|
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] Re: iSCSI:referenced task tagJulian, I am not sure you're referring to the same field as I was. I am referring to the named field in the TMF response PDU - why the non-existent ITT needs to be sent back by the target? Here is the text I am referring to: 9.6.2 Referenced Task Tag If the Request was ABORT TASK and the Response is "task does not exist", the Referenced Task Tag contains the Initiator Task Tag of the task that was to be aborted. In other cases, it MUST be set to 0xffffffff. Thanks. -- Mallikarjun Mallikarjun Chadalapaka Networked Storage Architecture Network Storage Solutions Organization Hewlett-Packard MS 5668 Roseville CA 95747 ----- Original Message ----- From: "Julian Satran" <Julian_Satran@il.ibm.com> To: "Mallikarjun C." <cbm@rose.hp.com> Cc: <ips@ece.cmu.edu> Sent: Thursday, February 28, 2002 11:42 PM Subject: Re: referenced task tag > The reason was to allow the two functions - "abort" and "fill-the-hole" be > performed by a the abort-task task management request. This is important > both on a single connection and on multiple (when switching connections > all the commands on the logged-out connection after expcmdsn have either > to be reissued or aborted). If we want to abort them selectively or > collectively we still have to fill the holes and the only way to do it is > ether abort them with task abort or fill them with nop (otherwise commands > from other connections are held-up). > Your previous note outlines correctly the mechanism and yes you have to > check for RefCmdSN only when the task is not there! > > Julo > > > > > "Mallikarjun C." <cbm@rose.hp.com> > 01-03-02 03:49 > > > To: Julian Satran/Haifa/IBM@IBMIL > cc: > Subject: referenced task tag > > > > Julian, > > I don't see the use of Referenced Task Tag field (9.6.2). > > Can you please comment on why we decided to retain it? > > Thanks! > > Mallikarjun > > > > >
Home Last updated: Fri Mar 01 16:18:07 2002 8974 messages in chronological order |