Re: [nfsv4] CB_COPY list of errors (WAS: server-side copyoffloadI-D)

["Erasani, Pranoop" <[e-mail]>]

> > > Assuming the actual s2s protocol is NFSv4.x, I do see many more 
> > > errors
> > to be represented here to make sense of the communication between 
> > client-destination,destination-source. At the minimum, I see the 
> > CB_COPY sending the same set of errors (and many many more) 
> as CB_STATUS.
> > 
> > I started with error codes. But, ended with the question 
> that, there 
> > could be some s2s protocol error codes that the clients 
> need to know.
> > What do you want to do with them? I took an example of a 
> case were s2s 
> > protocol isnot NFSv4.x. What would be the error code that 
> is returned 
> > back, to the client, let's say, if source does not accept the 
> > destination FTP requests.
> 
> If the source does not accept any copy offload requests from 
> the destination, then we have an error defined for COPY -
> "NFS4ERR_PARTNER_NOTSUPP:  The remote server does not support the
>       server-to-server copy offload protocol"
> 
> But even before that, when the client issues the COPY_NOTIFY 
> to source, it should return NFS4ERR_NOSUPP  -
> "NFS4ERR_NOSUPP:  The copy offload operation is not supported by the
>       NFS server receiving this request."

Okay. One more try. I'm not talking about the beginning's of the
workings of the protocol. Yes, I know COPY_NOTIFY, COPY can reject such
requests.

I'm talking about the actual srouce-destination transfer protocol. What
happens if say FTP (mid-way through the transfer) gets an error from the
source? It could be any of the FTP specific errors as described in
RFC765. How does the current protocol propose dealing with the actual
copy protocol errors and how would it enlighten the client about the
same? What sorta of abstract error codes (as defined by this spec) would
the actual copy protocol errors need to be mapped to, so that clients
get a meaningful errors back for an admin to make sense of?

- Pranoop

> 
> Regards,
> Anshul
> 
> > 
> > - Pranoop
> > 
> > > Regards,
> > > Anshul
> > > 
> > > >  
> > > > - Pranoop
> > > > _______________________________________________
> > > > nfsv4 mailing list
> > > > [e-mail]
> > > > https://www.ietf.org/mailman/listinfo/nfsv4
> > > 
> > > 
> 
>