SORT BY:

LIST ORDER
THREAD
AUTHOR
SUBJECT


SEARCH

IPS HOME


    [Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

    Re: iSCSI: proposal to solve the ISID issues



    John,
    
    >This is an
    > Implementation decision!
    
    I have been trying to point out that it is *not* an implementation decision
    *whether*  to require ISID-configurability (Note below).  OTOH, *how*
    to configure the ISIDs is an implementation decision (as Jim pointed out
    earlier), so also the decision of  *if* it should be made a public API.
    
    Note: And I say this based on three factors -
             - multiple NICs in the Node sharing the same Node Name
             - stated objective of allowing multiple sessions between a given
    initiator
                node and a target portal group.
            - ISID RULE, which forbids parallel nexus
    
    As a final note, Jim alluded to the possibility of "this naming authority
    can
    be elevated to the (SNIA-defined) coordinating entity or even the OS vendor"
    in future.  That possibility requires external ISID-configurability as well,
    AND requires the API to be public.
    
    I see that Jim and Julian had already appreciated where I am coming from.
    I hope I did a better job of convincing you this time.....
    --
    Mallikarjun
    
    Mallikarjun Chadalapaka
    Networked Storage Architecture
    Network Storage Solutions Organization
    Hewlett-Packard MS 5668
    Roseville CA 95747
    ----- Original Message -----
    From: "John Hufferd" <hufferd@us.ibm.com>
    To: <cbm@rose.hp.com>
    Cc: "Jim Hafner" <hafner@almaden.ibm.com>; <ips@ece.cmu.edu>
    Sent: Tuesday, October 30, 2001 1:48 PM
    Subject: Re: iSCSI: proposal to solve the ISID issues
    
