OCI lookup
Moderators: chulett, rschirm, roy
OCI lookup
Hello Guru's
I would like to know if any one else faced similar problem and if yes please pass the solution.
I did find some posts similar but not exact
coming to the problem
job fails at transformer stage due to abnormal termination at lookup and writes warring to log then aborts with core dump message.
errors from log
Abnormal termination of stage full_MbrBill_Ld..Transformer_27 detected
DataStage Job 533 Phantom 24912
jobnotify: Unknown error
[24937] DSD.StageRun full_MbrBill_Ld. full_MbrBill_Ld.Transformer_27 2 0/0/1 - core dumped.
DataStage Phantom Finished
job design
Oracle_OCI lookup
|
|
v
Source file ------->transformer--------> Oracle_OCI stage
lookup stage query type was set to user defined.
i changed the settings, query type to Generate Select cluse from columns then this works fine. it works fine only with this option
I need to know why is this happening. or is it known problem with OCI or datastage version
Thank you
I would like to know if any one else faced similar problem and if yes please pass the solution.
I did find some posts similar but not exact
coming to the problem
job fails at transformer stage due to abnormal termination at lookup and writes warring to log then aborts with core dump message.
errors from log
Abnormal termination of stage full_MbrBill_Ld..Transformer_27 detected
DataStage Job 533 Phantom 24912
jobnotify: Unknown error
[24937] DSD.StageRun full_MbrBill_Ld. full_MbrBill_Ld.Transformer_27 2 0/0/1 - core dumped.
DataStage Phantom Finished
job design
Oracle_OCI lookup
|
|
v
Source file ------->transformer--------> Oracle_OCI stage
lookup stage query type was set to user defined.
i changed the settings, query type to Generate Select cluse from columns then this works fine. it works fine only with this option
I need to know why is this happening. or is it known problem with OCI or datastage version
Thank you
lenny
Even though I'm not a Duke or a Ray I'll put something in here. Do a reset and see if the generated log entry is any help. A core dump in a transform stage usually means yuo didn something incorrectly that the compiler didn't detect. What are you doing apart from column 1:1 mapping?
<a href=http://www.worldcommunitygrid.org/team/ ... TZ9H4CGVP1 target="WCGWin">
</a>
</a>
Apology first
I am really sorry I did not mean to say that even though it sounds like it.
I did try reset. I did not get much info from that may be I missed, so I am going to paste
Log reset messages
DataStage Job 533 Phantom 24912
jobnotify: Unknown error
[24937] DSD.StageRun full_MbrBill_Ld. full_MbrBill_Ld.Transformer_27 2 0/0/1 - core dumped.
DataStage Phantom Finished
Job full_MbrBill_Ld aborted.
(sjWeeklyFull) <- full_MbrBill_Ld: Job under control finished.
Resetting Job full_MbrBill_Ld.
Resetting stage full_MbrBill_Ld..Dump_HipBill.
Resetting stage full_MbrBill_Ld..LC_HipVytra.
Finished Resetting stage full_MbrBill_Ld..Dump_HipBill.
Resetting stage full_MbrBill_Ld..Transformer_27.
Finished Resetting stage full_MbrBill_Ld..LC_HipVytra.
Finished Resetting stage full_MbrBill_Ld..Transformer_27.
Job full_MbrBill_Ld has been reset.
(sjWeeklyFull) <- full_MbrBill_Ld: Job under control finished.
I am really sorry I did not mean to say that even though it sounds like it.
I did try reset. I did not get much info from that may be I missed, so I am going to paste
Log reset messages
DataStage Job 533 Phantom 24912
jobnotify: Unknown error
[24937] DSD.StageRun full_MbrBill_Ld. full_MbrBill_Ld.Transformer_27 2 0/0/1 - core dumped.
DataStage Phantom Finished
Job full_MbrBill_Ld aborted.
(sjWeeklyFull) <- full_MbrBill_Ld: Job under control finished.
Resetting Job full_MbrBill_Ld.
Resetting stage full_MbrBill_Ld..Dump_HipBill.
