Hi,
There is a DataStage job that extract records from a view table in
Oracle database and perform data transformation before it loads to
Unidata table. The job uses the Oracle client connection to the
database server and it is using the OCI8 stage to connect to the Oracle
view table. This job is scheduled to run every 15 minutes throughout
the day from 8 am to 8 pm.
However, at certain runs during the day, the job status is
showing running, but we could see only zero records read in from the
source table via the DataStage monitor screen. This running status
will run forever if we did not stop the job manually and re-run the job
from the DataStage director. It seems the job just hangs without any
activity.
What is the exact cause to this problem? The job did not show any
abort status in the job log. This problem did not have an error message
in the job log which makes our debugging task difficult.
For your information, we are using DataStage version 3.6 and is
running on a HP-UX 11.00 machine.
Our customer is demanding for an explanation. Could this be due to
time-out operation when DataStage trying to get connection to the
Oracle database? Or could this be due to running out of resource on the
DataStage machine?
Would appreciate your advice. Thank you.
Regards,
Melvin
Job status "running" without record output
Melvin,
When this happens, I suggest you go into Oracle and see if DataStage has established a connection. If it has, check what the connection is doing. Is Oracle waiting on DataStage or is Oracle busy with the query.
David Barham
Information Technology Consultant
InformAtect Pty Ltd
Brisbane, Australia
e-mail: david@barham.hm
-----Original Message-----
From: Melvin Thong [mailto:melvin@adtel.esb.com.my]
Sent: Wednesday, 11 July 2001 4:38 PM
To: datastage-users@oliver.com
Subject: Job status "running" without record output
Hi,
There is a DataStage job that extract records from a view table in Oracle database and perform data transformation before it loads to Unidata table. The job uses the Oracle client connection to the database server and it is using the OCI8 stage to connect to the Oracle view table. This job is scheduled to run every 15 minutes throughout the day from 8 am to 8 pm.
However, at certain runs during the day, the job status is showing running, but we could see only zero records read in from the source table via the DataStage monitor screen. This running status will run forever if we did not stop the job manually and re-run the job from the DataStage director. It seems the job just hangs without any activity.
What is the exact cause to this problem? The job did not show any abort status in the job log. This problem did not have an error message in the job log which makes our debugging task difficult.
For your information, we are using DataStage version 3.6 and is running on a HP-UX 11.00 machine.
Our customer is demanding for an explanation. Could this be due to time-out operation when DataStage trying to get connection to the Oracle database? Or could this be due to running out of resource on the DataStage machine?
Would appreciate your advice. Thank you.
Regards,
Melvin
When this happens, I suggest you go into Oracle and see if DataStage has established a connection. If it has, check what the connection is doing. Is Oracle waiting on DataStage or is Oracle busy with the query.
David Barham
Information Technology Consultant
InformAtect Pty Ltd
Brisbane, Australia
e-mail: david@barham.hm
-----Original Message-----
From: Melvin Thong [mailto:melvin@adtel.esb.com.my]
Sent: Wednesday, 11 July 2001 4:38 PM
To: datastage-users@oliver.com
Subject: Job status "running" without record output
Hi,
There is a DataStage job that extract records from a view table in Oracle database and perform data transformation before it loads to Unidata table. The job uses the Oracle client connection to the database server and it is using the OCI8 stage to connect to the Oracle view table. This job is scheduled to run every 15 minutes throughout the day from 8 am to 8 pm.
However, at certain runs during the day, the job status is showing running, but we could see only zero records read in from the source table via the DataStage monitor screen. This running status will run forever if we did not stop the job manually and re-run the job from the DataStage director. It seems the job just hangs without any activity.
What is the exact cause to this problem? The job did not show any abort status in the job log. This problem did not have an error message in the job log which makes our debugging task difficult.
For your information, we are using DataStage version 3.6 and is running on a HP-UX 11.00 machine.
Our customer is demanding for an explanation. Could this be due to time-out operation when DataStage trying to get connection to the Oracle database? Or could this be due to running out of resource on the DataStage machine?
Would appreciate your advice. Thank you.
Regards,
Melvin
Hi
Sometimes, I meet this case at HP11.0
Sometimes, David Barham guess is right. but some other cases is happened at HP11.0
When You run a job, Datastage engine call process and this process folk a child process. this child process connect tho oracle and fatch the dataset. if this process is killed abnormalled, but datastage did not update status, then this job shows status running forever when you made job reset. Beacuse
Datastage manages Job status as filed on universe table.
then you clear status file at datastage director or use DS.TOOLS at TCL Mode (uv)
so to solve this problem
D.H Kim
----- Original Message -----
From: "Melvin Thong"
To:
Sent: Wednesday, July 11, 2001 3:38 PM
Subject: Job status "running" without record output
> Hi,
>
> There is a DataStage job that extract records from a view table in
> Oracle database and perform data transformation before it loads to
> Unidata table. The job uses the Oracle client connection to the
> database server and it is using the OCI8 stage to connect to the Oracle
> view table. This job is scheduled to run every 15 minutes throughout
> the day from 8 am to 8 pm.
>
> However, at certain runs during the day, the job status is
> showing running, but we could see only zero records read in from the
> source table via the DataStage monitor screen. This running status
> will run forever if we did not stop the job manually and re-run the job
> from the DataStage director. It seems the job just hangs without any
> activity.
>
> What is the exact cause to this problem? The job did not show any
> abort status in the job log. This problem did not have an error message
> in the job log which makes our debugging task difficult.
>
> For your information, we are using DataStage version 3.6 and is
> running on a HP-UX 11.00 machine.
>
> Our customer is demanding for an explanation. Could this be due to
> time-out operation when DataStage trying to get connection to the
> Oracle database? Or could this be due to running out of resource on the
> DataStage machine?
>
> Would appreciate your advice. Thank you.
>
> Regards,
> Melvin
>
Sometimes, I meet this case at HP11.0
Sometimes, David Barham guess is right. but some other cases is happened at HP11.0
When You run a job, Datastage engine call process and this process folk a child process. this child process connect tho oracle and fatch the dataset. if this process is killed abnormalled, but datastage did not update status, then this job shows status running forever when you made job reset. Beacuse
Datastage manages Job status as filed on universe table.
then you clear status file at datastage director or use DS.TOOLS at TCL Mode (uv)
so to solve this problem
D.H Kim
----- Original Message -----
From: "Melvin Thong"
To:
Sent: Wednesday, July 11, 2001 3:38 PM
Subject: Job status "running" without record output
> Hi,
>
> There is a DataStage job that extract records from a view table in
> Oracle database and perform data transformation before it loads to
> Unidata table. The job uses the Oracle client connection to the
> database server and it is using the OCI8 stage to connect to the Oracle
> view table. This job is scheduled to run every 15 minutes throughout
> the day from 8 am to 8 pm.
>
> However, at certain runs during the day, the job status is
> showing running, but we could see only zero records read in from the
> source table via the DataStage monitor screen. This running status
> will run forever if we did not stop the job manually and re-run the job
> from the DataStage director. It seems the job just hangs without any
> activity.
>
> What is the exact cause to this problem? The job did not show any
> abort status in the job log. This problem did not have an error message
> in the job log which makes our debugging task difficult.
>
> For your information, we are using DataStage version 3.6 and is
> running on a HP-UX 11.00 machine.
>
> Our customer is demanding for an explanation. Could this be due to
> time-out operation when DataStage trying to get connection to the
> Oracle database? Or could this be due to running out of resource on the
> DataStage machine?
>
> Would appreciate your advice. Thank you.
>
> Regards,
> Melvin
>