|
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] RE: Re: iSCSI Marker questionsSanjay, I understand the login issue and will send a text that takes care of this soon to the list. Dean's assumptions where wrong anyhow and assume the same knowledge of login space. The new text will say that the first marker appearing after login will be placed AS IF markers where in from time 0 so that the markers can be placed without knowing anything about the login length. Julo
Hi I would agree to the way Dean had thought of where would the first marker be. Assuming marker position is independent of Login phase, marker interval being 4k and login phase being 11k long. the logic, to find where the next marker would be, would be either multiple additions until it crosses 11k OR division of 11k by 4k. It still has to keep track of the FFP TcpSeqNo and start TcpSeqNo to know the Login phase length. I think the simpler way is to start marker after Login phase completes and the next marker is at FFP TcpSeqNo + marker interval which is what Dean stated. In this case you just need to add marker interval to the FFP TcpSeqNo to get the next marker position. Please let me know of your views, Regards Sanjay Goyal -----Original Message----- From: Julian Satran [mailto:Julian_Satran@il.ibm.com] Sent: Tuesday, December 11, 2001 7:54 PM To: ips@ece.cmu.edu Subject: Fw: Re: iSCSI Marker questions ----- Forwarded by Julian Satran/Haifa/IBM on 12-12-01 02:39 -----
Dean, owner-ips@ece.cmu.edu wrote on 11-12-2001 03:09:11: > The iSCSI Draft 9 Appendix C makes the following statements about > Markers and the Initial Marker-less Interval: > > "The offset to the next iSCSI PDU header is counted in terms > of the TCP stream data. Anything counted in the TCP > sequence-number is counted for the offset. Specifically this > includes any bytes "inserted" in the TCP stream by an UFL and > it excludes any other markers inserted between the one we are > examining and the next PDU header."... > > "To enable the connection setup including the login phase > negotiation, marking (if any) is started only at the first > marker interval after the end of the login phase." > > I understand that markers are not inserted until after login phase. > Am I correct to assume that the placement of the first marker > determined by the TCP sequence numbers on the final Login Request/ > Response PDUs, or is initial marker position determined by the > TCP sequence numbers at connection establishment? > > Assume the following interaction: > > I-> SYN (TCP sequenceNum=1000) -- irrelevant to this discussion? > > T-> SYN-ACK (TCP sequenceNum=2000) -- irrelevant to this discussion? > > I-> Login Request PDU, T=0,CSG=1,NSG=0: > InitiatorName=xxx > TargetName=yyy > SessionType=normal > ... > FMarker=send-receive > RFMarkInt=512,1024 > > T-> Login Response PDU, T=0,CSG=1,NSG=0: > ... > FMarker=send-receive > SFMarkInt=1024 > RFMarkInt=1024 > > I-> Login Request PDU, T=1,CSG=1,NSG=3: > SFMarkInt=1024 > (64-byte PDU... TCP sequenceNum=1301-1364) > > T-> Login Response PDU, T=1,CSG=1,NSG=3: > (48-byte PDU... TCP sequenceNum=2201-2248) > > The above interaction designates a 1024 x 4 = 4096-byte marker > interval in both directions. The first PDU byte sent by the > intitiator in full-feature mode will have sequenceNum=1365, and > the first byte sent by the target will have sequenceNum=2249. > > Assuming the markerless interval starts at the end of login > phase, the first two markers in each direction will have the > following TCP sequence numbers: > > TCP SeqNum of TCP SeqNum of > First Marker Second Marker > ------------ ------------- > Initiator: 5461-5468 9565-9572 > Target: 6345-6352 10449-10456 > No - the correct numbers are dependent only on the marker interval (not the length of the login phase) and are: Initiator 5096-5103 9200-9201 Target 6096-6103 10200-10201 > Is this the correct interpretation of marker usage in iSCSI > Draft 9, or does marker placement depend on the connection's > initial sequence numbers? > > Also, is "RFMarkInt=..." always considered an offer, and "SFMarkInt=" > considered a reply to that offer? If an offer is sent with "FMarker=..." > and "RFMarkInt=...", MUST the reply contain either "FMarker=no" or > BOTH "FMarker=yes" and "SFMarkInt=..."? > Fmarker is not boolean - legal values are no, send, receive, send-receive The sender and receiver must set the interval it wants/is ready to use otherwise the responder can't answer. I assume a normal dialogue may go like: I->FMarker=send-receive,RFMarkInt=1,4,SFMarkInt=1,512 T->FMarker=send-receive,RFMarkInt=8, SFMarkInt=2 Please observe that target answers with RFMarkInt to the initiators SFMarkInt and viceversa. I will attempt an example in draft 10 (last?). > Thanks, > Dean Scoville > QLogic Corp.
Home Last updated: Sun Dec 16 13:17:45 2001 8088 messages in chronological order |