    
    >
    > I do not like the word must even if it is in lower case.  This is an
    > Implementation decision! and many/most will not need Admin help.
    > If you were to say "Should allow for configuration and coordination of
    > ISIDs, by the iSCSI/HBA Driver,to allow for configuration and coordination
    > of ISIDs for...." I could go along with that. How that driver gets the
    > information is its business.
    >
    >
    >
    >
    >
    > .
    > .
    > .
    > John L. Hufferd
    > Senior Technical Staff Member (STSM)
    > IBM/SSG San Jose Ca
    > Main Office (408) 256-0403, Tie: 276-0403,  eFax: (408) 904-4688
    > Home Office (408) 997-6136
    > Internet address: hufferd@us.ibm.com
    >
    >
    > "Mallikarjun C." <cbm@rose.hp.com>@ece.cmu.edu on 10/30/2001 11:53:20 AM
    >
    > Please respond to cbm@rose.hp.com
    >
    > Sent by:  owner-ips@ece.cmu.edu
    >
    >
    > To:   Jim Hafner/Almaden/IBM@IBMUS
    > cc:   ips@ece.cmu.edu
    > Subject:  Re: iSCSI: proposal to solve the ISID issues
    >
    >
    >
    > Jim,
    >
    > I am in broad agreement with you on your suggestion,
    > and I agree that the ISID-configurability is a derivative
    > requirement - but my argument is that it's nevertheless
    > a crucial one to warrant making it explicit, but I can
    > live with the proposed compromise of making it a NOTE.
    >
    > Some comments -
    >
    > >There is a requirement
    > > that vendor NICs (managed by the same vendor driver) be configurable in
    > > their ISIDs.
    >
    > I am kind of perplexed on this underlying assumption that
    > a vendor NICs are "managed by the same vendor driver".  As
    > you know, this is not always true.  IMHO, regardless of
    > vendor and driver affiliations, an iSCSI initiator NIC MUST
    > have external ISID-configurability - that's a design constraint
    > on a NIC vendor (from the ISID rule)! [ Note that I am NOT
    > suggesting that the API be "public", it's upto the NIC vendor
    > on allowing third-party drivers - and I believe most would. ]
    >
    > With that, I would suggest the following wordsmithing, but
    > I will not press it any further...
    >
    >   NOTE: For correct behavior (in particular with respect to the ISID
    > rule), a
    >   vendor of iSCSI hardware implementations (e.g., NICs or HBAs) must
    > allow
    >   for configuration and coordination of ISIDs for all sessions managed
    > by
    >   multiple instances of that hardware within a given iSCSI Node.  Such
    >   configuration might be either static (e.g., partitioned across all the
    > NICs
    >   at initialization) or dynamic (e.g., on-line allocator).
    >
    > I hope the suggested wording (or something close to it) on the
    > Node name would make its way in.
    >
    > Finally, thanks for so patiently driving this issue!
    > --
    > Mallikarjun
    >
    >
    > Mallikarjun Chadalapaka
    > Networked Storage Architecture
    > Network Storage Solutions Organization
    > MS 5668   Hewlett-Packard, Roseville.
    > cbm@rose.hp.com
    >
    >
    > Jim Hafner wrote:
    > >
    > > John and Mallikarjun,
    > >
    > > I appreciate where Mallikarjun wants to go here.  There is a requirement
    > > that vendor NICs (managed by the same vendor driver) be configurable in
    > > their ISIDs.  However, that requirement is more a derivative conclusion
    > of
    > > the rules (for a good and correct implementation) than it is a protocol
    > > requirement.  A vendor that doesn't do this only steps on his own toes
    > when
    > > more than one of his cards is in a system!
    > >
    > > Perhaps we compromise and make this a big NOTE:
    > >
    > > NOTE: For correct behavior (in particular with respect to the ISID
    rule),
    > a
    > > vendor of iSCSI hardware implementations (e.g., NICs or HBAs) must
    > provide
    > > either driver software or external and public APIs to allow for
    > > configuration and coordination of ISIDs for all sessions managed by
    > > multiple instances of that hardware within a given iSCSI Node.  Such
    > > configuration might be either static (e.g., partitioned across all the
    > NICs
    > > at initialization) or dynamic (e.g., on-line allocator).
    > >
    > > We can make the iSCSI Node Name a stronger requirement (that is, outside
