Transmission of IP over InfiniBand (IPoIB)
draft-ietf-ipoib-ip-over-infiniband-09
Revision differences
Document history
Date | Rev. | By | Action |
---|---|---|---|
2012-08-22
|
09 | (System) | post-migration administrative database adjustment to the No Objection position for Thomas Narten |
2012-08-22
|
09 | (System) | post-migration administrative database adjustment to the No Objection position for Scott Hollenbeck |
2012-08-22
|
09 | (System) | post-migration administrative database adjustment to the No Objection position for Ted Hardie |
2012-08-22
|
09 | (System) | post-migration administrative database adjustment to the No Objection position for David Kessens |
2012-08-22
|
09 | (System) | post-migration administrative database adjustment to the No Objection position for Russ Housley |
2005-01-17
|
09 | Amy Vezza | State Changes to RFC Ed Queue from Approved-announcement sent by Amy Vezza |
2005-01-14
|
09 | Amy Vezza | IESG state changed to Approved-announcement sent |
2005-01-14
|
09 | Amy Vezza | IESG has approved the document |
2005-01-14
|
09 | Amy Vezza | Closed "Approve" ballot |
2005-01-14
|
09 | Margaret Cullen | State Changes to Approved-announcement to be sent from IESG Evaluation::AD Followup by Margaret Wasserman |
2005-01-14
|
09 | Margaret Cullen | Note field has been cleared by Margaret Wasserman |
2005-01-14
|
09 | Thomas Narten | [Ballot Position Update] Position for Thomas Narten has been changed to No Objection from Discuss by Thomas Narten |
2005-01-13
|
09 | (System) | New version available: draft-ietf-ipoib-ip-over-infiniband-09.txt |
2004-12-17
|
09 | (System) | Removed from agenda for telechat - 2004-12-16 |
2004-12-16
|
09 | Michelle Cotton | IANA Comments: Should the reference for hardware type 32 be changed to become this RFC? |
2004-12-13
|
09 | David Kessens | [Ballot Position Update] Position for David Kessens has been changed to No Objection from Discuss by David Kessens |
2004-12-13
|
09 | Ted Hardie | [Ballot Position Update] Position for Ted Hardie has been changed to No Objection from Discuss by Ted Hardie |
2004-12-13
|
09 | Ted Hardie | [Ballot comment] Since my discuss is clearing with no text changes, I'm entering the author's comments to me. I'm okay with these answers. >Ted Hardie: … [Ballot comment] Since my discuss is clearing with no text changes, I'm entering the author's comments to me. I'm okay with these answers. >Ted Hardie: >Discuss: >[2004-08-30] Section 10 and 11 confused me on a couple of points. >First, the emulation of >promiscuous mode seemed a bit odd; I can understand why you would use it >if you're medium supports it, but why emulate rather than send the packets >only to the nodes listening? The emulation is done on the RECEIVE end since we don't have control on all the senders. An earlier solution requiring all the senders to also send a copy of all multicast packets to the routers was rejected in favor of the emulation. Also, the emulation is not done for local multicast - the packets are sent out to the corresponding IB multicast address. When there are no local listeners the packet needs to be received by the router. It was decided by the WG to keep the senders simple and let the router do the emulation since the triggers are already present. > >I was also confused by this: > > Note that for an IPoIB link that spans more than one IB subnet > connected by IB routers, an adequate multicast forwarding support at > the IB level is required for multicast packets to reach listeners on > a remote IB subnet. The specific mechanism for this is beyond the > scope of IPoIB. > >I realize that the specifics are meant to be elsewhere, but this left me >unsure about whether it was possible to have IP multicast without using >the IB multicast at all. The term "remote IB subnet" also seemed like it >could be ambiguous if there are multicast nodes on multiple links. Not sure about the question. IP multicast is building on top of IB multicast, and hence won't work without IB mulitcast. This paragraph simply reiterates this fact, especially for those IPoIB links that span multicast IB links. > |
2004-12-12
|
09 | Sam Hartman | [Ballot Position Update] New position, No Objection, has been recorded for Sam Hartman by Sam Hartman |
2004-12-09
|
09 | Margaret Cullen | Placed on agenda for telechat - 2004-12-16 by Margaret Wasserman |
2004-12-09
|
09 | Margaret Cullen | [Note]: 'On agenda to check if Ted''s, David''s and Thomas'' issues have been addressed.' added by Margaret Wasserman |
2004-12-05
|
09 | Margaret Cullen | [Note]: 'Authors sent follow-up message to ADs with discusses to see if their issues have been addressed.' added by Margaret Wasserman |
