Where you have a process that takes a long time (and this looks like
it might), you might consider a "bounce" in your app. We use this all
the time and with nary a problem... This way, you don't have to
change the client's timeout, and even if the time increases for the
analysis past a "usual" time, you're still OK.
e.g., we use a redirect and bounce through the same action over and
over again (with some useful message to the user each time) until the
task is completed. This keeps the browser active and the Witango
server happy. You can set something up so that the first XXX of
array1 get checked, then bounce to the action, grabbing the next XXX
of array1, etc. Does require using user scoped variables, but is
definitely worth it.
We found this most useful when dealing with Windows IE, which had
real (lack of) persistence issues in our testing, years ago.
And this is a prime example of why it would be really ***GREAT*** to
add the <@MINUS tag to <@UNION and <@INTERSECT because often we have
to do some sort of loop like this because we cannot compare arrays
and just get the rows that are missing....
On Mar 15, 2006, at 6:27 PM, Driscoll, Kevin wrote:
Brief overview
Selects distinct part number and vendors (about 25000 records but only
8000 distinct
Saves to an array
Loops through them and gets other data from the file
Save that to a second array
Then loops through those
To determine if any of the lots were rejected
Then save any that had no rejections to an array
Formats the report and displays it.
Kevin Driscoll
Quote of the "minute"
-----------------------------------
The Democrats are the party that says government will make you
smarter,
taller, richer, and remove the crabgrass on your lawn. The Republicans
are the party that says government doesn't work and then they get
elected and prove it.
P.J. O'Rourke
-----------------------------------
-----Original Message-----
From: Phil Wade [mailto:[EMAIL PROTECTED]
Sent: Wednesday, March 15, 2006 6:21 PM
To: [email protected]
Subject: Re: Witango-Talk: Timeout after 6 minutes
Kevin,
It is normally (99.999% of the time) not needed. Just add it to the
plug-ins stanza in the clients.ini file and restart the web server.
e.g.
CLIENTIOTIMEOUT=600
What task is the taf doing?
Regards
Phil
On 16/03/2006, at 10:15 AM, Driscoll, Kevin wrote:
Phil looked at the clients.ini and there is no setting in there at
all
for timeout??
Kevin Driscoll
Quote of the "minute"
-----------------------------------
It is my ambition to say in ten sentences what others say in a whole
book.
Friedrich Nietzsche
-----------------------------------
-----Original Message-----
From: Phil Wade [mailto:[EMAIL PROTECTED]
Sent: Wednesday, March 15, 2006 5:33 PM
To: [email protected]
Subject: Re: Witango-Talk: Timeout after 6 minutes
Kevin,
It sounds like the taf file is getting tied up or looping on the
server and never returning a response to the client.
The 5.5 client errors are quite granular now. The T2K/5.0 client
would simply report a 200 OK with error HTML. The 5.5 client will
now
change the http status code as well as displaying it to one of the
following so you can trap it at the web server. This is what they
mean.
STATUS_OK=200; //no error
STATUS_ERR_UNKNOWN=600; // the error is unknown
STATUS_ERR_INITENV=601; // Witango client environment cannot be
initialized
STATUS_ERR_CONNECT=602; // the client has failed to connect to the
application server
STATUS_ERR_REQUEST=603; // the client has failed to allocate an
application server request
STATUS_ERR_SEND=604; // the client has failed to send the
application server request
STATUS_ERR_RESPONSE=605; // the client has received an invalid
response from the application server
STATUS_ERR_HTTPHEADER=606; // the HTTP response received from the
application server is invalid
STATUS_ERR_LICENSE=607; // reserved
STATUS_ERR_INTERNAL=608; // internal client error
STATUS_ERR_ERRORHTML=609; // the error HTML file (ERROR_HTML) is not
found
STATUS_ERR_CLIENT=610; // the client cannot be allocated by the
API
STATUS_ERR_EXCEPTION=611; // an exception, such as invalid memory
access, etc, has occurred STATUS_ERR_DESTROYREQ=612; // Cannot
destroy
the request STATUS_ERR_DESTROYCON=613; // Cannot destroy the
connection
The 605 error occurs when the client does not receive a response from
the server in time. By using the push, the http header and some of
the result html is returned to the witango client so it has
received a
valid response.
How long does it take for the taf to complete? Check the
witangoevents.log file for entries relating to files timing out or
stack traces.
You could also try setting the CLIENTIOTIMEOUT in the clients.ini.
This controls how long the client will wait for a response from the
server. It is measured in seconds and defaults to 301 seconds, 1
second longer than the server timeout for a taf. If the taf is meant
to run for longer than 300 seconds you could bump CLIENTIOTIMEOUT out
to 600 seconds.
Regards
Phil
On 16/03/2006, at 12:59 AM, Driscoll, Kevin wrote:
Tango is completing the process but I am getting a client Plug-in
error, but if I do a push on the file it does not fail.
Does anyone know what the problem could be?
I looked at the timeout on IIS and it was set to 600 secs.
I know that is 10 minutes but I changed to to 1600 sec and same
result.
Client/Plug-in ErrorError Code: 605
Kevin Driscoll
_____________________________________________________________________
_
__
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