    > of
    > > a NOTE).
    > >
    > > Does that help?
    > >
    > > Jim Hafner
    > >
    > > John Hufferd/San Jose/IBM@IBMUS@ece.cmu.edu on 10/29/2001 06:47:31 pm
    > >
    > > Sent by:  owner-ips@ece.cmu.edu
    > >
    > > To:   cbm@rose.hp.com
    > > cc:   ips@ece.cmu.edu
    > > Subject:  Re: iSCSI: proposal to solve the ISID issues
    > >
    > > I have no problem with the Must have a way to have the Node name set.
    > > However, I do not grasp the problem you state here:
    > >
    > > "I don't believe the type of the ISID matters, nor does
    > > the NIC capability (coordinating vs not).  If two initiator
    > > NICs share the same node name, they better not re-use the
    > > same ISID to a given target portal group.  I was trying
    > > to capture the consequences of this reality."
    > >
    > > If the ISID includes the Vendor ID, I do not see the problem, unless the
    > > vendor can not assign the rest of the ISID.  And I would say that it is
    > the
    > > vendors problem to work out.  If you are saying that with NICs (not
    HBAs)
    > > that the SW Drivers will have a problem coming up with the ISID, I do
    not
    > > understand that either since most software is built by a Vendor, and if
    > not
    > > the Random type should work.
    > >
    > > Now in case you are saying something about the creation of a second
    > Session
    > > from the Initiator, which is to access the same Target Node.  Then by
    > > definition, the ISID must be unique. (and it will probably need a Wedge
    > > Driver to operate correctly).  I would think that a vendor should be
    able
    > > to come up with a unique qualifier for the session since they have 64K
    > > numbers to chose from.  Remember the vendor only needs to coordinate the
    > > lower 2 bytes of the new ISID, with itself.  So if they can not do that,
    > > ... Oh well ...
    > >
    > > Are we on the same page yet?
    > >
    > > .
    > > .
    > > .
    > > John L. Hufferd
    > > Senior Technical Staff Member (STSM)
    > > IBM/SSG San Jose Ca
    > > Main Office (408) 256-0403, Tie: 276-0403,  eFax: (408) 904-4688
    > > Home Office (408) 997-6136
    > > Internet address: hufferd@us.ibm.com
    > >
    > > "Mallikarjun C." <cbm@rose.hp.com>@core.rose.hp.com on 10/29/2001
    > 06:35:55
    > > PM
    > >
    > > Please respond to cbm@rose.hp.com
    > >
    > > Sent by:  cbm@core.rose.hp.com
    > >
    > > To:   John Hufferd/San Jose/IBM@IBMUS
    > > cc:   ips@ece.cmu.edu
    > > Subject:  Re: iSCSI: proposal to solve the ISID issues
    > >
    > > John,
    > >
    > > I take it that you don't have an issue with the
    > > Node Name part of the wording.
    > >
    > > I thought I answered your question by saying -
    > >
    > > "My point though is that by banning a duplicate nexus (that an
    > > inadvertently re-used ISID allows), we are  making the ISID
    > > (dynamic/static) distribution an imminent hard requirement
    > > for all NICs in a given iSCSI initiator Node"
    > >
    > > The unspoken assumption above is that we would want to
    > > allow multiple sessions between an initiator node and
    > > a target portal group.
    > >
    > > I don't believe the type of the ISID matters, nor does
    > > the NIC capability (coordinating vs not).  If two initiator
    > > NICs share the same node name, they better not re-use the
    > > same ISID to a given target portal group.  I was trying
    > > to capture the consequences of this reality.
    > >
    > > Regards.
    > > --
    > > Mallikarjun
    > >
    > > Mallikarjun Chadalapaka
    > > Networked Storage Architecture
    > > Network Storage Solutions Organization
    > > MS 5668   Hewlett-Packard, Roseville.
    > > cbm@rose.hp.com
    > >
    > > John Hufferd wrote:
    > > >
    > > > Mallikarjun,
    > > > I do not understand the purpose for your proposed wordage
    > > >
    > > > "All iSCSIinitiator hardware implementations MUST in addition also
    > > support
    > > > the operational ISID values to be (either statically or dynamically)
    > > > externally configurable."
    > > >
    > > > Why is this so important that it is a MUST.  The purpose of the
    > proposal
    > > is
    > > > to eliminate the need for the above.  At most it should only apply to
    > the