2004-12-05
|
09 | Margaret Cullen | Authors submitted updated version and circulated responses to all discuss comments. Waiting for IESG members to review. |
2004-12-02
|
09 | Russ Housley | [Ballot Position Update] Position for Russ Housley has been changed to No Objection from Discuss by Russ Housley |
2004-12-01
|
09 | Scott Hollenbeck | [Ballot Position Update] Position for Scott Hollenbeck has been changed to No Objection from Discuss by Scott Hollenbeck |
2004-11-30
|
09 | (System) | Sub state has been changed to AD Follow up from New Id Needed |
2004-11-30
|
08 | (System) | New version available: draft-ietf-ipoib-ip-over-infiniband-08.txt |
2004-09-24
|
09 | Margaret Cullen | State Changes to IESG Evaluation::Revised ID Needed from IESG Evaluation::AD Followup by Margaret Wasserman |
2004-09-02
|
09 | Amy Vezza | State Changes to IESG Evaluation::AD Followup from IESG Evaluation by Amy Vezza |
2004-09-02
|
09 | Thomas Narten | [Ballot comment] > attributes, and how to setup basic broadcast and multicast services s/setup/set up/ > of its parameters are upto the implementation … [Ballot comment] > attributes, and how to setup basic broadcast and multicast services s/setup/set up/ > of its parameters are upto the implementation and/or the s/upto/up to/ References not split. |
2004-09-02
|
09 | Thomas Narten | [Ballot discuss] > In IPv6 subnets the MTU may be reduced by a Router Advertisement > [DISC] containing an MTU option which specifies … [Ballot discuss] > In IPv6 subnets the MTU may be reduced by a Router Advertisement > [DISC] containing an MTU option which specifies a smaller MTU, or by > manual configuration of each node. If a Router Advertisement > received on an IPoIB interface has an MTU option specifying an MTU > larger than the link MTU or larger than a manually configured value, > that MTU option may be logged to system management but must be > otherwise ignored. This could be worded better. I think that the intent of the IPv6 RA MTU is to allow the sys admin to also specify a higher (not just lower) value if it knows something about the implementations. But the above wording suggests that they can be ignored. What I think you want to say is that the MTU value in an RA MUST be used, if the implementation supports it (and then maybe say what happens otherwise). I don't think the above words say that. > a) Reserved Flags > > These 8 bits are reserved for future use. These bits MUST be > set to zero on send and ignored on receive unless specified > differently in a future document. IANA considerations? > 9.4 Cautionary Note on QPN Caching > > The link-layer address for IPoIB includes the QPN which might not be > constant across reboots or even across network interface resets. > Cached QPN entries, such as in static ARP entries or in RARP servers > will only work if the implementation(s) using these options ensure > that the QPN associated with an interface is invariant across > reboots/network resets. ARP has no caching lifetimes... Seems like this spec should include a recommendation that ARP caches be revalidated periodically, e.g., every few minutes or so. IPv6 does this by default, so there is no issue here. |
2004-09-02
|
09 | Thomas Narten | [Ballot Position Update] New position, Discuss, has been recorded for Thomas Narten by Thomas Narten |
2004-09-02
|
09 | Allison Mankin | [Ballot comment] No further objection. |
2004-09-02
|
09 | Bert Wijnen | [Ballot Position Update] Position for Bert Wijnen has been changed to No Objection from Undefined by Bert Wijnen |
2004-09-02
|
09 | Bert Wijnen | [Ballot comment] I have bno futher DISCUSS items (I guess there are enough already) Comments: - Refrences must be split in normative/informative |
2004-09-02
|
09 | Bert Wijnen | [Ballot Position Update] New position, Undefined, has been recorded for Bert Wijnen by Bert Wijnen |
2004-09-02
|
09 | Harald Alvestrand | [Ballot comment] Reviewed by Brian Carpenter, Gen-ART |
2004-09-02
|
09 | Harald Alvestrand | [Ballot Position Update] New position, No Objection, has been recorded for Harald Alvestrand by Harald Alvestrand |
2004-09-02
|
09 | Bill Fenner | [Ballot Position Update] New position, No Objection, has been recorded for Bill Fenner by Bill Fenner |
2004-09-02
|
09 | Jon Peterson | [Ballot Position Update] New position, No Objection, has been recorded for Jon Peterson by Jon Peterson |
2004-09-02
|
09 | Alex Zinin | [Ballot comment] All comments I had seem to be already included in other DISCUSS'es |
2004-09-02
|
09 | Alex Zinin | [Ballot Position Update] New position, No Objection, has been recorded for Alex Zinin by Alex Zinin |
