Skip to content

Commit

Permalink
Updated Hop Count discussion in DEX draft
Browse files Browse the repository at this point in the history
  • Loading branch information
@talmi
talmi committed Sep 22, 2019
1 parent 2031986 commit 80a20d2
Show file tree
Hide file tree
Showing 2 changed files with 81 additions and 77 deletions.
40 changes: 22 additions & 18 deletions drafts/draft-ioamteam-ippm-ioam-direct-export.xml
View file
Expand Up @@ -422,23 +422,6 @@ packets |Encapsu-| | Transit| | Transit| |Decapsu-|
</t>


<t>Note that if the IOAM-Trace-Type
<xref target="I-D.ietf-ippm-ioam-data"/> has the Hop_Lim/Node_ID bit
set, then exported packets include the Hop_Lim/Node_ID data field,
thus indicating the node from which the packet was exported.
If the DEX header had an explicit Hop Limit field it would be
possible for the collector to have the hop information even when the
Hop_Lim/Node_ID bit is not set. Furthermore, an explicit Hop Limit
field would count the number of IOAM-capable hops, while the Hop_Lim data
field uses the TTL/Hop Limit value from a lower layer protocol, which
may also count IOAM-incapable hops. Conversely, including such an
explicit field in the DEX header would require transit nodes to update
the DEX header in-flight, thus complicating the transit processing,
while one of the main goals of the DEX option was to simplify the
processing of transit nodes. Therefore, the DEX header does not
include an explicity Hop Limit field, and the Hop_Lim/Node_ID data
field can be used instead.</t>

</section>


Expand Down Expand Up @@ -515,7 +498,28 @@ packets |Encapsu-| | Transit| | Transit| |Decapsu-|

<t><list style="symbols">

<t>None at this point.</t>

<t>Hop Limit / Hop Count: in order to help correlate and order the
exported packets, it is possible to include a 1-octet
Hop Count field in the DEX header (presumably by claiming some space
from the Flags field). Its value starts from 0 at the encapsulating
node and is incremented by each IOAM transit node that supports
the DEX option. The Hop Count field value is also included in the
exported packet. An alternative approach is to use the

This comment has been minimized.

Sign in to view

Copy link
@brockners

brockners Oct 1, 2019

Collaborator

Per the discussion on this list: Explict inclusion of a "hop-limit" field that every node would need to update is out of scope for Direct Export. A mode which requires per hop updates is to be considered a variant of IOAM tracing and not direct export. Direct Export is to enable a mode for IOAM where nodes do not need to update IOAM fields.

If you want to keep the discussion in the paragraph, I'd suggest to move it to a dedicated section, e.g. "Related discussions".
Hop_Lim/Node_ID data field; if the IOAM-Trace-Type
<xref target="I-D.ietf-ippm-ioam-data"/> has the Hop_Lim/Node_ID bit
set, then exported packets include the Hop_Lim/Node_ID data field,
which contains the TTL/Hop Limit value from a lower layer protocol.
The main advantage of the Hop Count approach is that it counts the
number of IOAM-capable nodes without relying on the lower layer TTL,
and thus it explicitly allows to detect a case where an IOAM-capable
node fails to export packets to the collector.
The main advantage of the Hop_Lim/Node_ID approach is that it
provides information about the current hop count without requiring
each transit node to modify the DEX option, thus simplifying the data
plane functionality of Direct Exporting.
Thus, further discussion is required about the tradeoff between the
two alternatives.</t>

</list></t>

Expand Down
118 changes: 59 additions & 59 deletions drafts/versions/00/draft-ioamteam-ippm-ioam-direct-export-00.txt
View file
Expand Up @@ -5,7 +5,7 @@
IPPM H. Song
Internet-Draft Futurewei
Intended status: Standards Track B. Gafni
Expires: March 13, 2020 Mellanox Technologies, Inc.
Expires: March 25, 2020 Mellanox Technologies, Inc.
T. Zhou
Z. Li
Huawei
Expand All @@ -15,7 +15,7 @@ Expires: March 13, 2020 Mellanox Technologies, Inc.
Cisco
T. Mizrahi
Huawei Network.IO Innovation Lab
September 10, 2019
September 22, 2019


In-situ OAM Direct Exporting
Expand Down Expand Up @@ -46,14 +46,14 @@ Status of This Memo
time. It is inappropriate to use Internet-Drafts as reference
material or to cite them other than as "work in progress."

This Internet-Draft will expire on March 13, 2020.
This Internet-Draft will expire on March 25, 2020.






