Steve French <smfrench@xxxxxxxxx> writes: > Without this change the ioctl() fails with INVALID_PARAMETER. > Since SET_REPARSE_POINT has no output, set the max output response > size to zero. Note that this is more of a work around, I didn't intend to send this yet. I haven't tried to figure out which cases we were violating and where was the underlying issue. > If either InputCount, MaxInputResponse, or MaxOutputResponse is > greater than Connection.MaxTransactSize, the server SHOULD<306> fail > the request with STATUS_INVALID_PARAMETER. possible > The server MUST fail the request with STATUS_INVALID_PARAMETER in the following cases: > > * If InputOffset is greater than zero but less than (size of SMB2 > header + size of the SMB2 IOCTL request not including Buffer) or > if InputOffset is greater than (size of SMB2 header + size of the > SMB2 IOCTL request). input related, nope > * If OutputOffset is greater than zero but less than (size of SMB2 > header + size of the SMB2 IOCTL request not including Buffer) or if > OutputOffset is greater than (size of SMB2 header + size of the SMB2 > IOCTL request). doesn't involve MaxOutputResponse, nope > * If (InputOffset + InputCount) is greater than (size of SMB2 header + > size of the SMB2 IOCTL request). input related, nope > * If (OutputOffset + OutputCount) is greater than (size of SMB2 header > + size of the SMB2 IOCTL request). doesn't involve MaxOutputResponse, nope > * If OutputCount is greater than zero and OutputOffset is less > than (InputOffset + InputCount). doesn't involve MaxOutputResponse, nope So it seems to me only possible explanation is the first, i.e. MaxOutputResponse > Connection.MaxTransactSize, somehow. -- Aurélien Aptel / SUSE Labs Samba Team GPG: 1839 CB5F 9F5B FB9B AA97 8C99 03C8 A49B 521B D5D3 SUSE Linux GmbH, Maxfeldstraße 5, 90409 Nürnberg, Germany GF: Felix Imendörffer, Jane Smithard, Graham Norton, HRB 21284 (AG Nürnberg)
