Focal Point
[CASE-OPENED] EDA Code 35804 and BAD STATUS CODE (-12)

This topic can be found at:
https://forums.informationbuilders.com/eve/forums/a/tpc/f/7971057331/m/785102512

August 20, 2009, 09:18 AM
Pascal Bellerose
[CASE-OPENED] EDA Code 35804 and BAD STATUS CODE (-12)
Hi,

We have several alert reports launched by Report Caster around 2:00AM.

We also have some other alert reports that are launched by Report Caster every 15 minutes.

Not every nights but at least once a week we get errors between 1:00AM and 4:00AM stating an EDA Code 35804 and a BAD STATUS CODE (-12) received from a remote adapter.

The remote adapter is connected to a Full Function Server (FFS). Both the WFS and the FFS were up and running at the moment we received the error.

We suspect a problem with agents availability on the FFS side. But we don't receive any "No resource for default" notification, as we would normally when there is no more agents available.

Is there a way to collect agents usage on a server an store this statistic for trend analysis?

Thanks

This message has been edited. Last edited by: Kerry,


WFS/FFS/DM 7.6.4 / WFS/FFS/DM 7.6.8 / WFS/FFS/DM 7.6.9 / AS400 V5R4M0 / HTML / iSM 5.5sp2
August 20, 2009, 09:31 AM
GinnyJakes
A couple of things. Statistics are available from the WebFOCUS console, Workpace/Monitor/Statistics.

You could also try scrubbing EDAPRINT for information. This file is stored in /WebFOCUS/ibi/srv76/wfs/bin.

Generally a -12 indicates that there was too much data coming back from the remote server and there was a timeout. I think you would get another error if you were out of agents. If you think that is the case, you could increase your max agents a bit.


Ginny
---------------------------------
Prod: WF 7.7.01 Dev: WF 7.6.9-11
Admin, MRE,self-service; adapters: Teradata, DB2, Oracle, SQL Server, Essbase, ESRI, FlexEnable, Google
August 20, 2009, 11:05 AM
Pascal Bellerose
Statistics are no use because I can't figure out what was the number of agents running at the time the error occured.

As for the timeout issue, the fex returns almost the same number of records each time it is run (every 15 seconds). Normally it runs without a flaw, but sometimes we get the error and not necessarily always for the same report. it could be either one of the alert reports that runs once a day around 2:00AM or it could be one of the alert reports that runs every 15 minutes.

To be more precise, we usually get this kind of error from the "once a day" alert reports. Today is the first time we get it from an alert report that runs every 15 minutes.

Thanks,


WFS/FFS/DM 7.6.4 / WFS/FFS/DM 7.6.8 / WFS/FFS/DM 7.6.9 / AS400 V5R4M0 / HTML / iSM 5.5sp2
August 20, 2009, 12:19 PM
GinnyJakes
You can get agent usage out of EDAPRINT.


Ginny
---------------------------------
Prod: WF 7.7.01 Dev: WF 7.6.9-11
Admin, MRE,self-service; adapters: Teradata, DB2, Oracle, SQL Server, Essbase, ESRI, FlexEnable, Google
August 20, 2009, 12:29 PM
Pascal Bellerose
Ginny,

I guess this would need programming a tool that could analyze EDAPRINT and create an history of agents usage.

Other than that, any ideas on how I could have this kind of information? Resource Management for example?

Thanks


WFS/FFS/DM 7.6.4 / WFS/FFS/DM 7.6.8 / WFS/FFS/DM 7.6.9 / AS400 V5R4M0 / HTML / iSM 5.5sp2
August 20, 2009, 02:28 PM
GinnyJakes
With WebFOCUS you can write your own tool to analyze edaprint. Here is a sample of the contents.
quote:

08/19/2009 21:28:33 request by cmrpip021868 for explicit connect to agent (DEFAULT)
08/19/2009 21:28:33 connecting cmrpip021868 tscomid=259,sesid=21700
08/19/2009 21:28:33 processing cmrpip021868 u=xxxxxxxx,g=webfocus,l=USR
08/19/2009 21:28:34 request by t3rp1179658 to notify disconnect of sesid=21700
08/19/2009 21:28:34 disconnect cmrpip021868 tscomid=259,sesid=21700
08/19/2009 21:28:48 accepting cmrpip021869 tcp=139.51.76.101:60167
08/19/2009 21:28:48 request by cmrpip021869 for explicit connect to agent (DEFAULT)
08/19/2009 21:28:48 connecting cmrpip021869 tscomid=259,sesid=21701
08/19/2009 21:28:48 processing cmrpip021869 u=xxxxxxxx,g=webfocus,l=USR
08/19/2009 21:28:48 accepting FDS connection tcp=127.0.0.1:60168


Using a custom master and some substringing, etc. you can determine how long an agent was active per session etc. There are also some reports available from the web console but they don't seem to relate to the tscomid.


Ginny
---------------------------------
Prod: WF 7.7.01 Dev: WF 7.6.9-11
Admin, MRE,self-service; adapters: Teradata, DB2, Oracle, SQL Server, Essbase, ESRI, FlexEnable, Google
August 25, 2009, 12:32 PM
Pascal Bellerose
GinnyJakes,

Do you have this kind of report? If so could you post the code so I can submit this to our developers?

Thanks,

Thanks,


WFS/FFS/DM 7.6.4 / WFS/FFS/DM 7.6.8 / WFS/FFS/DM 7.6.9 / AS400 V5R4M0 / HTML / iSM 5.5sp2
August 25, 2009, 12:34 PM
GinnyJakes
No, I do not. Sorry.


Ginny
---------------------------------
Prod: WF 7.7.01 Dev: WF 7.6.9-11
Admin, MRE,self-service; adapters: Teradata, DB2, Oracle, SQL Server, Essbase, ESRI, FlexEnable, Google
August 25, 2009, 12:46 PM
Pascal Bellerose
I have an update on the case though:

This morning I received "BAD STATUS CODE (-33)" with a EDA Code 35804.

The only info I found on the error is this case: BAD STATUS CODE (-33)
But it relates to a "LOCAL" adapter.

In my case, it is a remote adapter and the remote server kept sending "NO RESOURCE FOR DEFAULT" notifications (600 emails exactly from 2:10AM to 2:47AM)

It is the same reports that last week were crashing with a BAD STATUS CODE (-12)

Any ideas?


WFS/FFS/DM 7.6.4 / WFS/FFS/DM 7.6.8 / WFS/FFS/DM 7.6.9 / AS400 V5R4M0 / HTML / iSM 5.5sp2
August 25, 2009, 02:05 PM
GinnyJakes
-33 User resource limit reached.

You are going to have to log onto the web console for the remote server and see what is going on. Look at agent settings and max settings. Try running your program directly from the web console.


Ginny
---------------------------------
Prod: WF 7.7.01 Dev: WF 7.6.9-11
Admin, MRE,self-service; adapters: Teradata, DB2, Oracle, SQL Server, Essbase, ESRI, FlexEnable, Google