Failed a keylookup for record
Moderators: chulett, rschirm, roy
Failed a keylookup for record
Hi,
I am getting hte following error when executing a job
"Failed a keylookup for record"
The lookup key is of Integer type. There are 10 primary records and required 10 records key is present in the Lookup table. But still the job fails.
Can anyone throw more light on this?
Thanks
Senthil
I am getting hte following error when executing a job
"Failed a keylookup for record"
The lookup key is of Integer type. There are 10 primary records and required 10 records key is present in the Lookup table. But still the job fails.
Can anyone throw more light on this?
Thanks
Senthil
hello ThilSe,
the information isn't really enought to give an answer in this case; from the comments it seems that this used to work at 7.0 and no longer does after going to 7.5.1. It really boils down to DS thinking that the key isn't present while you think it is.
Normally problems finding keys on lookups boil down the partitioning in the job - what partitioning are you using on your lookup? Is this a lookup set or a different type? Does the error happen when you use a 1-node configuration?
the information isn't really enought to give an answer in this case; from the comments it seems that this used to work at 7.0 and no longer does after going to 7.5.1. It really boils down to DS thinking that the key isn't present while you think it is.
Normally problems finding keys on lookups boil down the partitioning in the job - what partitioning are you using on your lookup? Is this a lookup set or a different type? Does the error happen when you use a 1-node configuration?
ArndW,
I forgot to add this previously.
All the stages in the job are set as "Sequential ".
My job looks like this:
DB2Primary
|
|
`'
Lookup---->DB2Target
^
|
|
DB2Lkup
The same job runs correctly in test environment which is also DS7.5 but it aborts in production.
In prod, I have set the primary keys properly and also manually verified the records. The lookup keys match properly.
I can also view the records in both the DB2Primary and DB2Lkup stages without any warnings. The SQL query runs properly.
But when i execute the job it aborts saying Lookup failed.
Thanks/Regards
Senthil
I forgot to add this previously.
All the stages in the job are set as "Sequential ".
My job looks like this:
DB2Primary
|
|
`'
Lookup---->DB2Target
^
|
|
DB2Lkup
The same job runs correctly in test environment which is also DS7.5 but it aborts in production.
In prod, I have set the primary keys properly and also manually verified the records. The lookup keys match properly.
I can also view the records in both the DB2Primary and DB2Lkup stages without any warnings. The SQL query runs properly.
But when i execute the job it aborts saying Lookup failed.
Thanks/Regards
Senthil
-
- Participant
- Posts: 33
- Joined: Sat Jul 02, 2005 2:27 am
- Location: Bangalore
Hi Senthil,
Before running the jobs do the following points
1. For the lookup to work properly in DS 7.5 the lookup keys should be represented as keys. To do this check the key column in the input tab for the lookup keys. ( Bottom left corner, column name, key, SQL Type... check the key column for the Look up keys).
2. Verify if hash partitioning without sorting is done on the keys for both source and reference
3. Verify whether the datatypes of the keys are matching for both source and target.
After this run and the job and let this forum be informed of how it went through.
Before running the jobs do the following points
1. For the lookup to work properly in DS 7.5 the lookup keys should be represented as keys. To do this check the key column in the input tab for the lookup keys. ( Bottom left corner, column name, key, SQL Type... check the key column for the Look up keys).
2. Verify if hash partitioning without sorting is done on the keys for both source and reference
3. Verify whether the datatypes of the keys are matching for both source and target.
After this run and the job and let this forum be informed of how it went through.
warm regards,
Ajith GK
Ajith GK
Madhavan,
Thanks for your inputs
I have ensured the following:
1. The key column in the input column is checked.
2. The EXECUTION MODE property in the Advanced options tab of ALL the stages is set to SEQUENTIAL (this is due to business requirements). So the partition property has been DISABLED. I cannot set it.
3. The data types are matching.
4. Additionally, there is also KEY Expression property, in the Input lookup link of lookup stage, has been set to the key column.
But the job is throwing an error.
Will there be any problem by setting the EXECUTION MODE property as SEQUENTIAL?
When Execution mode is set as sequential in all the stages, will the data be partitioned?
Thanks/Regards
Senthil
Thanks for your inputs
I have ensured the following:
1. The key column in the input column is checked.
2. The EXECUTION MODE property in the Advanced options tab of ALL the stages is set to SEQUENTIAL (this is due to business requirements). So the partition property has been DISABLED. I cannot set it.
3. The data types are matching.
4. Additionally, there is also KEY Expression property, in the Input lookup link of lookup stage, has been set to the key column.
But the job is throwing an error.
Will there be any problem by setting the EXECUTION MODE property as SEQUENTIAL?
When Execution mode is set as sequential in all the stages, will the data be partitioned?
Thanks/Regards
Senthil
-
- Participant
- Posts: 54607
- Joined: Wed Oct 23, 2002 10:52 pm
- Location: Sydney, Australia
- Contact:
The same job runs correctly in test environment which is also DS7.5 but it aborts in production.
You must tell us what's different between the two environments (which may be DB2 client configuration).
"Nothing" is not the correct answer.
You must tell us what's different between the two environments (which may be DB2 client configuration).
"Nothing" is not the correct answer.
IBM Software Services Group
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.