Resetting stage full_MbrBill_Ld..LC_HipVytra.
Finished Resetting stage full_MbrBill_Ld..Dump_HipBill.
Resetting stage full_MbrBill_Ld..Transformer_27.
Finished Resetting stage full_MbrBill_Ld..LC_HipVytra.
Finished Resetting stage full_MbrBill_Ld..Transformer_27.
Job full_MbrBill_Ld has been reset.
(sjWeeklyFull) <- full_MbrBill_Ld: Job under control finished.
apart from mapping doing lookup and nothing else...What are you doing apart from column 1:1 mapping?
lenny
custom SQLkumar_s wrote:what is the Query that you are using to do a lookup. Paste both the query, user defined and Generated. ...
SELECT MRE_0003_MBR_ID_SITE,MRE_3001_MBR_ID_1_3,MRE_3001_MBR_ID_4_5,MRE_3001_MBR_ID_6_9,lpad(MRE_3001_MBR_SEQ_N,2,0)
FROM CCI_hipdwstage1.MbrElg MbrElg
WHERE MRE_0003_MBR_ID_SITE=:1
AND MRE_3001_MBR_ID_1_3=:2
AND MRE_3001_MBR_ID_4_5=:3
AND MRE_3001_MBR_ID_6_9=:4
AND lpad(MRE_3001_MBR_SEQ_N,2,0)=:5
group by MbrElg.MRE_0003_MBR_ID_SITE,MbrElg.MRE_3001_MBR_ID_1_3,MRE_3001_MBR_ID_4_5,MbrElg.MRE_3001_MBR_ID_6_9,MbrElg.MRE_3001_MBR_SEQ_N
Generated SQL
SELECT MRE_0003_MBR_ID_SITE,MRE_3001_MBR_ID_1_3,MRE_3001_MBR_ID_4_5,MRE_3001_MBR_ID_6_9,lpad(MRE_3001_MBR_SEQ_N,2,0)
FROM #$EDW_Stage1_Schema#.MbrElg MbrElg
WHERE MRE_0003_MBR_ID_SITE=:1
AND MRE_3001_MBR_ID_1_3=:2
AND MRE_3001_MBR_ID_4_5=:3
AND MRE_3001_MBR_ID_6_9=:4
AND lpad(MRE_3001_MBR_SEQ_N,2,0)=:5
group by MbrElg.MRE_0003_MBR_ID_SITE,MbrElg.MRE_3001_MBR_ID_1_3,MRE_3001_MBR_ID_4_5,MbrElg.MRE_3001_MBR_ID_6_9,MbrElg.MRE_3001_MBR_SEQ_N
lenny
If you have Oracle ODBC configured you could try using that stage to see if you get success or a better error message. The "core" for the DataStage engine means it is likely in Oracle, as a bad transform stage would normally give a DS/Basic error location before coring out.
If you have a savvy UNIX person around and have "dbx" on your machine you can check the stack to see in which program/library the execution was when the job died. Actually, I think you can do that yourself if you have access to dbx, just call up "dbx /path/to/your/core" and then the command "corefile" (I think) which might be sufficient.
If you have a savvy UNIX person around and have "dbx" on your machine you can check the stack to see in which program/library the execution was when the job died. Actually, I think you can do that yourself if you have access to dbx, just call up "dbx /path/to/your/core" and then the command "corefile" (I think) which might be sufficient.
<a href=http://www.worldcommunitygrid.org/team/ ... TZ9H4CGVP1 target="WCGWin">
</a>
</a>
No, that's not used. I mean the Oracle client on the DataStage server. Forgot to ask - is the Oracle client version on the DataStage server a different version from what is installed on your actual database?
Ultimately, unless we manage to fix this, any kind of answer about core dumps will need to come from your official support provider - not us. Have you opened a case with them yet?
Ultimately, unless we manage to fix this, any kind of answer about core dumps will need to come from your official support provider - not us. Have you opened a case with them yet?
-craig
"You can never have too many knives" -- Logan Nine Fingers
"You can never have too many knives" -- Logan Nine Fingers