    > > > vendors (Probably Software and at Universities and IT shops) that use
    > the
    > > > RANDOM ISID type code.
    > > >
    > > > I do not see the need elsewhere.  At most it could be "MAY".
    > > >
    > > > .
    > > > .
    > > > .
    > > > John L. Hufferd
    > > > Senior Technical Staff Member (STSM)
    > > > IBM/SSG San Jose Ca
    > > > Main Office (408) 256-0403, Tie: 276-0403,  eFax: (408) 904-4688
    > > > Home Office (408) 997-6136
    > > > Internet address: hufferd@us.ibm.com
    > > >
    > > > "Mallikarjun C." <cbm@rose.hp.com>@ece.cmu.edu on 10/29/2001 05:24:20
    > PM
    > > >
    > > > Please respond to cbm@rose.hp.com
    > > >
    > > > Sent by:  owner-ips@ece.cmu.edu
    > > >
    > > > To:   Jim Hafner/Almaden/IBM@IBMUS
    > > > cc:   ips@ece.cmu.edu
    > > > Subject:  Re: iSCSI: proposal to solve the ISID issues
    > > >
    > > > Jim,
    > > >
    > > > Thanks for the quick response.
    > > >
    > > > As much as I would like to avoid consuming more bandwidth
    > > > on this topic, let me make one last strawman for wording below.
    > > >
    > > > > What words do you suggest?
    > > > ...
    > > > > And a lot depends on how
    > > > > functional the NICs are: (a) just network nics, (b) protocol nics,
    > but
    > > > not
    > > > > session coordinating nics (c) full session coordination nics (with
    > > driver
    > > > > assist) (d) fully autonomous session nics.   How do we make a
    > > requirement
    > > > > that fits all the cases correctly?
    > > >
    > > > I have been implying iSCSI NICs in all my postings so far -
    > > > sorry, may be that wasn't very clear.  "iSCSI Node name" comment
    > > > obviously wasn't targeted at simple network NICs.  So clearly (a)
    > > > is out of discussion.  But you are right - I wasn't very precise.
    > > > Here's a strawman -
    > > >
    > > >    All iSCSI hardware implementations (as in NICs) MUST allow
    > > >    the iSCSI Node Name to be externally configurable.  All iSCSI
    > > >    initiator hardware implementations MUST in addition also
    > > >    support the the operational ISID values to be (either statically
    > > >    or dynamically) externally configurable.
    > > >
    > > > > Making them configurable (at
    > > > > initialization time, e.g., as a range of values) or making them
    > dynamic
    > > > > (the driver generates them on demand) both fit the mold.  So, I
    don't
    > > > think
    > > > > this one is a hard requirement.
    > > >
    > > > I agree that my earlier "range" wording was too constraining.
    > > > My point though is that by banning a duplicate nexus (that an
    > > > inadvertently re-used ISID allows), we are  making the ISID
    > > > (dynamic/static) distribution an imminent hard requirement
    > > > for all NICs in a given iSCSI initiator Node - except it is
    > > > not sufficiently explicit.  Please consider the above proposed
    > > > wording.
    > > >
    > > > > Well, I don't think so.  If each vendor implements "conservative
    > reuse"
    > > > > within their own ISID namespace on their own NICs, then you get this
    > > > > property system wide.
    > > >
    > > > You are right, I guess I briefly overlooked the fact that ISID
    > > > includes the vendor identifier within, per the new proposal.
    > > >
    > > > I am personally okay with making "conservative reuse" a
    > > > SHOULD.
    > > >
    > > > Regards.
    > > > --
    > > > Mallikarjun
    > > >
    > > > Mallikarjun Chadalapaka
    > > > Networked Storage Architecture
    > > > Network Storage Solutions Organization
    > > > MS 5668   Hewlett-Packard, Roseville.
    > > > cbm@rose.hp.com
    > > >
    > > > Jim Hafner wrote:
    > > > >
    > > > > Mallikarjun,
    > > > >
    > > > > Comments in line below
    > > > >
    > > > > Jim Hafner
    > > > >
    > > > > "Mallikarjun C." <cbm@rose.hp.com>@core.rose.hp.com on 10/29/2001
    > > > 12:17:49
    > > > > pm
    > > > >
    > > > > Please respond to cbm@rose.hp.com
    > > > >
    > > > > Sent by:  cbm@core.rose.hp.com
    > > > >
    > > > > To:   Jim Hafner/Almaden/IBM@IBMUS
    > > > > cc:   ips@ece.cmu.edu
    > > > > Subject:  Re: iSCSI: proposal to solve the ISID issues
    > > > >
    > > > > Jim,
    > > > >
    > > > > Thanks for the excellent summary.  Some questions/comments -
    > > > >
    > > > > - I assume that the NICs must enable the configuration of
    > > > >   iSCSI Node name even in this proposal.  If it is so, please
    > > > >   call this out as a hard requirement in the main modelling
    > > > >   discussion.
    > > > >
    > > > > <JLH>
    > > > > This requirement doesn't change from before (but how it gets written
    > > into
    > > > > the spec may differ -- and we've had this discussion before).  If a
    > > > vendor
    > > > > doesn't allow configuration of his NIC's iSCSI Node name, then his
    > NICs
    > > > > will be acting as their own iSCSI Node (that is, not representing
    the
    > > > > system they live in).  One can argue that this is a legitimate
    > > > > implementation (just as writing user-level code that uses its own
    > iSCSI
    > > > > Node name).  So a "hard requirement" may be a bit too strong.
    > However,
    > > I
    > > > > will go with the will of the group on this point.
    > > > >
    > > > > What words do you suggest?
    > > > > </JLH>
    > > > > - In the case of multiple NICs from the same vendor operating
    > > > >   in an end node, it appears to me that the proposal implicitly
    > > > >   assumes an ISID-range to be configurable among those NICs
    > > > >   (perhaps in vendor-unique ways, which is always what I expected).
    > > > >   If this is a correct inference, there is again a hard
    > > > >   requirement on the NICs to make the ISID range configurable.
    > > > >   Please comment on any subtle qualitative change from the
    > > > >   earlier situation that I may be missing.
    > > > >
    > > > > <JLH>
    > > > > Well, the point is that the vendor can manage his own namespace for
    > his
    > > > > NICs anyway he/she/it wants to.  Making them configurable (at
    > > > > initialization time, e.g., as a range of values) or making them
    > dynamic
    > > > > (the driver generates them on demand) both fit the mold.  So, I
    don't
    > > > think
    > > > > this one is a hard requirement (though that is probably how most
    > > vendors
    > > > > will implement their NICs).  In effect, the proposal gives vendors
    > more
    > > > > flexibility in their own space, without causing heterogenous
    > > > > interoperability problems within the host.  And a lot depends on how
    > > > > functional the NICs are: (a) just network nics, (b) protocol nics,
    > but
    > > > not
    > > > > session coordinating nics (c) full session coordination nics (with
    > > driver
    > > > > assist) (d) fully autonomous session nics.   How do we make a
    > > requirement
    > > > > that fits all the cases correctly?  You clearly have in mind a
    > specific
    > > > > level of implementation within the NICs, but that may not be
    > > everybody's
    > > > > model.
    > > > > </JLH>
    > > > >
    > > > > > (2-e) Its debatable whether "conservative reuse" is a MUST or a
    > > > > > SHOULD.  My personal opinion is "SHOULD", because many systems,
    > > > > > particularly low-end that don't use reservations, can function
    more
    > > or
    > > > > > less OK without it.
    > > > >
    > > > > It seems we're attempting to set ourselves up for future in
    > > > > discussing the above requirement.  Some questions -
    > > > >         - It appears to me that the "conservative reuse" can not
    > > > >           be enforced for initiators hosting NICs from different
    > > > >           vendors (since the proposal allows ISID namespaces to
    > > > >           be totally non-overlapping between non-homogenous NICs).
    > > > >           Is this a fair assessment?
    > > > > <JLH>
    > > > > Well, I don't think so.  If each vendor implements "conservative
    > reuse"
    > > > > within their own ISID namespace on their own NICs, then you get this
    > > > > property system wide.  As before, by owning their own piece of the
    > ISID
    > > > > namespace, they can implement what they want.  So, you may have a
    > > > situation
    > > > > where some of your installed nics have this property and some don't.
    > > > > You'll find out (if your environment really needs conservative
    reuse)
    > > > that
    > > > > you haven't got it, and it becomes a marketing/purchase spec
    > > requirement.
    > > > > </JLH>
    > > > >      - Do you see a particular disadvantage for low-end systems
    > > > >           if it's mandated (aside from the fact that they may be
    able
    > > > >           to live without it)?
    > > > > <JLH>
    > > > > No, but I don't see any way to really enforce it (or even really
    test
    > > > it).
    > > > > It's not a requirement of the protocol, per se.
    > > > > </JLH>
    > > > >      - Do you see any corner cases not being met (for future)
    > > > >           if we don't make it a MUST (since you said "more or less
    > > OK")?
    > > > > <JLH>
    > > > > No, I can't see that far into the future! :-{).  One reason I'm
    being
    > > > cagey
    > > > > here is that I'm finding it difficult to find the right words to
    > > specify
    > > > > this as a hard requirement (but I'm no technical writer either!).
    > > > > </JLH>
    > > > >
    > > > > Regards.
    > > > > --
    > > > > Mallikarjun
    > > > >
    > > > > Mallikarjun Chadalapaka
    > > > > Networked Storage Architecture
    > > > > Network Storage Solutions Organization
    > > > > MS 5668   Hewlett-Packard, Roseville.
    > > > > cbm@rose.hp.com
    > > > >
    > > > > Jim Hafner wrote:
    > > > > >
    > > > > > Folks, (David Black in particular).
    > > > > >
    > > > > > Apologies for the long note, but the issue is complicated.
    > > > > >
    > > > > > The NDTeam have a proposal to resolve the concerns surrounding use
    > of
    > > > > > ISIDs as essential components (along with iSCSI Initiator Name) of
    > > > > > SCSI Initiator Portname.  (This was rooted in private discussions
    > > > > > between John Hufferd, Julian and myself -- and came about as a
    > result
    > > > > > of the lengthy (and boring) discussions mostly myself and David
    > > > > > Black.)
    > > > > >
    > > > > > There are two somewhat orthogonal issues involved in this
    > discussion:
    > > > > >
    > > > > > 1) How to coordinate ISIDs in the initiator to minimize the risk
    of
    > > > > > accidentally breaking the ISID RULE (no parallel nexus) when
    > > > > > independent vendors co-exist in the same OS.
    > > > > >
    > > > > > 2) What ISID "reuse" rules should be specified to facilitate
    > current
    > > > > > and future (soon?) SCSI reservation semantics (and also internal
    OS
    > > > > > views of SCSI Initiator Ports).
    > > > > >
    > > > > > To address these two issues, we (NDT) propose the following:
    > > > > >
    > > > > > 1) Enlarge the ISID field in the Login Request and Response to
    > 6bytes
    > > > > > and structure it with a component that corresponds to a "naming
    > > > > > authority" (essentially the vendor generating the ISID).  So
    > vendors
    > > > > > each have a piece of the ISID namespace to work with their own
    > > > > > components (HBAs, SW, etc). More details below.
    > > > > >
    > > > > > 2) ISIDs (within the namespace of a given iSCSIName) SHOULD be
    used
    > > as
    > > > > > conservatively as possible ("conservative reuse").  What this
    means
    > > is
    > > > > > that a given ISID should be reused for as many sessions (across
    > > > > > multiple target portal groups) as possible (but never to the same
    > > > > > target portal group twice -- that's the ISID RULE).
    > > > > >
    > > > > > NOTES and ADVANTAGES:
    > > > > >
    > > > > > (1-a) Each vendor owns his own piece of the ISID namespace, so in
    > > > > > effect, at the vendor-level, this provides a "partitioning" of
    that
    > > > > > namespace.
    > > > > >
    > > > > > (1-b) We don't need to specify an OS infrastructure requirement
    for
    > > > > > configuration of ISIDs -- each vendor can do it any way it chooses
    > > > > > (within its naming authority).
    > > > > >
    > > > > > (1-c) Breaking of the ISID RULE will only occur if a vendor messes
    > up
    > > > > > its own implementation.
    > > > > >
    > > > > > (1-d) This is not fundamentally different from David Black's
    > > > > > suggestion for a "key"; it just (a) defines it precisely and (b)
    > > > > > embeds it in an already defined (but enlarged) field.
    > > > > >
    > > > > > (1-e) Looking towards the future, as common ISID coordination
    > > > > > mechanisms are implemented between vendors (say, SNIA banding
    > > together
    > > > > > to define a precise API), this "naming authority" can be elevated
    > to
    > > > > > the coordinating entity or even the OS vendor.
    > > > > >
    > > > > > (1-f) ISIDs have no global uniqueness property.  They need only be
    > > > > > unique between a named iSCSI initiator and iSCSI target portal
    > group.
    > > > > > That means that a vendor implementation can use the same algorithm
    > to
    > > > > > generate ISIDs (under its naming authority) in every machine (OS).
    > > > > >
    > > > > > (2-a) If a SCSI target (or logical unit) is holding state (say
    > > > > > persistent reservation) for a SCSI Initiator Port (named by iSCSI
    > > Name
    > > > > > and ISID), and the associated nexus/session is dropped for some
    > > > > > reason, "reuse" by that initiator of that ISID will restore to the
    > > > > > resulting new session that state (with no other action on the part
    > of
    > > > > > the upper SCSI layers).
    > > > > >
    > > > > > (2-b) "Reuse" of an ISID on different sessions to (necessarily)
    > > > > > different SCSI Target Ports (iSCSI Target portal groups), will
    > enable
    > > > > > the SCSI target/logical unit to recognize a common SCSI Initiator
    > > Port
    > > > > > for those two paths.  This facilitates future changes to SCSI
    > > > > > reservations (at least).
    > > > > >
    > > > > > (2-c) An initiator driver implementated with "conservative reuse"
    > can
    > > > > > present to the OS a stable and fairly static view of the SCSI
    > > > > > Initiator Ports (one for each ISID).  Current OS driver stacks
    > > > > > (including current wedge drivers) that are built on the
    presumption
    > > of
    > > > > > such a stable view, will not need modification to handle the more
    > > > > > dynamic nature of iSCSI's SCSI Initiator Port. [Note: the
    existence
    > > of
    > > > > > a SCSI Initiator Port presented to the OS does not *require* that
    > > this
    > > > > > port discover all possible targets; what the initiator builds
    > > sessions
    > > > > > to using a specific ISID will determine what is presented to the
    OS
    > > as
    > > > > > "devices" and LUs visible to that SCSI Initiator Port (could be
    > none
    > > > > > if that ISID is never actually used!).]
    > > > > >
    > > > > > (2-d) (Related to (2-c)): Adding a new (pseudo-static) SCSI
    > Initiator
    > > > > > Port is as simple as configuring another ISID!
    > > > > >
    > > > > > (2-e) Its debatable whether "conservative reuse" is a MUST or a
    > > > > > SHOULD.  My personal opinion is "SHOULD", because many systems,
    > > > > > particularly low-end that don't use reservations, can function
    more
    > > or
    > > > > > less OK without it.  This is an open question.
    > > > > >
    > > > > > DETAILS:
    > > > > >
    > > > > > 1) ISID Structure:
    > > > > > The 6byte ISID structure would be split into three parts:
    > > > > >    "type" field (1byte) -- identifies the format of the naming
    > > > > >                            authority field
    > > > > >    "naming authority" field (3bytes) -- identifies the vendor (or
    > > > > >                            group of vendors) generating this ISID
    > > > > >    "qualifier" (2bytes) -- the free-space for ISID uniqueness
    > > > > >
    > > > > > The type field takes on three defined values with all other values
    > > > > > reserved:
    > > > > >          Type    naming authority format
    > > > > >          00h     IEEE OUI
    > > > > >          01h     IANA Enterprise Number (EN)
    > > > > >          02h     "Random"
    > > > > >
    > > > > > The first types two provide a mechanism to uniquely (world wide)
    > > > > > identify the naming authority.  A vendor with one or more OUIs and
    > > > > > possibly also Enterprise number MUST use at least one of these
    > > numbers
    > > > > > when it generates ISIDs.
    > > > > >
    > > > > > The "Random" type is for the case where the ISID generator (SW or
    > HW)
    > > > > > is provided by an entity that has no OUI or EN.  This includes,
    for
    > > > > > example,
    > > > > > -- a user-written program that builds sessions (and has access to
    > the
    > > > > > system level iSCSI Name)
    > > > > > -- a university or other organization providing the component
    > > > > > -- a testing tool
    > > > > >
    > > > > > For the "Random" type, the naming authority field should be a
    > random
    > > > > > or pseudo-random number. (See below on how this affects
    > "conservative
    > > > > > reuse").
    > > > > >
    > > > > > [Note: the "type" field needn't be this big, but NDT felt that (a)
    > > > > > 2bits was insufficient for the future, (b) the "type" field should
    > be
    > > > > > first, and (c) the "naming authority" field should be
    > byte-aligned.]
    > > > > >
    > > > > > 2) Conservative Reuse
    > > > > >
    > > > > > The "conservative reuse" principle of this proposal just says that
    > > > > > SCSI Initiator Portnames should be viewed as static things (as
    much
    > > as
    > > > > > possible) and reused (when feasible) to give the most stable
    > > > > > presentation of SCSI Initiator Ports to both the OS above and to
    > SCSU
    > > > > > targets across the wire.
    > > > > >
    > > > > > Whereas the current draft says "The ISID RULE does not preclude
    the
    > > > > > use of the same ISID to different Target portal groups", the
    > > > > > "conservative reuse" requirement would say that such reuse (using
    > the
    > > > > > same ISID to many target portal groups) is a SHOULD (or is that
    > > > > > MUST?).  So, in one implementation, each iSCSI Initiator Portal
    > group
    > > > > > would get configured with iSCSI Name, and a (small) set of ISIDs.
    > > The
    > > > > > portal group might take this set of ISIDs as an ordered list.  The
    > > > > > first session to a Target portal group would use the first ISID,
    > the
    > > > > > second to the *same* target portal group would use the second in
    > the
    > > > > > list, etc.  The number of ISIDs would then define a configuration
    > > > > > parameter that can be interpreted as the maximum number of
    sessions
    > > > > > that can be built to a given target portal group.
    > > > > >
    > > > > > To facilitate "conservative reuse", the "qualifier" field of a set
    > of
    > > > > > ISIDs SHOULD be generated using either a repeatable algorithm
    > > > > > (non-random or pseudo-random but based on a fixed seed) or any
    > > > > > algorithm and stored in a persistent location (e.g., registry or
    > /etc
    > > > > > file).
    > > > > >
    > > > > > For the "Random" type, "conservative reuse" may not be an issue
    > (say,
    > > > > > user application that doesn't care about reservations, etc.).
    When
    > > it
    > > > > > is, the "naming authority" field SHOULD also be generated by a
    > > > > > mechanism similar to that for the "qualifier" field as specified
    > > above
    > > > > > (e.g., defined in the SW at compilation time.)
    > > > > >
    > > > > > DOCUMENTATING CHANGES:
    > > > > >
    > > > > > The iscsi drafts would need the following minor changes to support
    > > > > > this proposal:
    > > > > >
    > > > > > NDT doc
    > > > > > (a) define the structure of the ISID field (as described above)
    > > > > > (b) add some notes about implementation in the presense of
    > > > > > "conservative reuse" (e.g., that ISID should be configured once,
    > say
    > > > > > at install time, then reused on subsequent reboots, even for
    > "Random"
    > > > > > type).
    > > > > >
    > > > > > iSCSI MAIN doc:
    > > > > > (a) move the CID field in Login Request to another reserved field.
    > > > > > (b) expand the ISID field in Login Request and Response to
    > encompass
    > > > > > the previous 4 bytes.
    > > > > > (c) add a sentence where the ISID field is defined indicating that
    > > > > > this field is a structured value, showing the structure (as above)
    > > and
    > > > > > point to the NDT document.
    > > > > > (d) add some text to "Note to Implementors" on "conservative
    > reuse".
    > > > > >
    > > > > > Comments?
    > > > > >
    > > > > > Jim Hafner
    > >
    > > --
    > > Mallikarjun
    > >
    > > Mallikarjun Chadalapaka
    > > Networked Storage Architecture
    > > Network Storage Solutions Organization
    > > MS 5668   Hewlett-Packard, Roseville.
    > > cbm@rose.hp.com
    >
    >
    >
    >
    
    


Home

Last updated: Wed Oct 31 04:17:29 2001
7462 messages in chronological order