DI Job Execution Hangs
It took a while to build our DI Linux Job Server, so we initially installed a local DI Job Server on each of the laptops in Windows.
We're still having to work out some firewall rules as well-
Here is the DI issue. We design a job in the Designer, and can execute it on the local job server (on the laptop) and it runs fine and to completion.
Becuase of the firewall rules, we can't execute a job from within the Designer, but rather, we execute it from the DI Web Admin Console.
When we execute the same job on the Linux Job Server, it appears to be running, but the trace log reads this:
(11.7) 01-07-09 10:51:55 (21168:3086001856) JOB: The initial environment locale <eng_us.utf-8> has been coerced to <Unicode (UTF-16)> ().
(11.7) 01-07-09 10:51:55 (21168:3086001856) ORA_LIB: Loaded Oracle Client library <libclntsh.so> is in the path: </u01/app/oracle/product/10.2.0/client_1/lib>.
(11.7) 01-07-09 10:51:55 (21168:3086001856) JOB: Reading job <89857676_72d8_4bf9_818f_24dc45457e2c> from the repository; Server version is <11.7.3.5>; Repository version is
<11.7.0.0000>.
(11.7) 01-07-09 10:51:55 (21168:3086001856) JOB: Current directory of job <89857676_72d8_4bf9_818f_24dc45457e2c> is </opt/DataIntegrator/NDDI/bin>.
(11.7) 01-07-09 10:51:55 (21168:3086001856) JOB: Starting job on job server host <DataIntegrator-v1.dc.nd.edu>, port <3500>.
(11.7) 01-07-09 10:51:55 (21168:3086001856) JOB: Job <JOB_ADV> of runid <20090107105155211683086001856> is initiated by user <boadmin>.
(11.7) 01-07-09 10:51:55 (21168:3086001856) JOB: Processing job <JOB_ADV>.
And never completes.
This is also the log in the error_JobService.log file:
01-07-09 11:55:02 (20754:-1208867136) JSERVICE: Sleeping for 30sec
01-07-09 11:55:32 (20754:-1208867136) JSERVICE: Polling the JobServer(s) ..
01-07-09 11:55:32 (20754:-1208867136) JSERVICE: Polling the Data Integrator Web Server..
01-07-09 11:55:32 (20754:-1208867136) JSERVICE: Polling the AccessServer(s) ..
01-07-09 11:55:32 (20754:-1208867136) JSERVICE: Sleeping for 30sec
01-07-09 11:56:02 (20754:-1208867136) JSERVICE: Polling the JobServer(s) ..
01-07-09 11:56:02 (20754:-1208867136) JSERVICE: Polling the Data Integrator Web Server..
01-07-09 11:56:02 (20754:-1208867136) JSERVICE: Polling the AccessServer(s) ..
01-07-09 11:56:02 (20754:-1208867136) JSERVICE: Sleeping for 30sec
01-07-09 11:56:32 (20754:-1208867136) JSERVICE: Polling the JobServer(s) ..
01-07-09 11:56:32 (20754:-1208867136) JSERVICE: Polling the Data Integrator Web Server..
01-07-09 11:56:32 (20754:-1208867136) JSERVICE: Polling the AccessServer(s) ..
01-07-09 11:56:32 (20754:-1208867136) JSERVICE: Sleeping for 30sec
Additionally, for the PID noted above, 21168, there is an entry in the stack_trace.txt log in the log directory that reads this:
boadmin@DataIntegrator-v1 log$ cat stack_trace.txt
=== Start Stack Trace dump. PID: 14417===
=== End Stack Trace dump. PID: 14417 ===
=== Start Stack Trace dump. PID: 17290===
=== End Stack Trace dump. PID: 17290 ===
=== Start Stack Trace dump. PID: 21168===
=== End Stack Trace dump. PID: 21168 ===
boadmin@DataIntegrator-v1 log$
Why is the job just hanging like that when we execute it on the Linux job server? Is there something else that needs to be done?