2004-09-01
|
09 | Allison Mankin | [Ballot Position Update] New position, No Objection, has been recorded for Allison Mankin by Allison Mankin |
2004-09-01
|
09 | David Kessens | [Ballot discuss] I agree with most other issues already raised by the other DISCUSSes. Below are (additional) discussion points as received from the ops directorare … [Ballot discuss] I agree with most other issues already raised by the other DISCUSSes. Below are (additional) discussion points as received from the ops directorare by Pekka Savola: substantial ----------- 4.0 Multicast Mapping [...] A special case exists for the IPv4 limited broadcast address "255.255.255.255" [HOSTS]. The address SHALL be mapped to the "broadcast-GID", which is defined as follows: ==> should the same mapping be done to the all-nodes IPv6 link-local multicast address ff02::1 ? 9.3 Address Resolution in IPv6 Subnets [...] The source/target address option is specified as follows: Length: 3 Link-layer address: The link-layer address is as specified in section 9.1.1 and depicted in figure 5. ==> section 9.1.1 specifies that the length of the link-layer address is 20 bytes (note, it never says this, you have to calculate it yourself from the figure, please fix this!). The SLLA/TLLA option length in RFC2461, however, is defined as the amount in the number of 8 octets, i.e., the length here would be 24 bytes, of which 2 bytes is consumed by the length/type fields. That leaves two unspecified bytes, which would implicitly be at the end. Where do you want to put them (does 32 bit alignment matter?) ? THis requires more text. editorial --------- ==> fails a number of nits, like the old boilerplate, too long lines, reference split, etc. 3.0 InfiniBand Datalink [...] The packet transmission and routing within the IB fabric is also affected by additional parameters such as the traffic class (TClass), hop limit (HopLimit), service level (SL) and the flow label (FlowLabel). ==> what is the "service level" ? The rest have a clear definition at least in IPv6 land, though not necessarily with IPv4.. 11.0 IP Multicast Routing IP multicast routing requires multicast routers to receive a copy of every link multicast packet on a locally connected link [IPMULT, IP6MLD]. For Ethernet this is usually achieved by turning on the promiscuous multicast mode on a locally connected Ethernet interface. ==> IP6MLD is an inappropriate reference here, because MLD/IGMP protocols do not require for the routers to get the copies of all multicast packets. On the other hand, if a node wishes to *send* a multicast packet, it obviously needs to be received by the router (in case it needs to be forwarded off-link). Further, the reference to promiscuous mode may be slightly wrong in a technical sense. At least some Ethernet driver implementations I've seen support either multicast filters (up to some number of groups), or passing all the multicast traffic (which is a subset of promiscuous mode) upwards. 13.0 Security Considerations Controlled Q_Keys SHOULD be used in all transmissions. This is to prevent non-privileged software from fabricating IP datagrams. ==> (the same in section 4.1) I'm concerned about the security model provided by that approach. Is this something similar to how some applications could treat e.g., packets sourced from a "privileged TCP/UDP port" (<1024) as being "secure"? This was considered broken already 10 years ago. This could work (to a degree) if you had an implicilit assumption that only trustworthy nodes would be granted access to an IB link. Note that there is no security whatsoever on who is able to join to an IB link. In other words, this paragraph probably needs to be expanded to not to give the reader a false sense of security. |
2004-09-01
|
09 | David Kessens | [Ballot Position Update] New position, Discuss, has been recorded for David Kessens by David Kessens |
2004-09-01
|
09 | Russ Housley | [Ballot comment] In section 5 and section 7: s/upto/up to/ In section 12: s/should log error messages/SHOULD log error messages/ |
2004-09-01
|
09 | Russ Housley | [Ballot discuss] Section 4 says: > > For IPv4, the group ID is only 28-bit long and the rest of the > … [Ballot discuss] Section 4 says: > > For IPv4, the group ID is only 28-bit long and the rest of the > bits are filled with 0. > How? Is the 28-bit value left justified or right justified? because of the structure used in Figure 2, I suspect the zeros are on the left and the 28-bit value is on the right. Section 5 says: > > It is advisable to ensure that the creation and deletion of > the broadcast group are under administrative control. > I agree. Why isn't there a MUST statement that requires this to be supported? The RFC 2119 term "RECOMMENDED" is not used here either. This is the minimum change to resolve this comment. Section 6 specifies a reserved field, but it does not describe any processing rules for this field. What value should the transmitter use? I assume that the receiver will ignore any value that is present, but this is not stated. I am looking for words similar to those used in section 9.1.1. |
