Retry delay within an Exception path

Do you know if there is a way to set a retry-delay within an exception path. I tried using a timer control, but it doesn't seem to work. I noticed that a retry-count and retry-delay can be set for the overall process. Is there a way to determine which retry the process is executing. For example I want to send an email to production support, but only on the final retry and the retries need to be delayed. Any ideas?

if (this.context.getExceptionInfo().getRetriesRemaining() == 0 )

Similar Messages

Help with frozen process, retry delay and transaction rolled back

Hi all,


Is there a way to create a JPD that will be able to retry a specific logic for several times with delay in between and freeze on the last retry. Also, at the same time, not rollback any database commits?


I have tried several ways to approach this but did not work.


1) create a JPD with transaction block with start node = "freeze on failure". Set retry count and delay in the transaction block. The outcome is, the JPD freezes, retries correctly, unfreezed properly, but all DB transactions are rolled back


2) create a JPD without transaction block, and set start node with "freeze on failure", group the nodes that I would like to perform retry and in the exception path add in the retry count. Use timer in the exception path to introduct the delay. The outcome of this is: JPD freezes, retries correctly, unfreezed INCORRECTLY (when trying to unfreeze this process, it will start at the ontimeout method of the timer and does not start from the beginning of the jpd), DB transactions are commited correctly


Any help or suggestions would be much appreciated...

Thanks!


Carol

The issue may be due to the transaction timeout, verify the configured timeouts and the processing time of the process B.
Try increasing Sync Max Time Out and the other timeouts accordingly and test it.
Refer the below URL for the details of configuring the timeouts.
http://www.albinsblog.com/2011/11/oracle-soa-11g-configure-transaction_20.html
Regards
Albin I

Max delay and Min delay on same data path

Hello,
I am trying to set up timing constraints (both max and min delay) on a data path between two combinatorial elements.
I use set_max_delay -from [get_pins pin name] -to [get_pins pin name] 6 -datapath_only
set_min_delay -from [get_pins pin name] -to [get_pins pin name] 4
I checked the timing report and I see that the max_delay constraint is satisfied but the min_delay is not.What can be the reason behind this?
Also, when i comment out the first line (max_delay constraint), I see that min_delay constraint is satisfied.
Can't a datapath have both max and min delay constraints?
Any help will be really appreciated.
Thanks!
-Kamath

Be very careful about trying to set max/min delays on a portion of a path (i.e. not from a valid startpoint to a valid endpoint). Doing so is called path segmentation. When you do so, the max/min delay gets applied to the portion of the path you specified, but the rest of the path (between the startpoint and the set_max_delay -from and from the set_max_delay -to to the endpoint) are no longer constrained at all.
There are extremely few reasons to want to do this. I second the question "Why do you want to do this?"
That being said, you can set both a max and min delay on the same path (or path segment, if you insist on segmenting the path). The set_max_delay overrides the setup requirement, and the set_min_delay overrides the hold requirement. If the two values are too close together the tools will not be able to satisfy both; all timing checks are done at both process corners (max PVT and min PVT) - the ratio from max to min is generally around 3:1, so if the set_max_delay is less than 3x the set_min_delay the tools will be unable to satisfy both (due to real, physical limitations).
Avrum

Process transaction retry count retry delay ERROR

Hi
I've an asynchronous process (listening to a channel). The subscription is synchronous (but i also try it whit an asynchronous subscription).
The matter is:
I've a transaction in the process so i can set the retry count to 23 and the retry delay to 3600 sec.
But when i test it i've got a "dramatic" problem. It only retries 6 times and the time between each retry is zero, it automatically retries.
i don`t know what can i do ... some help ?
zanks in advance
Taja

Hi again
the problem is solved.
Recently we have migrated de DB to a new machine. We exported ALL, data, tables, everithing.
All we do to solve the problem is delete the distributed async queues and create them again.
Redeploy and woala ...
it works again, 23 retries with an interval of 3600 sec
i hope this will be useful for someone.
regards
Taja

The name ("common name") of a valid code-signing certificate in a keychain within your keychain path. A missing or invalid certificate will cause a build error. [CODE_SIGN_IDENTITY]

The name ("common name") of a valid code-signing certificate in a keychain within your keychain path. A missing or invalid certificate will cause a build error. [CODE_SIGN_IDENTITY]

If you could ask a coherent question, maybe...
Perhaps you should be posting in the developers forums...

