Re: Performance on 8CPU's and 32GB of RAM
От | Scott Marlowe |
---|---|
Тема | Re: Performance on 8CPU's and 32GB of RAM |
Дата | |
Msg-id | dcc563d10709051712q7f531ecew9fb31ed65dfc2dfa@mail.gmail.com обсуждение исходный текст |
Ответ на | Re: Performance on 8CPU's and 32GB of RAM (Ansgar -59cobalt- Wiechers <lists@planetcobalt.net>) |
Ответы |
Re: Performance on 8CPU's and 32GB of RAM
|
Список | pgsql-performance |
On 9/5/07, Ansgar -59cobalt- Wiechers <lists@planetcobalt.net> wrote: > On 2007-09-05 Scott Marlowe wrote: > > On 9/5/07, Ansgar -59cobalt- Wiechers <lists@planetcobalt.net> wrote: > >> On 2007-09-05 Scott Marlowe wrote: > >>> And there's the issue that with windows / NTFS that when one process > >>> opens a file for read, it locks it for all other users. This means > >>> that things like virus scanners can cause odd, unpredictable > >>> failures of your database. > >> > >> Uh... what? Locking isn't done by the filesystem but by applications > >> (which certainly can decide to not lock a file when opening it). And > >> no one in his right mind would ever have a virus scanner access the > >> files of a running database, regardless of operating system or > >> filesystem. > > > > Exactly, the default is to lock the file. The application has to > > explicitly NOT lock it. It's the opposite of linux. > > Yes. So? It's still up to the application, and it still has nothing at > all to do with the filesystem. And if you look at my original reply, you'll see that I said WINDOWS / NTFS. not just NTFS. i.e. it's a windowsism. > > > And be careful, you're insulting a LOT of people who have come on this > > list with the exact problem of having their anti-virus scramble the > > brain of their postgresql installation. It's a far more common > > problem than it should be. > > How does that make it any less stup^Wintellectually challenged? It doesn't. It's just not necessary to insult people to make a point.
В списке pgsql-performance по дате отправления: