Re: Revisited: Transactions, insert unique.

> Ok so I'm biased to how MySQL does it (it's simple and has a good chance of
> working well). Yes it shifts a lot to the application. But if people have
> to do things like do their multiple select for updates in the right order
> (to prevent deadlocks), they might as well start using something like this
> instead (or fix their architecture if possible ;) ).
>
> And it's likely to be faster! Anyone else here like this arbitrary lock
> thingy?
>
> I'm very interested to know of other ways to achieve good serialisation,
> especially database centric methods.
>
> Cheerio,
>
> Link.
>
> p.s. Would anyone actually need timeouts of a day (86400) or greater?

Are you asking for sub-second timeout values?  If so, we could add that.

--
  Bruce Momjian                        |  http://www.op.net/~candle
  pgman@candle.pha.pa.us               |  (610) 853-3000
  +  If your life is a hard drive,     |  830 Blythe Avenue
  +  Christ can be your backup.        |  Drexel Hill, Pennsylvania 19026