Subject | Re: [firebird-support] Re: Deadlock, Insert question |
---|---|
Author | Ann W. Harrison |
Post date | 2005-09-06T19:22:09Z |
Adam wrote:
was chosen in 1982 based on actual performance tests. However, machine
performance at expected loads at that time has nothing to do with either
modern performance or modern loads.
Regards,
Ann
>You could try reducing the deadlock timeout from 10 seconds to 1. Ten
> For us, the waiting game caused some UI issues. It was more convenient
> to inform the user that someone else was making changes to the record
> and to try again later, than to freeze to UI in the (usually
> unrealistic) hope the other transaction failed.
>
was chosen in 1982 based on actual performance tests. However, machine
performance at expected loads at that time has nothing to do with either
modern performance or modern loads.
Regards,
Ann