2004-09-01
|
09 | Russ Housley | [Ballot Position Update] New position, Discuss, has been recorded for Russ Housley by Russ Housley |
2004-08-30
|
09 | Ted Hardie | [Ballot discuss] Section 10 and 11 confused me on a couple of points. First, the emulation of promiscuous mode seemed a bit odd; I can … [Ballot discuss] Section 10 and 11 confused me on a couple of points. First, the emulation of promiscuous mode seemed a bit odd; I can understand why you would use it if you're medium supports it, but why emulate rather than send the packets only to the nodes listening? I was also confused by this: Note that for an IPoIB link that spans more than one IB subnet connected by IB routers, an adequate multicast forwarding support at the IB level is required for multicast packets to reach listeners on a remote IB subnet. The specific mechanism for this is beyond the scope of IPoIB. I realize that the specifics are meant to be elsewhere, but this left me unsure about whether it was possible to have IP multicast without using the IB multicast at all. The term "remote IB subnet" also seemed like it could be ambiguous if there are multicast nodes on multiple links. |
2004-08-30
|
09 | Ted Hardie | [Ballot Position Update] New position, Discuss, has been recorded for Ted Hardie by Ted Hardie |
2004-08-30
|
09 | Steven Bellovin | [Ballot discuss] DISCUSS: Section 4 says: For IPv4, the group ID is only 28-bit long and the rest of the bits … [Ballot discuss] DISCUSS: Section 4 says: For IPv4, the group ID is only 28-bit long and the rest of the bits are filled with 0. State explicitly where the 0 bits are inserted, i.e., which end of the 80-bit field. The exmaple seems to imply that it's 52 bits of 0s followed by the group ID, but that isn't clear. How can a node do an expanding ring search for the scope bits? What results tell you if you've gone too far or not far enough? 5.0: It is not clear that this: The method creation of the broadcast group and the assignment/choice of its parameters are upto the implementation and/or the administrator of the IPoIB subnet. will yield interoperable implementations. What if one implementation uses administrative control while another has some implementation-specific scheme? I suggest that administrative control be mandatory, while other schemes are optional and implementation-dependent. 6.0: It says that "implementation MUST be able to handle packets received with or without the use of GRH." What about sending? It seems that implementations MUST be able to send with GRH. 7.0 says " Larger and smaller MTUs MAY be supported". Smaller than 1280 violates the IPv6 spec, I believe. This should be clarified here. 10.0: Implementations should cache the information about the existence of an IB multicast group, its MLID and other attributes. What are typical cache lifetimes? What would suggest longer or shorter lifetimes? What other events should cause cache flushes? |
2004-08-30
|
09 | Amy Vezza | [Ballot Position Update] New position, Discuss, has been recorded for Steve Bellovin by Amy Vezza |
2004-08-16
|
09 | Scott Hollenbeck | [Ballot discuss] References need to be split normative/informative. The reference to the InfiniBand Architecture Specification should then be listed as a normative reference, and a … [Ballot discuss] References need to be split normative/informative. The reference to the InfiniBand Architecture Specification should then be listed as a normative reference, and a better locator than www.infinibandta.org should be provided -- there is no copy of the specification on that page, just a link to "Specifications". It would also be helpful if people didn't have to register with the web site to download a copy of the specification. RFC 2119 should be cited as a reference in section 1.0. 2119 is listed, but not cited (yes, this is a nit). In section 4 it says "For IPv4, the group ID is only 28-bit long and the rest of the bits are filled with 0." Please explain if the leading or trailing bits are to be padded with 0. |
2004-08-16
|
09 | Scott Hollenbeck | [Ballot Position Update] New position, Discuss, has been recorded for Scott Hollenbeck by Scott Hollenbeck |
