|
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] Re: ISCSI: need to fix handling of partial data transfersDevice operation is not in the iSCSI domain. You can find the information you need in SCSI. It comes complete with device type, capabilities, alignment required(if any), block sizes and much more (enough to fill many pages for every device class). If you want the information you require just ask for it by SCSI means. A perusal of the SCSI standards for several classes of devices might help. Julo Paul Koning <ni1d@arrl.net> To: Julian Satran/Haifa/IBM@IBMIL Sent by: cc: ips@ece.cmu.edu owner-ips@ece.cmu Subject: Re: ISCSI: need to fix handling of partial data .edu transfers 16-03-02 18:24 Please respond to Paul Koning Excerpt of message (sent 16 March 2002) by Julian Satran: > > Paul, > > We have different opinions on basics. > If I enter into an appliance shop and ask for a bottle of milk - I get an > "error message". > It may fall in one of two categories: > > "protocol error" (in humanes - "you are an idiot") > "we don't keep milk" > > > I agreed that the first was not nice and I will change it. > But there are errors from which you have no recovery - the devices can't > work with each other. I think we're actually pretty much in agreement here. My point is this: Right now the spec says "you can ask for milk in an appliance store, but the shopkeeper is allowed to tell you 'no'". I'm proposing to say "you cannot ask for milk in an appliance store". The outcome is the same either way: the devices discover that they can't talk to each other. But the second rule is simpler. So that's why I proposed to make the rule "you must answer with exactly the data length asked for (R2T) or negotiated (unsolicited data length)". paul
Home Last updated: Mon Mar 18 09:18:15 2002 9175 messages in chronological order |