Hi
We have jobs failing in production because it is unable to create datasets based on the defined configuration file. We have a Datastage box and two DB2 boxes where the scratch space is defined. This is the first time we are running the jobs in this environment.
Below is the error.
/test/test.ds: Error when checking operator:
/test/test.ds has 20 partitions, but only 4 are
accessible from the nodes in the configuration file.
[datamgr/deleteop.C:165]
I checked for all the dataset directories defined in the config file exist with right permissions.
Any help to debug this issue would be great.
Unable to create a Dataset
Moderators: chulett, rschirm, roy
-
- Participant
- Posts: 54607
- Joined: Wed Oct 23, 2002 10:52 pm
- Location: Sydney, Australia
- Contact:
Given that the Data Set already exists, it must be accessed using a compatible configuration file (one that refers to the 20 nodes with which it was created). The orchadmin command has options to bypass this requirement but the Data Set stage does not (except in a grid environment in version 8.1) as far as I am aware.
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.
Resolved.
The dataset was originally created using a different configuration file on different boxes where the scratch was defined. When we moved over to new boxes, the job tried to overwrite the existing dataset, but was unable to because of the mismatch in the config files.
I manually went and deleted the old files from all the nodes and then was able to run the job successfully. The admin is working on setting up orchadmin currently.
The dataset was originally created using a different configuration file on different boxes where the scratch was defined. When we moved over to new boxes, the job tried to overwrite the existing dataset, but was unable to because of the mismatch in the config files.
I manually went and deleted the old files from all the nodes and then was able to run the job successfully. The admin is working on setting up orchadmin currently.