Song, et al. Expires March 13, 2020 [Page 1]
Song, et al. Expires March 25, 2020 [Page 1]

Internet-Draft IOAM Direct Exporting September 2019

Expand Down Expand Up @@ -83,11 +83,11 @@ Table of Contents
3.1. Overview . . . . . . . . . . . . . . . . . . . . . . . . 3
3.2. The DEX Option Format . . . . . . . . . . . . . . . . . . 5
4. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 6
4.1. IOAM Type . . . . . . . . . . . . . . . . . . . . . . . . 7
4.2. IOAM DEX Flags . . . . . . . . . . . . . . . . . . . . . 7
4.1. IOAM Type . . . . . . . . . . . . . . . . . . . . . . . . 6
4.2. IOAM DEX Flags . . . . . . . . . . . . . . . . . . . . . 6
5. Performance Considerations . . . . . . . . . . . . . . . . . 7
6. Security Considerations . . . . . . . . . . . . . . . . . . . 7
7. Topics for Further Discussion . . . . . . . . . . . . . . . . 8
7. Topics for Further Discussion . . . . . . . . . . . . . . . . 7
8. References . . . . . . . . . . . . . . . . . . . . . . . . . 8
8.1. Normative References . . . . . . . . . . . . . . . . . . 8
8.2. Informative References . . . . . . . . . . . . . . . . . 8
Expand All @@ -109,7 +109,7 @@ Table of Contents



Song, et al. Expires March 13, 2020 [Page 2]
Song, et al. Expires March 25, 2020 [Page 2]

Internet-Draft IOAM Direct Exporting September 2019

Expand Down Expand Up @@ -165,7 +165,7 @@ Internet-Draft IOAM Direct Exporting September 2019



Song, et al. Expires March 13, 2020 [Page 3]
Song, et al. Expires March 25, 2020 [Page 3]

Internet-Draft IOAM Direct Exporting September 2019

Expand Down Expand Up @@ -221,7 +221,7 @@ Internet-Draft IOAM Direct Exporting September 2019



Song, et al. Expires March 13, 2020 [Page 4]
Song, et al. Expires March 25, 2020 [Page 4]

Internet-Draft IOAM Direct Exporting September 2019

Expand Down Expand Up @@ -277,7 +277,7 @@ Internet-Draft IOAM Direct Exporting September 2019



Song, et al. Expires March 13, 2020 [Page 5]
Song, et al. Expires March 25, 2020 [Page 5]

Internet-Draft IOAM Direct Exporting September 2019

Expand Down Expand Up @@ -309,35 +309,8 @@ Internet-Draft IOAM Direct Exporting September 2019
provides a convenient approach to correlate the
exported data from the same user packet.

Note that if the IOAM-Trace-Type [I-D.ietf-ippm-ioam-data] has the
Hop_Lim/Node_ID bit set, then exported packets include the Hop_Lim/
Node_ID data field, thus indicating the node from which the packet
was exported. If the DEX header had an explicit Hop Limit field it
would be possible for the collector to have the hop information even
when the Hop_Lim/Node_ID bit is not set. Furthermore, an explicit
Hop Limit field would count the number of IOAM-capable hops, while
the Hop_Lim data field uses the TTL/Hop Limit value from a lower
layer protocol, which may also count IOAM-incapable hops.
Conversely, including such an explicit field in the DEX header would
require transit nodes to update the DEX header in-flight, thus
complicating the transit processing, while one of the main goals of
the DEX option was to simplify the processing of transit nodes.
Therefore, the DEX header does not include an explicity Hop Limit
field, and the Hop_Lim/Node_ID data field can be used instead.

4. IANA Considerations







Song, et al. Expires March 13, 2020 [Page 6]

Internet-Draft IOAM Direct Exporting September 2019


4.1. IOAM Type

The "IOAM Type Registry" was defined in Section 7.2 of
Expand All @@ -354,6 +327,17 @@ Internet-Draft IOAM Direct Exporting September 2019
registry includes 16 flag bits. Allocation should be performed based
on the "RFC Required" procedure, as defined in [RFC8126].







Song, et al. Expires March 25, 2020 [Page 6]

Internet-Draft IOAM Direct Exporting September 2019


5. Performance Considerations

The DEX option triggers exported packets to be exported to a
Expand Down Expand Up @@ -385,15 +369,6 @@ Internet-Draft IOAM Direct Exporting September 2019
amplifies the effect on the network bandwidth and on the collector's
load.





