From: Dmitry Y. <di...@us...> - 2003-10-15 06:42:36
|
Hi, > I can reproduce this in a single processor box. Just let a > long running > transaction/large fetch query running, connect using another > instance and > try to query again. That?s it "lock manager out of room". I?m > not sending anything (like logs, etc) because you people are > investingating the problem, > but I?m sure something is wrong with the win32 lock manager (yet). AFAIK, Nickolay fixed the last known issues with the win32 lock manager after RC6. The fixed binary was sent to interested persons privately and nobody has reported any errors back. So, do you use the original RC6 build or a later one? Dmitry |