This is a discussion on About index_build within the pgsql Hackers forums, part of the PostgreSQL category; --> Hi, I have found comments in backend/catalog/index.c, that index_build closes the passed rels ( heapRelation and indexRelation ). But ...
| |||||||
| FAQ | Members List | Calendar | Search | Today's Posts | Mark Forums Read |
| ||||
| Hi, I have found comments in backend/catalog/index.c, that index_build closes the passed rels ( heapRelation and indexRelation ). But in backend/access/nbtree/nbtree.c I see something like this: if (IsNormalProcessingMode()) { .... heap_close(heap, NoLock); index_close(index); .... } Why it is not like if (IsNormalProcessingMode()) { .... heap_close(heap, NoLock); index_close(index); .... } else { heap_close(heap, NoLock); index_close(index); } Is it a bug or feature ? Alexey Slynko E-mail: slynko@tronet.ru ---------------------------(end of broadcast)--------------------------- TIP 7: don't forget to increase your free space map settings |
| ||||
| Alexey Slynko <slynko@tronet.ru> writes: > I have found comments in backend/catalog/index.c, that > index_build closes the passed rels ( heapRelation and indexRelation ). Bootstrap mode does it differently. I've never really felt motivated to understand why in any detail. Presumably it was once *necessary* for bootstrap to behave like that, and maybe it still is or maybe not. It ain't broke so why fix it ... regards, tom lane ---------------------------(end of broadcast)--------------------------- TIP 2: you can get off all lists at once with the unregister command (send "unregister YourEmailAddressHere" to majordomo@postgresql.org) |