Song, et al. Expires March 13, 2020 [Page 7]

Internet-Draft IOAM Direct Exporting September 2019


In order to mitigate the attacks described above, it should be
possible for IOAM-enabled devices to limit the exported IOAM data to
a configurable rate.
Expand All @@ -405,7 +380,34 @@ Internet-Draft IOAM Direct Exporting September 2019

7. Topics for Further Discussion

o None at this point.
o Hop Limit / Hop Count: in order to help correlate and order the
exported packets, it is possible to include a 1-octet Hop Count
field in the DEX header (presumably by claiming some space from
the Flags field). Its value starts from 0 at the encapsulating
node and is incremented by each IOAM transit node that supports
the DEX option. The Hop Count field value is also included in the



Song, et al. Expires March 25, 2020 [Page 7]

Internet-Draft IOAM Direct Exporting September 2019


exported packet. An alternative approach is to use the Hop_Lim/
Node_ID data field; if the IOAM-Trace-Type
[I-D.ietf-ippm-ioam-data] has the Hop_Lim/Node_ID bit set, then
exported packets include the Hop_Lim/Node_ID data field, which
contains the TTL/Hop Limit value from a lower layer protocol. The
main advantage of the Hop Count approach is that it counts the
number of IOAM-capable nodes without relying on the lower layer
TTL, and thus it explicitly allows to detect a case where an IOAM-
capable node fails to export packets to the collector. The main
advantage of the Hop_Lim/Node_ID approach is that it provides
information about the current hop count without requiring each
transit node to modify the DEX option, thus simplifying the data
plane functionality of Direct Exporting. Thus, further discussion
is required about the tradeoff between the two alternatives.

8. References

Expand All @@ -416,7 +418,7 @@ Internet-Draft IOAM Direct Exporting September 2019
Leddy, J., Youell, S., Mizrahi, T., Mozes, D., Lapukhov,
P., Chang, R., daniel.bernier@bell.ca, d., and J. Lemon,
"Data Fields for In-situ OAM", draft-ietf-ippm-ioam-
data-06 (work in progress), July 2019.
data-07 (work in progress), September 2019.

[I-D.mizrahi-ippm-ioam-flags]
Mizrahi, T., Brockners, F., Bhandari, S., Sivakolundu, R.,
Expand All @@ -443,19 +445,15 @@ Internet-Draft IOAM Direct Exporting September 2019





Song, et al. Expires March 13, 2020 [Page 8]
Song, et al. Expires March 25, 2020 [Page 8]

Internet-Draft IOAM Direct Exporting September 2019


[I-D.ietf-sfc-ioam-nsh]
Brockners, F., Bhandari, S., Govindan, V., Pignataro, C.,
Gredler, H., Leddy, J., Youell, S., Mizrahi, T., Mozes,
D., Lapukhov, P., and R. Chang, "Network Service Header
Brockners, F. and S. Bhandari, "Network Service Header
(NSH) Encapsulation for In-situ OAM (IOAM) Data", draft-
ietf-sfc-ioam-nsh-01 (work in progress), March 2019.
ietf-sfc-ioam-nsh-02 (work in progress), September 2019.

[I-D.ioametal-ippm-6man-ioam-ipv6-options]
Bhandari, S., Brockners, F., Pignataro, C., Gredler, H.,
Expand Down Expand Up @@ -501,7 +499,9 @@ Authors' Addresses



Song, et al. Expires March 13, 2020 [Page 9]


Song, et al. Expires March 25, 2020 [Page 9]

Internet-Draft IOAM Direct Exporting September 2019

Expand Down Expand Up @@ -557,7 +557,7 @@ Internet-Draft IOAM Direct Exporting September 2019



Song, et al. Expires March 13, 2020 [Page 10]
Song, et al. Expires March 25, 2020 [Page 10]

Internet-Draft IOAM Direct Exporting September 2019

Expand Down Expand Up @@ -613,4 +613,4 @@ Internet-Draft IOAM Direct Exporting September 2019



Song, et al. Expires March 13, 2020 [Page 11]
Song, et al. Expires March 25, 2020 [Page 11]

1 comment on commit 80a20d2

@talmi
Copy link
Collaborator Author

@talmi talmi commented on 80a20d2 Sep 22, 2019

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The Hop Count issue was updated based on the discussion in the last virtual meeting, and based on the comments sent to the mailing list from Haoyu. Hopefully the current text captures the discussion.

Summary of the last virtual meeting:
https://mailarchive.ietf.org/arch/msg/ippm-ioam-ix-dt/rHBfFHftx1JaTVA2gQoOWOq1hfk

Haoyu's comments:
https://mailarchive.ietf.org/arch/msg/ippm-ioam-ix-dt/TcYgy-uKNFjV5sJc0Vy5rEAnKRA

Cheers,
Tal.

Please sign in to comment.