right before the  failing  action then it fails on the duplicated 
action).

It seems as if it is something like a maxaction limitation but 
I've   got maxaction set to 0, which if I understand correctly is 
unlimited   (I tested setting maxaction to 4000, way more than I 
need, but that   didn't change anything). The failure is happening 
quickly which  rules  out time out problems, and it is not deep in 
branch actions  which  rules that out.

If I look on the console, I see the SQL query being parsed, then 
a   line including system information and the License number.
Mac OX 10.3 running Witango 5.0.x (current version) under Apache. 
I   don't see a response from the database.

Anyone know what this is and how to fix or can steer me towards 
more   diagnostics?

Thanks,

Maggie

* * * * * * * * * * * * * * * * * * * *
* Margaret Pinckard
* [EMAIL PROTECTED]
* Principal Research Associate
* End-use Forecasting Group
* E.O. Lawrence Berkeley National Laboratory
_ 
__ _
TO UNSUBSCRIBE: Go to http://www.witango.com/maillist.taf


__ 
__
TO UNSUBSCRIBE: Go to http://www.witango.com/maillist.taf


--
* * * * * * * * * * * * * * * * * * * * * Margaret Pinckard
* [EMAIL PROTECTED]
* Principal Research Associate
* End-use Forecasting Group
* E.O. Lawrence Berkeley National Laboratory
___ 
_
TO UNSU

Re: Witango-Talk: Invalid Response

[MJPinckard]

Actually when this occurs, I get no debugging, even though it is turned 
on... so no debugging is available to peruse.

witangoevents.log shows nothing after the last time the server started up...

any other ideas on where to look?

Maggie

Jon van der Raadt wrote:

Do you have debug turned on -- sometimes that will help determine the  
problem.  You can also look in the witangoevents.log It will record a  
crash if that is happening...

On Dec 4, 2003, at 7:26 PM, MJPinckard wrote:

I'm getting a Client/Plug-in error message and hope someone knows  
what's going on. Here's the message.

  Client/Plug-in Error

  The response received from the application server is invalid.

It always happens at the same location in my application process, 
but  doesn't seem to be tied to any specific code (e.g. if I remove 
the  search action causing the error, the next search action fails, 
if I  duplicate a earlier successful search action right before the 
failing  action then it fails on the duplicated action).

It seems as if it is something like a maxaction limitation but I've  
got maxaction set to 0, which if I understand correctly is unlimited  
(I tested setting maxaction to 4000, way more than I need, but that  
didn't change anything). The failure is happening quickly which 
rules  out time out problems, and it is not deep in branch actions 
which  rules that out.

If I look on the console, I see the SQL query being parsed, then a  
line including system information and the License number.
Mac OX 10.3 running Witango 5.0.x (current version) under Apache. I  
don't see a response from the database.

Anyone know what this is and how to fix or can steer me towards more  
diagnostics?

Thanks,

Maggie

* * * * * * * * * * * * * * * * * * * *
* Margaret Pinckard
* [EMAIL PROTECTED]
* Principal Research Associate
* End-use Forecasting Group
* E.O. Lawrence Berkeley National Laboratory
___ 
_
TO UNSUBSCRIBE: Go to http://www.witango.com/maillist.taf



TO UNSUBSCRIBE: Go to http://www.witango.com/maillist.taf


--
* * * * * * * * * * * * * * * * * * * * 
* Margaret Pinckard
* [EMAIL PROTECTED]
* Principal Research Associate
* End-use Forecasting Group
* E.O. Lawrence Berkeley National Laboratory


TO UNSUBSCRIBE: Go to http://www.witango.com/maillist.taf

Witango-Talk: Invalid Response

[MJPinckard]

I'm getting a Client/Plug-in error message and hope someone knows what's 
going on. Here's the message.

  Client/Plug-in Error

  The response received from the application server is invalid.

It always happens at the same location in my application process, but 
doesn't seem to be tied to any specific code (e.g. if I remove the 
search action causing the error, the next search action fails, if I 
duplicate a earlier successful search action right before the failing 
action then it fails on the duplicated action).

It seems as if it is something like a maxaction limitation but I've got 
maxaction set to 0, which if I understand correctly is unlimited (I 
tested setting maxaction to 4000, way more than I need, but that didn't 
change anything). The failure is happening quickly which rules out time 
out problems, and it is not deep in branch actions which rules that out.

If I look on the console, I see the SQL query being parsed, then a line 
including system information and the License number.
Mac OX 10.3 running Witango 5.0.x (current version) under Apache. I 
don't see a response from the database.

Anyone know what this is and how to fix or can steer me towards more 
diagnostics?

Thanks,

Maggie

* * * * * * * * * * * * * * * * * * * *
* Margaret Pinckard
* [EMAIL PROTECTED]
* Principal Research Associate
* End-use Forecasting Group
* E.O. Lawrence Berkeley National Laboratory

TO UNSUBSCRIBE: Go to http://www.witango.com/maillist.taf

Witango-Talk: Connecting to Oracle with OS X

[MJPinckard]

I'm attempting to migrate my site from Tango 3.5.2 (isn't that scary) to 
Witango 5.

At this point, I'm trying to get the Dev. Studio to connect to my Oracle 
datasource, but am stuck. There was a discussion back in May about this 
problem, but I didn't see a resolution posted. I'd quote the discussion, 
but the archive isn't responding right now. By following all the 
recommendations in the discussions, I've managed to get the Dev. Studio 
(5.0.9) to know that Oracle exists, but can't open the datasource, it 
asks for username and password, but is unable to connect "verify that 
the datasource is properly configured and the database server is 
on-line". Obviously something is still wrong, but I don't know what to do.

My plan had been to use Oracle OCI to connect, but frankly I don't care 
how it connects, I just want it to connect. If anyone has a set of 
instructions (including directions on where to get any needed drivers) 
on how to connect to Oracle in any way (OCI, ODBC, JDBC) I really 
appreciate the help.

My new OS X server is due in any day, so I'd like to get this resolved.

Thanks,

Maggie

--
* * * * * * * * * * * * * * * * * * * * 
* Margaret Pinckard
* [EMAIL PROTECTED]
* Principal Research Associate
* End-use Forecasting Group
* E.O. Lawrence Berkeley National Laboratory


TO UNSUBSCRIBE: Go to http://www.witango.com/maillist.taf