Thanks!
is the job hanging or it has crashed ?
do you see this pid in the output of following command
ps -aef
looks like its crashing, check if there are any core dump file in the $LINK_DIR/bin folder, check the ulimit -a to find if core file size is not set to 0
what is the exact version of DI 11.7.0.0 or 11.7.2.x, or 11.7.3.x
try running the job by enabling trace all option, this can give some information
Similar Messages
-
Test Execution hangs while executing a test from Command prompt option
I have done the following
1) Created a script Test driver
2) Added different scripts as Script->Properties->Assets to the driver script
3) Call each of this asset script from TestDriver using command getScript("<<ScrptName>>".run(1, true, true, true);
4) Now calling this TestDriver script from command prompt with command
e:\\OracleATS\\agent
runScript.bat E:\\INTEGRATION_SERVICES\\Identity_management\\Automation\\Scripts\\MasterScripts\\TestDriver
TestDriver.jwg
The script execution hangs at "Initializing VU 1 for Script TestDriver. After few mintues of seeing this message i get error "Execution Timed Out".
Do I need to keep openscript UI open for script execution from command line ?
Is there any way to increase this Execution Time out ? May be it is taking time to load the internal script assets. AS other scripts are running fine on the same machine from command prompt itself.
Complete error is as follows
C:\Documents and Settings\oracle>e:\\OracleATS\\agent\\runScript.bat E:\\INTEGRATION_SERVICES\\Identity_management\\Automation\\Scripts\\MasterScripts\\TestDriv
er\\TestDriver.jwg
Running "TestDriver" ...
Agent started. Available commands:
stop - Stop the virtual user after it finishes the current iteration.
abort - Abort the virtual user cleanly, before it finishes the current itera
tion.
exit - Terminate the process immediately
You may type the above commands at any time.
12:17:53,418 INFO [1] Initialized script service "oracle.oats.scripting.modules.utilities.api.UtilitiesService"
12:17:53,418 INFO [1] Initialized script service "oracle.oats.scripting.modules.browser.api.BrowserService"
12:17:53,418 INFO [1] Initialized script service "oracle.oats.scripting.modules.functionalTest.api.FunctionalTestService"
12:17:53,465 INFO [1] Initialized script service "oracle.oats.scripting.modules.webdom.api.WebDomService"
12:17:53,856 INFO [1] Initialized script service "oracle.oats.scripting.modules.datatable.api.DataTableService"
12:17:53,856 INFO [1] Initializing VU 1 for script TestDriver
Execution Timed Out.Hi e_raja_sekar,
You wrote:
public static void main(String args){should be
public static void main(String args[]){Granted, the below code was tested on Windows XP using J2SE SDK 1.4.1_02 (and not 1.2.2 like Sridhar is using), but when I compiled and executed this code:
public class BadMain {
public static void main(String args) {
System.out.println("Hello World");
}The output I got was:
Exception in thread "main" java.lang.NoSuchMethodError: mainSo I don't think this is Sridhar's problem (but I could be wrong) since he says that he gets no error message.
Cheers,
Avi. -
Schedule Job Execution problem
Hi All,
The schedule jobs are taking more time.... to complete
We have schdule jobs which will run for every hour,soo to execute the job sometimes it is taking 1hr and sometime 30mins
please help me to have the Job execution total time to be constant to near 30mins.preethi wrote:
Hi All,
The schedule jobs are taking more time.... to complete
We have schdule jobs which will run for every hour,soo to execute the job sometimes it is taking 1hr and sometime 30mins
please help me to have the Job execution total time to be constant to near 30mins.Thread: HOW TO: Post a SQL statement tuning request - template posting
HOW TO: Post a SQL statement tuning request - template posting -
Sql Server Agent Job Execution
Hi All,
I want to run a sql agent job (one job is execution one package) in such a way that as soon as job1 completes , it should be run again (without any time loss)
I am folowing Below approach:
Job 1 has 2 steps , in both the steps I am calling same package to execute . After that on Success/Failure action. I am selecting the same step.it is running in an infinite loop and continue fetching data.
is there any dis-advantages on the above approach?
Is there any other approach to achive the Same?
Thanks in Advance....Hello,
The two-step recursive (on success) technique should be fine. I would just recommend the introduction of short-circuit”techniques that stops a runaway job in the event of logical issues (the “on failure” action will still catch the general error issues).
Other patterns that are sometimes used for continual-execution requirements:
Service Broker (for example, continually watch the queue and act on new messages)
T-SQL home-grown job that runs in a loop and then executes the job based on some condition – and would require you to check job execution state
For a continuously running package watching for new files in a folder, one example is detailed in the following blog post:
Continuously watching files with WMI Event Watcher Task
Best Regards, Joe Sack, MVP, SQL MCM | SackHQ.com -
Enable server group in BPC job execution
Dear Experts
[Note 1476057 - Enable server group in BPC job execution|https://websmp130.sap-ag.de/sap/bc/bsp/spn/sapnotes/index2.htm?numm=0001476057&nlang=E]
After upgrade, the server group logic will be :
1, Create a correct server group with server assigned,
job will be executed in specified server group;
2, Create a server group without server assigned,
job will be successfully submitted to BW,
but will not be executed until server is assigned to the server group;
3, Incorrectly specify a server group in web admin,
client will get error message when run package.
