On May 15, 5:30 pm, Ronny <nitely..._at_ist-einmalig.de> wrote:
> > Possibly because archivelogs are in the MB size range and backup
> > pieces can be 2 or more GB in size. I'd hate to put SQL*Net to that
> > stress test given all it's doing now.
>
> > David Fitzjarrell
>
> David,
>
> the volume of network traffic is nearly the same, if using SQL*Net,
> scp, ftp, NFS or some other tool. Oracle always describes its software
> to be highly scalable. SQL*Net when used in conjunction with "imp",
> "exp", Data-Pump and SQL*Loader makes no problems, even if transfering
> large amounts (2 GB and more) of data. It appears to me that network
> load is not the reason why.
>
> Ronny
I don't think "size matters" in this case. All it matters is that
there is no "oracle" running on the target before instantiating the
standby and you cannot instantiate it until you get the files over -
chicken and egg type of thing - therefore "nobody" can handle the
files. Keep in mind that local processes cannot access remote
resources unless they are made local at the O/S level.
Received on Tue May 15 2007 - 16:38:22 CDT