Retry whole procedure on exception

Hi everyone,
I have a process which calls several OBPM procedures.
Within these OBPM procedures there are several automatics which translate bpm variables (mostly strings / group of strings) to introspected xml variables and the response back to a bpm variable. Sometimes the WS callout won't give the desired result (technical OK, functional NOK; e.g.: customer not found) and I'm throwing an funtional exception when this is the case.
The problem that I'm having right now, is that when I set the correct BPM variables (e.g. customer number, which is the procedure's input) the exception is sent back to the automatic where the exception occurred and will still fail because the WS response is still incorrect.
I would like to retry the whole procedure and not just the automatic within the procedure where the exception was thrown.
I've tried Action = repeat (which saved the instance variables in the screenflow) / retry (throws back to the automatic where the exception occurred) but I'm a bit stuck right now.
How could I solve this?
Thanks in advance,
Joris.

To anyone having the same problem;
I didn't get this to work as intended.
A workaround for me is to create an in- and outgoing instance variable which contains a success/failure status and to have an conditional transition (procresult != "success") from the procedure which would lead to functional exception handling.
The exception handling screenflow would be able to abort and retry so the only unconditional transition is going back to the failed procedure.

LP and Plogue Bidule: Delay in the MIDi path. Solutions?

I'm trying to use Plogue Bidule with LP in rewire mode to getting around the RAM limit of Logic.
Now, all would seem to work fine except for a MIDI issue I'm encountering and I can't solve:
When I hit my physical keyboard connected to Logic, having to command the instrument (connected in Rewire mode) in PBidule, the MIDI command arrives to Bidule various milliseconds after my action, making it impossible to play without getting crazy...
At first I thought there was an audio latency, but after, seeing the Kontakt 3 GUI keyboard, I noticed that my hammer actions were transfered to K3 GUI keyboard in delay...
Has anyone encountered this issue? (and has a solution?)
Tnx

Yeahh!!!!
After a week when nobody on Plogue forum can tell me anytyhing minimum useful..
I posted my issue on Logic forum and ... voilà: problem solved.
Thank you very much DTS! You saved me!

Delays in "BEGIN-EXCEPTION"??

Do they cause delays the use of "BEGIN-EXCEPTION" in procedures or functions of ORACLE?
Is true???

If there is a difference, then it is hardly measurable. Consider this 1 Mill loops example
SQL*Plus: Release 9.2.0.2.0 - Production on Thu Mar 6 14:05:46 2008
Copyright (c) 1982, 2002, Oracle Corporation. All rights reserved.
Connected to:
Oracle9i Enterprise Edition Release 9.2.0.8.0 - 64bit Production
With the Partitioning, OLAP and Oracle Data Mining options
JServer Release 9.2.0.8.0 - Production
SQL> set time on
14:07:23 SQL> set timing on
14:07:33 SQL>
14:07:34 SQL>
14:07:46 SQL> begin
14:07:46   2    for i in 1..1000000 loop
14:07:46   3      null;
14:07:46   4    end loop;
14:07:46   5 end;
14:07:46   6 /
PL/SQL procedure successfully completed.
Elapsed: 00:00:00.03
14:07:47 SQL>
14:07:47 SQL> begin
14:07:47   2    for i in 1..1000000 loop
14:07:47   3      begin
14:07:47   4      begin begin begin null; end; end; end;
14:07:47   5   exception
14:07:47   6      when no_data_found then null;
14:07:47   7   end;
14:07:47   8    end loop;
14:07:47   9 end;
14:07:47 10 /
PL/SQL procedure successfully completed.
Elapsed: 00:00:00.03
14:07:47 SQL>
14:07:48 SQL> As you can see I was not able to measure how much longer 4 Mill begin end block including 1 million exception blocks compared to 1 begin end block are.

Process Flow - delay within flow during execution

Hi
I have two process flows e.g. PF1 and PF1_1. PF1_1 is executed as a subprocess within PF1.
On execution of PF1 the scheduled job runs the mappings up to the end of PF1_1 and then pauses for approx. 1 to 2 minutes for (what appears to be) no apparent reason. The mappings do not have analyze set and looking on the DB there is no activity related to the job.
Can anyone help on the reason for the delay and how to avoid it?
I've done a bit of digging on Scheduled Jobs and I'm wondering if it's something to do with the PRIOIRTY the job is given? At the moment the job has a priority of 3 or Medium in OEM i.e. the defaults.
Any help much appreciated.
Cheers
Si

Hi Oleg
PF1 and PF1_1 were just for my example, the actuals are PF_SS_CLEANSE_1 and PF_SS_AGENT.
Added Instance_Label and ordered by End Date Desc
BEGIN_DATE     END_DATE     ITEM_KEY     ITEM_TYPE     PROCESS_NAME     ACTIVITY_NAME     INSTANCE_LABEL     ACTIVITY_ITEM_TYPE
23/12/2009 14:14:34     23/12/2009 14:14:34     WB_IK_20091223_141001_6227307     LOAD_AD     PF_SS_AGENT     END     END_SUCCESS     OWB_10G
23/12/2009 14:13:41     23/12/2009 14:14:34     WB_IK_20091223_141001_6227307     LOAD_AD     PF_SS_AGENT     A7A5D54FC0CECD7F5E04013AC2D287     PF_SS_EMAIL_SUCESS     LOAD_AD
23/12/2009 14:10:01     23/12/2009 14:14:34     WB_IK_20091223_141001_6227307     LOAD_AD     ROOT     PF_SS_AGENT     PF_SS_AGENT     LOAD_AD
23/12/2009 14:13:40     23/12/2009 14:13:41     WB_IK_20091223_141001_6227307     LOAD_AD     PF_SS_AGENT     A63D3B8BE1271F097E04013AC2D284     APPEND_DB_NAME_TO_TEXT_SUCCESS     LOAD_AD
23/12/2009 14:13:39     23/12/2009 14:13:40     WB_IK_20091223_141001_6227307     LOAD_AD     PF_SS_AGENT     A5A74F8CAE5156A24E04013AC2D287     GATHER_TABLE_GROUP_STATS     LOAD_AD
23/12/2009 14:13:15     23/12/2009 14:13:39     WB_IK_20091223_141001_6227307     LOAD_AD     PF_SS_AGENT     AND     AND_SUCCESS     OWB_10G
23/12/2009 14:13:14     23/12/2009 14:13:15     WB_IK_20091223_141001_6227307     LOAD_AD     PF_SS_AGENT     A74628F90794268C4E04013AC2D283     UPD_D_AGENT_CURR_ROW     LOAD_AD
23/12/2009 14:13:12     23/12/2009 14:13:14     WB_IK_20091223_141001_6227307     LOAD_AD     PF_SS_AGENT     A74628F8FC6BD68C4E04013AC2D283     UPD_D_AGENT     LOAD_AD
23/12/2009 14:12:49     23/12/2009 14:13:12     WB_IK_20091223_141001_6227307     LOAD_AD     PF_SS_AGENT     OR     OR1_2     OWB_10G
23/12/2009 14:12:44     23/12/2009 14:12:49     WB_IK_20091223_141001_6227307     LOAD_AD     PF_SS_AGENT     A7530A4866A5CFBFCE04013AC2D285     LOAD_D_AGENT     LOAD_AD
23/12/2009 14:12:20     23/12/2009 14:12:44     WB_IK_20091223_141001_6227307     LOAD_AD     PF_SS_AGENT     SWITCH     ROUTE1     OWB_10G
23/12/2009 14:12:20     23/12/2009 14:12:20     WB_IK_20091223_141001_6227307     LOAD_AD     PF_SS_AGENT     ROUTE     ROUTE     OWB_10G
23/12/2009 14:11:00     23/12/2009 14:12:20     WB_IK_20091223_141001_6227307     LOAD_AD     PF_SS_AGENT     P02_PF_SS_CLEANSE_1     PF_SS_CLEANSE_1     LOAD_AD