As I understand the group server is defined in BW Server. Where should I specify this group server in BPC. I could not find the parameter in BPC Web Admin - Appset Parameters or BPC Server (.Net Server) - Server Options.
Appreciate if somebody can help ?
Thanks
Ashish
Edited by: Ashish Narway on Oct 14, 2010 2:55 PMHI John,
The group settings in WGM SL and Leopard are different.
Yes, they are. I booted to my eval copy of SL Server and ran through the settings in WGM, Server Admin and Server Preferences; as you pointed out, all the mail list admin tools I could find are in the Mail service in Server Admin.
The only thing that may be of interest that I could find is that for the group mail list, the email address for the group is group_shortname@ServerDNSname, not groupshortname@yourdomain (this is in the Mail Services Admin Guide for 10.6). This might explain why your email to the group may have been rejected?
-Doug -
Got error in job execution, but the standalone procedure runs fine.
I got error in job execution. But it runs fine as standalone procedure. Where could it be wrong?
_>exec dbms_job.run(145373);
BEGIN dbms_job.run(145373); END;
ERROR at line 1:
ORA-12011: execution of 1 jobs failed
ORA-06512: at "SYS.DBMS_IJOB", line 481
ORA-06512: at "SYS.DBMS_JOB", line 275
ORA-06512: at line 1
I checked the alert log file
ORA-06550: line 1, column 101:
PLS-00103: Encountered the symbol "RPTIMPORT" when expecting one of the following:
:= . ( @ % ;
Thu May 2 12:59:34 2013
Errors in file /oracle/admin/PDAMLPR1/udump/pdamlpr11_ora_15574.trc:
ORA-12012: error on auto execute of job 146925
ORA-06550: line 1, column 133:
PLS-00103: Encountered the symbol "" when expecting one of the following:
:= . ( @ % ;
The symbol ";" was substituted for "" to continue.
The trace file:
*** ACTION NAME:() 2013-05-02 12:43:48.261
*** MODULE NAME:(SQL*Plus) 2013-05-02 12:43:48.261
*** SERVICE NAME:(SYS$USERS) 2013-05-02 12:43:48.261
*** SESSION ID:(1811.10644) 2013-05-02 12:43:48.261
*** 2013-05-02 12:43:48.261
ORA-12012: error on auto execute of job 145373
ORA-06550: line 1, column 133:
PLS-00103: Encountered the symbol "" when expecting one of the following:
:= . ( @ % ;
The symbol ";" was substituted for "" to continue.
*** ACTION NAME:(********* Schedule Job: Report C) 2013-05-02 12:47:58.178
*** MODULE NAME:(PATIENT_REMEDIATE_JOB) 2013-05-02 12:47:58.178
*** 2013-05-02 12:47:58.178
ORA-12012: error on auto execute of job 145373
ORA-06550: line 1, column 133:
PLS-00103: Encountered the symbol "" when expecting one of the following:
:= . ( @ % ;
The symbol ";" was substituted for "" to continue.
*** 2013-05-02 12:52:43.409
ORA-12012: error on auto execute of job 146924
ORA-06550: line 1, column 101:
PLS-00103: Encountered the symbol "RPTIMPORT" when expecting one of the following:
:= . ( @ % ;
*** 2013-05-02 12:59:34.651
ORA-12012: error on auto execute of job 146925
ORA-06550: line 1, column 133:
PLS-00103: Encountered the symbol "" when expecting one of the following:
:= . ( @ % ;
The symbol ";" was substituted for "" to continue.
~
~
The job was created as the following with 30 min interval:
The job number: 145373
DECLARE
X NUMBER;
BEGIN
SYS.DBMS_JOB.SUBMIT
( job => X
,what => 'RPTIMPORT.pr_exec_pop_ddl21_data_irt'
,next_date => to_date(sysdate)
,interval => '/*30:Mins*/ sysdate + 30/(60*24)'
,no_parse => TRUE
SYS.DBMS_OUTPUT.PUT_LINE('Job Number is: ' || to_char(x));
END;
Thanks for help!the issue is resolved. I missed to put ";" at end of what statement. Thank all.