2004-08-14
|
09 | Margaret Cullen | State Changes to IESG Evaluation from Waiting for Writeup::AD Followup by Margaret Wasserman |
2004-08-14
|
09 | Margaret Cullen | Placed on agenda for telechat - 2004-09-02 by Margaret Wasserman |
2004-08-14
|
09 | Margaret Cullen | [Ballot Position Update] New position, Yes, has been recorded for Margaret Wasserman |
2004-08-14
|
09 | Margaret Cullen | Ballot has been issued by Margaret Wasserman |
2004-08-14
|
09 | Margaret Cullen | Created "Approve" ballot |
2004-08-14
|
09 | Margaret Cullen | Note field has been cleared by Margaret Wasserman |
2004-08-13
|
09 | (System) | Sub state has been changed to AD Follow up from New Id Needed |
2004-08-13
|
07 | (System) | New version available: draft-ietf-ipoib-ip-over-infiniband-07.txt |
2004-05-26
|
09 | Margaret Cullen | [Note]: 'Update needed to clarify status of the "u" bit in IB GUIDs, so that we know whether or not to toggle it when making … [Note]: 'Update needed to clarify status of the "u" bit in IB GUIDs, so that we know whether or not to toggle it when making IPv6 IIDs. The IBTA needs to clarify their draft first, and work is currently underway in the IBTA link WG to do so.' added by Margaret Wasserman |
2004-05-22
|
09 | Margaret Cullen | [Note]: 'Update needed to clarify status of the "u" bit in IB GUIDs, so that we know whether or not to toggle it when making … [Note]: 'Update needed to clarify status of the "u" bit in IB GUIDs, so that we know whether or not to toggle it when making IPv6 IIDs.' added by Margaret Wasserman |
2004-04-08
|
09 | Margaret Cullen | State Changes to Waiting for Writeup::Revised ID Needed from Waiting for Writeup by Margaret Wasserman |
2004-03-29
|
09 | (System) | State has been changed to Waiting for Writeup from In Last Call by system |
2004-03-15
|
09 | Amy Vezza | Last call sent |
2004-03-15
|
09 | Amy Vezza | State Changes to In Last Call from Last Call Requested by Amy Vezza |
2004-03-14
|
09 | Margaret Cullen | Last Call was requested by Margaret Wasserman |
2004-03-14
|
09 | Margaret Cullen | Intended Status has been changed to Proposed Standard from Standard |
2004-03-14
|
09 | Margaret Cullen | [Note]: 'Comments posted to the mailing list on 2002-10-21 regarding all three IPoIB documents. Followup discussion has been ongoing since then. Documents considered to be … [Note]: 'Comments posted to the mailing list on 2002-10-21 regarding all three IPoIB documents. Followup discussion has been ongoing since then. Documents considered to be in WG hands.' has been cleared by Margaret Wasserman |
2004-03-14
|
09 | Margaret Cullen | Last Call was requested by Margaret Wasserman |
2004-03-14
|
09 | Margaret Cullen | State Changes to Last Call Requested from Publication Requested by Margaret Wasserman |
2004-03-14
|
09 | (System) | Ballot writeup text was added |
2004-03-14
|
09 | (System) | Last call text was added |
2004-03-14
|
09 | (System) | Ballot approval text was added |
2004-03-08
|
09 | Dinara Suleymanova | State Changes to Publication Requested from AD is watching by Dinara Suleymanova |
2004-03-08
|
09 | Dinara Suleymanova | Shepherding AD has been changed to Margaret Wasserman from Thomas Narten |
2004-03-08
|
09 | Dinara Suleymanova | Intended Status has been changed to Standard from Proposed Standard |
2004-01-21
|
06 | (System) | New version available: draft-ietf-ipoib-ip-over-infiniband-06.txt |
2003-12-01
|
05 | (System) | New version available: draft-ietf-ipoib-ip-over-infiniband-05.txt |
2003-06-23
|
09 | Thomas Narten | Comments posted to the mailing list on 2002-10-21 regarding all three IPoIB documents. Followup discussion has been ongoing since then. Documents considered to be in … Comments posted to the mailing list on 2002-10-21 regarding all three IPoIB documents. Followup discussion has been ongoing since then. Documents considered to be in WG hands. |
2003-06-23
|
09 | Thomas Narten | State Changes to AD is watching from AD Evaluation by Narten, Thomas |
2003-05-01
|
04 | (System) | New version available: draft-ietf-ipoib-ip-over-infiniband-04.txt |
2003-04-18
|
03 | (System) | New version available: draft-ietf-ipoib-ip-over-infiniband-03.txt |
2003-03-06
|
02 | (System) | New version available: draft-ietf-ipoib-ip-over-infiniband-02.txt |
2002-10-21
|
09 | Thomas Narten | State Changes to AD Evaluation -- AD Followup from Publication Requested by narten |
2002-09-06
|
09 | Stephen Coya | Draft Added by scoya |
2002-05-28
|
01 | (System) | New version available: draft-ietf-ipoib-ip-over-infiniband-01.txt |
2002-02-07
|
00 | (System) | New version available: draft-ietf-ipoib-ip-over-infiniband-00.txt |