Internal hash tables - remove limit on amount of data.
As per case TS006348841 currently internal hash tables have a limitation on the amount of data that can be copied into the internal hash table "At the end all the rows/columns must be copied into a hash table. And this is too much data." E.g. A gr...
Implement CREATE OR REPLACE option for triggers/views/synonyms/row types
Since a similar option for stored procedures is available starting with version 14.10 it would be handy if this syntax could also be introduced for triggers, views, synonyms and row types. Currently you first have to drop a trigger/view/synonym or...
I would like the ability to define a global statement and global cursor that can be shared between routines within a single session. This would allow for modular code and create tremendous increases in performance and memoory usuage
Add sql causing change when dictionary entries become dirty.
When onstat -g dic entries become dirty add sql causing this to the onstat -g dic output.This would allow diagnosing issues better. This does not need to be persisted to storage. When sysprocplan entries become invalid this would also allow diagno...
Table repack is a powerful feature but can generate a lot of logical logs. When repacking a large table, which may take many hours, the number of logs created can easily overwhelm the log backup processes and/or cause replication lag.
Potential co...
In Europe we have strong laws (GDPR) protecting personal data. Nobody, even programers or DBA can access personal data if not from an audited program who grants individual access to each user able to see this information. As application examples, ...
HQ: Add a 'Hot tables" sub-menu in the Performance tab, same for LOCKS
It would be a good thing to have, in HQ's performance sub-menu, a 'Hot tables' (and indexes) section which can be based on the output of onstat -g ppf.
The idea is to have a view on the tables that are more active, and eventually click on one tabl...
oncheck should report all bad rowids in an index. Current oncheck stops at the first bad rowid, this makes it impossible to tell if there is more than 1 bad rowid in an index.
Enhance onstat -g prc information: Last recompile datetime Add sql statement that causes this entry to last recompile if changed Add location column (module/line) similar to onstat -g afr output for last change to this entry if an object changed s...