-
Job execution fails using SQLServerAgent account but runs successfully under another windows personal account.
Both have same security access on server and DB.
The Windows account is mapped to dbo on the database .
The package uses procedures and views which are using encryption by certificates.
Please advice on how to provide security so that job runs from the agentHi Prady26,
If the SQL Server Agent Service account and the proxy account generated based on a Windows user account have the same permissions on both the SQL Server database and the File System of the server, it may be related to the protection level of the package.
If the package is set to encrypt with UserKey, the job will fail unless it runs under the security context of the Windows user that designs the package in BIDS/SSDT. You can check the
ProtectionLevel of the package in the BIDS/SSDT to see if it is EncryptSensitiveWithUserKey or EncryptAllWithUserKey, if it is, re-deploy/reinstall the package and set the package protection
level to “Relay on server storage for encryption”.
If the issue persists, post the error message in the job history for further analysis.
Regards,
Mike Yin
If you have any feedback on our support, please click
here
Mike Yin
TechNet Community Support -
Find aging of concurrent jobs execution for last one year in R12
DB:11.2.0.3.0
EBS: 12.1.3
O/S: Red Hat Linux
Hi all,
Could anyone please share the resolution for the below question in R12 ?
1) Considering only 3 weeks of data is maintained in FND tables, which means we cannot use fnd_concurrent_requests to get list of jobs submitted in last one year. Is there any other way what we can get to find out aging of jobs execution for last one year?
Thanks for your time!
Regards,Locking this thread - Pl continue the discussion in your original post - Identifying customizations in R12
Unless you have backups of the deleted data, this is not possible -
Error on 11g 2 during external job execution using credential
Hi,
I need a help from you.
I am getting below error on 11g 2 during external job execution using credentials -
EXTERNAL_LOG_ID="job_805812_412409",
ORA-27369: job of type EXECUTABLE failed with exit code: Arg list too long
STANDARD_ERROR="Launching external job failed: Invalid username or password"
Find credentials creation & other required details -
EXECUTE DBMS_SCHEDULER.create_credential(credential_name =>'SYS.SCRIPTS',username =>'scripts',password => 'u5342222');
BEGIN
DBMS_SCHEDULER.CREATE_JOB(
job_name => 'AUTOLOAD_JOB',
job_type => 'EXECUTABLE',
job_action => '/qpshell/dev/scripts/autoload.sh',
repeat_interval => 'freq=MINUTELY;interval=5;byhour=1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22;',
enabled => true,
auto_drop => false,
credential_name => 'SYS.SCRIPTS');
end;
-rw-r--r-- 1 root oinstall 1571 Nov 11 14:42 /u01/app/oracle/product/11.2.0/dbhome_1/rdbms/admin/externaljob.ora
content of externaljob.ora -
run_user = nobody
run_group = nobody
run_user=scripts
run_group=qvcscript
-rwsr-x--- 1 root oinstall 3068424 Nov 11 13:13 extjob
-rwxr-xr-x 1 oracle oinstall 3068424 Nov 11 13:13 extjobo
root@knudwdbs2:/u01/app/oracle/product/11.2.0/dbhome_1/bin> ls -ltr extjob*
-rwx------ 1 oracle oinstall 3068424 Nov 11 12:45 extjoboO
-rwsr-x--- 1 root oinstall 3068424 Nov 11 12:45 extjobO
-rwxr-xr-x 1 oracle oinstall 3068424 Nov 11 13:13 extjobo
-rwsr-x--- 1 root oinstall 3068424 Nov 11 13:13 extjob
Kindly suggest how it can be fixed. Early response much appreciated.
Thanks
Ajay
--Hi,
When you are logged in to the OS, can you do an "su" to the "scripts" user using this exact password (passwords are case-sensitive) ?
Also make sure that you have libpam.so located somewhere in $LD_LIBRARY_PATH and that it points to a real file.
external jobs with credentials use jssu, but I don't think permissions on jssu are the problem here (it should be setuid to root).
Hope this helps,
Ravi. -
Job executions lost in EM Grid Control
Hello,
We have lost all executions from the job execution pane except the backup jobs. We have lost also all scheduled jobs. Any idea about ?
Thanks,
Manuel.Did you create the email Notification Schedule?