23/12/2009 14:12:19     23/12/2009 14:12:19     WB_IK_20091223_141101_6227379     LOAD_AD     PF_SS_CLEANSE_1     END     END_SUCCESS     OWB_10G
23/12/2009 14:11:07     23/12/2009 14:12:19     WB_IK_20091223_141101_6227379     LOAD_AD     PF_SS_CLEANSE_1     AND     AND_SUCCESS     OWB_10G
23/12/2009 14:11:01     23/12/2009 14:12:19     WB_IK_20091223_141101_6227379     LOAD_AD     ROOT     PF_SS_CLEANSE_1     PF_SS_CLEANSE_1     LOAD_AD
23/12/2009 14:11:06     23/12/2009 14:11:07     WB_IK_20091223_141101_6227379     LOAD_AD     PF_SS_CLEANSE_1     A5B90750E3C6A81B2E04013AC2D287     CLEAN_TEAM     LOAD_AD
23/12/2009 14:11:05     23/12/2009 14:11:07     WB_IK_20091223_141101_6227379     LOAD_AD     PF_SS_CLEANSE_1     A5B9139CD54AF3314E04013AC2D280     CLEAN_TEAM_GROUP     LOAD_AD
23/12/2009 14:11:05     23/12/2009 14:11:05     WB_IK_20091223_141101_6227379     LOAD_AD     PF_SS_CLEANSE_1     FORK     FORK     OWB_10G
23/12/2009 14:11:04     23/12/2009 14:11:05     WB_IK_20091223_141101_6227379     LOAD_AD     PF_SS_CLEANSE_1     A5B9139CD5C4C3314E04013AC2D280     CLEAN_AGENT     LOAD_AD
23/12/2009 14:11:04     23/12/2009 14:11:05     WB_IK_20091223_141101_6227379     LOAD_AD     PF_SS_CLEANSE_1     A74B4104CC78F95CBE04013AC2D281     CLEAN_TEAM_GROUP_HISTORY     LOAD_AD
23/12/2009 14:11:04     23/12/2009 14:11:04     WB_IK_20091223_141101_6227379     LOAD_AD     PF_SS_CLEANSE_1     FORK     FORK_1     OWB_10G
23/12/2009 14:11:03     23/12/2009 14:11:04     WB_IK_20091223_141101_6227379     LOAD_AD     PF_SS_CLEANSE_1     A7465C6DA7FE4CB89E04013AC2D285     CLEAN_AGENT_TEAM_HISTORY     LOAD_AD
23/12/2009 14:11:01     23/12/2009 14:11:03     WB_IK_20091223_141101_6227379     LOAD_AD     PF_SS_CLEANSE_1     A74A0BC144826AECCE04013AC2D280     CLEAN_AGENT_TEAM_HISTORY_DATES     LOAD_AD
23/12/2009 14:11:01     23/12/2009 14:11:01     WB_IK_20091223_141101_6227379     LOAD_AD     PF_SS_CLEANSE_1     START     START1     OWB_10G
23/12/2009 14:10:59     23/12/2009 14:11:00     WB_IK_20091223_141001_6227307     LOAD_AD     PF_SS_AGENT     A5673591317F622DCE04013AC2D287     GET_EXECUTION_ID     LOAD_AD
23/12/2009 14:10:03     23/12/2009 14:10:59     WB_IK_20091223_141001_6227307     LOAD_AD     PF_SS_AGENT     A7A5D54FC0CECD7F5E04013AC2D287     PF_SS_EMAIL_START     LOAD_AD
23/12/2009 14:10:02     23/12/2009 14:10:03     WB_IK_20091223_141001_6227307     LOAD_AD     PF_SS_AGENT     A63D3B8BE1271F097E04013AC2D284     APPEND_DB_NAME_TO_TEXT     LOAD_AD
23/12/2009 14:10:01     23/12/2009 14:10:02     WB_IK_20091223_141001_6227307     LOAD_AD     PF_SS_AGENT     A63D3B8BE125BF097E04013AC2D284     GET_DATABASE_NAME     LOAD_AD
23/12/2009 14:10:01     23/12/2009 14:10:01     WB_IK_20091223_141001_6227307     LOAD_AD     PF_SS_AGENT     START     START1     OWB_10GThis is a mini schedule that I use to maintain a single dimension. The main schedule to load multiple dimensions and facts has even larger delays.
Cheers
Si
Edited by: ScoobySi on Dec 24, 2009 9:39 AM

