RE: RE: Transfer libraries from one box to another

Rob:

AFAIK, SAVRSTxxx commands do not create a local savefile and then transfer the 
savefile. (I'm almost certain I read a mention of this within the past couple 
months or so.) Without implying that IBM did it in a good way, I'd probably try 
doing it with the QaneSava/QaneRsta save/restore APIs with a sockets 
connection. The save/restore exit programs would pass the blocks across the 
socket. Since I've never seen it done, maybe there's a reason that won't work 
-- still, it seems easy enough.

IIRC, the SAVRSTxxx commands _did_ create intermediate savefiles back in 
earlier releases. But I think that's changed somewhere along the line.

Tom Liotta

midrange-l-request@xxxxxxxxxxxx wrote:

>  10. RE: Transfer libraries from one box to another (rob)
>
>date: Fri, 14 Oct 2005 11:51:51 -0500
>
>I can't see how the SAVRST... commands would need any less space than the 
>ftp methodology.  I would think it would take more space.
>Under the covers...
>1 - saves to some temp space, like a save file
>2 - stores the temp space into "sna" space.  Think, you can do a SNDNETF 
>and while the file is transmitting, delete the save file and it will not 
>disrupt the SNDNETF, right?  The eternal optimist might hope that IBM 
>combined this temp save space and the 'sna' space with the SAVRST... 
>family of commands, but I don't know.
>
>"Dwayne Allison" <Dwayne.Allison> 
>
>If we had the space on the old box to save and than transfer, I could
>see your vision. 
>
>-----Original Message-----
>From: midrange-l-bounces@xxxxxxxxxxxx
>[mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of Peter Vidal
>
>Now that we are talking about that, what do you think is the FASTEST 
>option in accomplishing the task of moving the library to your remote 
>location:
>
>(*) SAVLIB to *SAVF / FTP / RESTORE
>(*) SAVLIB to *SAVF / SNDNETF / RSTNETF
>(*) SAVRSTLIB