-
ODI Interface execution hanging
Hi,
suddenly some interfaces execution hanging, what is the reason for this issue, and how i can overcome this issue?
thanksThe interface is loading from SQL table to oracle table, My IKM are : IKM Oracle Incremental Update and LKM SQL to Oracle,
How can I know it's using memory engine or not ?
what do you mean by ODI Agent is hosed? I am using (Local Agent)
thanks -
Background Job execution enabled via pro-active monitoring
Hi community,
We're interested in CPS Redwood to help us in proactive monitoring of background jobs (Ad Hoc and Periodic) and impacts in system performance.
The idea would be to have somehow a tool that monitor the number of available working process and system performance. When the system workload reaches a certain level, ex: 95%, this tool would intercept new jobs execution. After this, when the system workload goes down and the resources are free, the system would release all jobs to normal run again.
Is it possible via CPS Redwood?
Is there any other way to have something automatic like this?
Thanks,
Jose Antonio.Hi Jose,
This is possible indeed using CPS. CPS can monitor various load metrics of your system, then adapt the scheduling logic if thresholds are exceeded.
In your scenario, you would basically always intercept all jobs, but normally not enforce any limits so effectively that means all jobs still run immediately. If the load is too high, CPS would enforce a limit, so then jobs will be delayed until the load is lower again.
Regards,
Anton. -
Jobs are hanging in running status
Hey ,
I have some overnight jobs that refresh materialized views over a dblink from a remote db. I came this morning and noticed that some of the jobs where hanging in RUNNING state with "more data from dblink" event but there was no corresponding sessions in the remote db. It seems like the refresh process was already done (by the size of the mviews). Tracing the session is useless (i've tried) since no actual job is being done. The alert log gave no clue.
any ideas on what the problem is or what to do in order to find it?
could it be a locking issue ?
ThanksHi,
I think you may need to use normal database hang diagnostic methods to try to figure out what is going on.
The first thing to do is look in the dba_scheduler_running_jobs view, there should be sessions associates with the running view (you can also check that there are real j0 slave processes running the jobs). Once you have checked that there is a real session that is still running the job maybe you could check the v$session view and any other v$ views that should be able to tell you what SQL statement the job is hanging on. You may also be able to check locks held by those sessions.
You may also be able to use Oracle's ADDM reporting tools to get some diagnostic info.
Another possible thing to try is running the job action by hand manually in pl/sql and seeing whether that hangs and possibly figure out where.
Where do you see the "more data from dblink" message ?
-Ravi -
How to suspend (all) job execution ?
Can I define jobs but suspend their execution (until I enable job execution again) ?
If yes, how do I suspend them ?Hi,
You can visit here:
http://download-uk.oracle.com/docs/cd/B19306_01/server.102/b14231/scheduse.htm#i1009724
and also visit here:
http://download-uk.oracle.com/docs/cd/B19306_01/server.102/b14231/scheduse.htm#i1009725
Adith -
Is there a way to generate job execution scripts from the command line?
The customer is planning to change DI production server configuraiton and needs to regenerate job execution scripts for hundreds of jobs. Their question is - instead of going through web admin and doing it one by one, is there a way to run a batch command and generate all jobs with the same job server information from one repository? I looked at the "al_engine" command but didn't find the options to do so. Did I miss something? This is for a DI 11.5.3 on HPUX.
Thanks,
LarryI also suggested that but the customer was so afraid of messing things up that they wanted to find out if there is another way. Thanks a lot for the confirmation.
Larry
Maybe you are looking for
-
Help in downloading IPhoto 6 upgrade
When I try to install upgrade IPhoto 6, it tells me that it cannot find IPhoto 5 on the Applications page. I looked and there it was. It opens, goes to desktop and back...but IPhoto 6 says it is not there. Thanks for any help.
-
SolMan 7.1 Setup Monitoring - Custom Template is not display to Assign
Hi All, SAP Solution Manager Configuration: Technical Monitoring - System Monitoring While creating custom template, i am facing issue, Please check it out below screens, After done, all setup, the custom template should display to apply and active.B
-
What to do next. Install failed, Upgrade Failed, From DVD X.4.6
What to do next. Loaded 9.2.2 did the firmware update, all were Successful, meets Min requirements as for as i can read
-
Sometimes a Front Row playlist appears in iTunes after using Front Row to play music. Why does this happen? Is it a feature? Can I make it stop? Does anybody else experience this? Thanks, Bob
-
Workorder creation reference to an existing Notification.
Hi Everybody, Pl let me know whether any BAPI or FM is there to create a Work Order with reference to an existing Notification. This is in Plant maintenance area. Thanks in advance. praveen k