How to move up within a URL path?

Recently started using URLs in my programs. Works great. However, all the
time I have been wondering how to move up in that URL path. For example,
if my url1=("http://www.something.com/") I normally end up at the index.html.
But how could one set up a relative url reaching a file or directory at the same
level as that index.html? I suppose one should go up onel level and then
specify the name of the required file.
I tried to use things like url2= new URL(url1,"../myfile") but that doesn't seem
to work.
Anybody has an idea? Thanks in advance.
Dirk

martin@work wrote:
Hi,
you can't, if your web server provides the minimum of security. In the web server configuration you normally map a base url to a physical directory on the server and the web server does its very best to assure that you can only access files in that physical directory and below, and that all other files on the server in any upper or parallel directories are not accessible.
If you want to access files in a parallel directory then you have to create a further mapping in the web server which maps another base URL to that directory.
Martin To reinforce what Martin says: if you cannot do this with your web browser, don't expect to do it with Java.
On the other hand, if you can do it in your web browser, we misunderstood your question and please
reformulate it using actual URL's (or at least the portion following the host name).
Edited by: baftos on Apr 25, 2008 9:28 AM

How to set FileSystemWatcher powershell script to exclude folders within the root path being monitored?

All,
I want to use a pre-fab script to monitor the OS folders on a Windows 2008 R2 Domain Controller and send any changes into the Event Log. But there are two folders in the file path I want to exclude because changes to them would fill up the Event Logs
too quickly. Thus, I want to monitor all folder in C:\Windows but want to exclude C:\Windows\Temp and C:\Windows\debug.
The script I want to use is below.
#This script uses the .NET FileSystemWatcher class to monitor file events in folder(s).
#The advantage of this method over using WMI eventing is that this can monitor sub-folders.
#The -Action parameter can contain any valid Powershell commands.
I have just included two for example.
#The script can be set to a wildcard filter, and IncludeSubdirectories can be changed to $true.
#You need not subscribe to all three types of event. All three are shown for example.
# Version 1.1
$folder = 'C:\Windows' # Enter the root path you want to monitor.
$filter = '*.*' # You can enter a wildcard filter here.
# In the following line, you can change 'IncludeSubdirectories to $true if required.
$fsw = New-Object IO.FileSystemWatcher $folder, $filter -Property @{IncludeSubdirectories = $true;NotifyFilter = [IO.NotifyFilters]'FileName, LastWrite'}
# Here, all three events are registerd. You need only subscribe to events that you need:
Register-ObjectEvent $fsw Created -SourceIdentifier FileCreated -Action {
$name = $Event.SourceEventArgs.Name
$changeType = $Event.SourceEventArgs.ChangeType
$timeStamp = $Event.TimeGenerated
Write-Host "The file '$name' was $changeType at $timeStamp" -fore green
Out-File -FilePath c:\scripts\filechange\outlog.txt -Append -InputObject "The file '$name' was $changeType at $timeStamp"}
Register-ObjectEvent $fsw Deleted -SourceIdentifier FileDeleted -Action {
$name = $Event.SourceEventArgs.Name
$changeType = $Event.SourceEventArgs.ChangeType
$timeStamp = $Event.TimeGenerated
Write-Host "The file '$name' was $changeType at $timeStamp" -fore red
Out-File -FilePath c:\scripts\filechange\outlog.txt -Append -InputObject "The file '$name' was $changeType at $timeStamp"}
Register-ObjectEvent $fsw Changed -SourceIdentifier FileChanged -Action {
$name = $Event.SourceEventArgs.Name
$changeType = $Event.SourceEventArgs.ChangeType
$timeStamp = $Event.TimeGenerated
Write-Host "The file '$name' was $changeType at $timeStamp" -fore white
Out-File -FilePath c:\scripts\filechange\outlog.txt -Append -InputObject "The file '$name' was $changeType at $timeStamp"}
# To stop the monitoring, run the following commands:
# Unregister-Event FileDeleted
# Unregister-Event FileCreated
# Unregister-Event FileChanged
Is there a command line I can use to exclude thos two folders?
Thanks

