Have you raise the support ticket for this? If not, please raise it and BMC might release patch for this issue if it is know bug.
I have already raised this issue with support when I came across this for first time, This is closed with below workaround / solution. But my concern is it is hard to ask customer to occasional reboot of server.
I got below reply from support on this -
there could be released db memory space which are either re-used or getting closed. As suggested by IBM, rebooting Cognos Server once/twice in a month will help to release such resource and also help the product to run fine down the line.
I am just looking for any other possibilities to resolve it rather than reboot.
Did the support engineer who worked on the ticket you open reference a defect ID? I would suggest opening a new ticket referencing the previous one and request that the look further into the issue as the workaround is not viable.
1 of 1 people found this helpful
I would check the bsara user indexes status first for any performance, then ETL execution, then data size.
So, are you on Oracle db?, if so as a BSARA_DW user, could you please run below?
SELECT * INDEX_NAME, STATUS FROM USER_INDEXES WHERE STATUS<>'VALID';
Thanks Alan, Isaac
@Isaac, Yes I have oracle db and I have tried things you mentioned. rest are the things looks fine even support also performed all the steps to validate the issue.
I will work with support requesting for further updates and fix.