First, thank you to everyone who has been helping me. I've gotten a little bit further in what I need to do. I still have two problems that I really need help in sorting out:
1. The script will run on Windows 7 with no errors but will not run on a Server 2008 R2 Domain Controller. On the DC, I get an error for the SourceIdentifier: "A parameter cannot be found that mataches parameter name 'SourceIdentifier'."
2. Even though the script runs on Windows 7 with no errors, I tested it by making changes to the Temp and debug folders and the changes were still being recorded in the logs.
A copy of the updated script I am running is below:
#This script uses the .NET FileSystemWatcher class to monitor file events in folder(s).
#The advantage of this method over using WMI eventing is that this can monitor sub-folders.
#The -Action parameter can contain any valid Powershell commands. I have just included two for example.
#The script can be set to a wildcard filter, and IncludeSubdirectories can be changed to $true.
#You need not subscribe to all three types of event. All three are shown for example.
# Version 1.1
$folder = 'C:\Windows' # Enter the root path you want to monitor.
$filter = '*.*' # You can enter a wildcard filter here.
# In the following line, you can change 'IncludeSubdirectories to $true if required.
$fsw = New-Object IO.FileSystemWatcher $folder, $filter -Property @{IncludeSubdirectories = $true;NotifyFilter = [IO.NotifyFilters]'FileName, LastWrite'}
# Here, all three events are registerd. You need only subscribe to events that you need:
$Script:excludedItems = 'C:\Windows\Temp','C:\Windows\debug'
Register-ObjectEvent $fsw Created -SourceIdentifier FileCreated -Action {
If ($excludedItems -notcontains $Event.SourceEventArgs.Name) {
$name = $Event.SourceEventArgs.Name
$changeType = $Event.SourceEventArgs.ChangeType
$timeStamp = $Event.TimeGenerated
Write-Host "The file '$name' was $changeType at $timeStamp" -fore green
Out-File -FilePath c:\scripts\filechange\outlog.txt -Append -InputObject "The file '$name' was $changeType at $timeStamp"} }
Register-ObjectEvent $fsw Deleted -SourceIdentifier FileDeleted -Action {
If ($excludedItems -notcontains $Event.SourceEventArgs.Name) {
$name = $Event.SourceEventArgs.Name
$changeType = $Event.SourceEventArgs.ChangeType
$timeStamp = $Event.TimeGenerated
Write-Host "The file '$name' was $changeType at $timeStamp" -fore red
Out-File -FilePath c:\scripts\filechange\outlog.txt -Append -InputObject "The file '$name' was $changeType at $timeStamp"} }
Register-ObjectEvent $fsw Changed -SourceIdentifier FileChanged -Action {
If ($excludedItems -notcontains $Event.SourceEventArgs.Name) {
$name = $Event.SourceEventArgs.Name
$changeType = $Event.SourceEventArgs.ChangeType
$timeStamp = $Event.TimeGenerated
Write-Host "The file '$name' was $changeType at $timeStamp" -fore white
Out-File -FilePath c:\scripts\filechange\outlog.txt -Append -InputObject "The file '$name' was $changeType at $timeStamp"} }
# To stop the monitoring, run the following commands:
# Unregister-Event FileDeleted
# Unregister-Event FileCreated
# Unregister-Event FileChanged
Thanks in advance for any help you can give.

Rotating an image within a clipping path in Photoshop CS5

When you have taken a photo of an image, then open it up in Photoshop and realise its on a bit of an angle but you make a clipping path around it first and save it, then afterwards want to rotate it - Is there a way of linking the path with the image layer so they rotate together in one step?,
Or can you only rotate the image layer first then the path separately as thats what I ended up doing?
Hoping someone can help me and answer my question

If your clipping path is stored in the Paths panel, then no. But you can select the path then go to Layer > Vector Mask > Current Path. Not transform the layer and the path is transformed, too. You can then open the Paths panel and drag the vector path onto the New Path button to duplicate it and make it a clipping path for the entire image.

Delay within internal loops - nested loops

Hi all,
I have generated report using nested loops.if i give some 200 hits ,it generates the report within 90 seconds.but if the hit limit increases above 500,the loop becomes infinite.i have checked the datas as well as select Queries,everything is fine.why it happens.please help me if u know the solution.
thanks in advance,
raj

hi
There are couple of solutions to this
a) try defining a sarted internal table for the inside table with the keys that you define in the where clause of your inside loop.
b) You can try the below code
loop at itab1.
read table itab2 with key .......
l_tabix = sy-tabix.
loop at itab2 from l_tabix.
<check the validity condition>
endloop.
this could kill about 3/4 of the time before
Hope this Helps
Anirban

Steps within Exception branch / Handler

I'm trying to send an Alert message or close a S/A bridge within an Exception branch (handler) and none of these seems to works.
Do any steps work at all in exception branch. Can we do some clean up work before the exception handler does its thing.
A bit more on what I'm trying to do.
1. Open S/A bridge via HTTP
2. Transform source to target message of RFC type
3. Make an RFC call in a branch with Exception handler and my exception is ABAP_Abend. We can call over 100 different ABAP programs via the RFC FM based on the data type and some programs Abend in SAP/target system
4.If RFC is successful, map response to HTTP response and this works fine
5. If RFC fails due to ABAP_Abend, I would like to close S/A bridge which waits indefinitely otherwise and send an alert message (via steps in exception handler). But this does not seem to work.
Any ideas on how to take care of some clean up work before the system does the exception handling.
Thank you for your responses.
Parimala

Hi,
We use the S/A bridge and exception branch to trigger alerts upon error during synchronous calls. Our process gets terminated and the alerts are thrown upon exception. But we do not use RFC - we make http calls to third-party systems.
I am not familiar with the ABAP-Abend. In case of Abend, do you still get a response back from the RFC? If you still get a response from RFC, you need to inspect the response (switch step is one way of doing it) and raise the exception to transfer the control to the exception branch.
Regards,
Suresh.

Find out who catches exception from within exception

Hi,
I am writing some generic Exception for our project. The aim is to simplify exception handling by making the exceptions log the necessary information with Log4J.
It seems pretty easy to find out which class actually threw the exception (top of stack trace) but I would much rather like to log the exception in the name of the class that will catch the exception.
The problem is that the stacktrace (it can be retrieved within an exception as of J2SE 1.4) does not contain information about which class will actually be catching the exception.
Is there any way to accomplish that? Would be very grateful if someone could point me into the right direction if there is a possibility.
Thank you very much in advance
Greets Robbi

Thanks for the quick reply.
What I actually really wanted is to log from within the Exception. Right now I am doing this from within the Exceptions constructor.
I see your point with the constructor being supposed to be lightweight, but in most cases the logging will be turned off and log4J does not take a lot of time to find that out.
The logging within the catch blocks is what we wanted to circumvent as that would be the same in all catch blocks. Additionally it would be easy to simply change the topmost Exception in our hierarchy to log somewhere else instead.
Performance is not too crucial on our application, but traceability is, as a lot of money is tied to a single transaction. That is why we decided to for Exceptions to log, as thus even when a developer would forget to log, the exception is still logged.
Anyway, I thought that it is quite likely not possible to determine who catches an exception from within that exception. I guess the virtual machine wouldn't like to unseal that information. Maybe this is something that might work in a future Java version.
Thanks for the input anyway.
Regards Robbi

Retry delay within an Exception path

Similar Messages

Maybe you are looking for