Agendas

4/11/24

• Complete the Entity Aggregation Field Dispositions
• Review additional changes for incorporation
• Re-review the diagram after correction

3/28/24

• Reviewed Trailers

3/14/24

• Touch base on error code parameter
• Review Entity Aggregation
• Review specific fields, especially with an eye to aggregation rules

2/29/24

• Review Entity Aggregation
• Review specific fields, especially with an eye to aggregation rules
• Discuss any interest in Proof-of-Concept from SVTA QoE WG

2/1/24

• Reiterate open chair position and beg for authors
• Work through list of parameters, assigning profiles and working through text where available.
• Start with the sc field, discussing how to "leave out" fields without confusing processors.

11/16/23

• Reiterate open chair position and beg for authors
• Work through list of parameters, assigning profiles and working through text where available.

10/19/23

• Open chair position & call for authors
• Document draft status
• Continue developing profile definitions
  • Finalize minimal definition - focus on http status code and/or ps-error
  • Finalize the basic profile - enough to get timing, errors, org
  • Finalize "extended" and "verbose" profiles
  • What other profiles do we need vs. menu approach?
  • How do these profiles interact with the non-http call graph

9/28/23

• Open chair position & call for authors
• Specification structure
  • Goals & scope + boiler plate
  • HTTP pull scope + DNS + how it interoperates with push, object lineage/transformations, export
  • Hammer out all language that describes the structure of the trace-data header, minus parameters
    • Explicit section addressing dependencies - cmcd/cmsd, trace context, cache status, proxy status
    • Introduce profiles conceptually first with definitions
  • Define list of parameters & values with profile designations
  • Table for parameters with profile designations where appropriate
  • Example Annex section in the same doc
    • Deep dives with explanations & diagrams
• Specification timeline - first draft by the end of the year, publish by the end of Q1

8/31/23

• Continue developing profile definitions
• Finalize the basic profile - enough to get timing, errors, org
• Finalize "extended" and "verbose" profiles
• What other profiles do we need vs. menu approach?
• How do these profiles interact with the non-http call graph

8/17/23

• Continue developing profile definitions
• Finalize the minimal profile - include org or not? other keys?
• Finalize the basic profile
• What other profiles do we need vs. menu approach?

8/3/23

• Data volume management - solve the problem of wire & cost
  • Sampling - session level, on error, etc.
  • On demand - full data set vs. sampled
  • Verbosity levels

7/20/23

(carried over from the last session due to lack of quorum)

• Full participation example, where everybody participates
  • Example should have a failure an a fallback
  • But everybody participates
  • Initial CDN request failure, failing over to an edge on a different CDN
• Time permitting: review text that implements the changes we've discussed

7/6/23

• Full participation example, where everybody participates
  • Example should have a failure an a fallback
  • But everybody participates
  • Initial CDN request failure, failing over to an edge on a different CDN
• Time permitting: review text that implements the changes we've discussed

6/22/23

• Continued discussion of Session ID, Request ID parameter handling
  • Interactions with cmcd.sid, tracecontext.traceparent
  • Example of the entire request graph, diagrams, narrative

6/8/23

• Session ID & Request ID handling - leveraging CMCD sid, trace parent, etc. (Chris's new example)
• Relating auxiliary data for fulfilling a request to the request graph itself

5/25/23

• Goals review
• Document status & management
• Intent-based tracing & flows (e.g. DRM license challenges, client-side CDN switching, CSAI, etc.)

5/18/23

• Multiple sources with different IDs integrate at different stages (push workflow & multi-camera)
• Discussion roadmap refinement

4/20/23

• Review trace parent/trace id generation and propagation semantics
• Interaction of trace id handling with push flows
• Further triage on the discussion roadmap

4/6/23

• Discussion Roadmap - tune & prioritize?
• Review examples using trace context & trace data together
• Use case: sequence of stacked cdns with live transcoding at multiple intermediary points between cdns
• Trace context on request only or do we need it in response headers
• Interaction with push flows?

3/23/23

• Continue to discuss non-participating clients and request id generation in that context.
• This includes possible use of trace id from trace context.

3/9/23

• Review new document set
• Continue discussion of non-participating clients
• Discuss non-participating CDNs & origins

1/26/23

• Continued discussions
  • scenario deep dives to vet header design and surface additional data elements
  • the relationship between RT headers and other standards-based headers (CMSD, Proxy Status, etc)

1/26/23

• Scenario deep dives

1/12/23

• Check in - where are we now & what's next?
• Configurations, request flows, & scenario itemization
• Scenario deep dives (1 or 2 scenarios)

12/8/22

• Parameter definition sharing with CMSD-dynamic, Proxy Status, Cache Status
• Continued scenario exploration - request collapsing, Client-side CDN switching

11/10/22

• Continued discussion from last week on header structure.
• Brainstorming parameters — Collecting ideas for data we can attach to requests and responses.

10/26/22

• Revised header structure - continued review
• Timing-related claims - examples

10/13/22

• Header structure - revised next hop format + diagrams
  • Simple + origin failover examples

9/15/22

• Short topics
  • Chris out for this session
  • Next meeting canceled - will resume on 10/13
• Tracing scenarios
  • Configurations
  • Request flows

9/1/22

• Continued discussion of
  • Numbered parameters
  • Timing-related claims
  • Trace data structure and semantics
• Session & request ids
• Scenario brainstorm

8/18/22

• General comments on the introduction
• Timing-related claims
• Review currently defined keys
• Session & request ids
• Trace data structure and semantics

8/4/22

• Streaming Media Request Tracing Draft

  • Skeleton/outline - Chris

  • Discussion topics from outline

  • Examples

    • Durations vs. timestamps
      • timestamp precision - nanosecs vs. microsecs, vs. millisecs
    • Representing non-linear request flows (redirects, retries)
    • Representing request collapsing scenarios - trailers?

7/21/22

• Charter review
• Scoping decisions

7/7/22

• Proposed focus areas for the working group
• Prior work quick review & conclusions
• Alignment on charter

6/16/22

• Prior Work: Open Telemetry - Chris Lemmons

6/2/22

• Quick Item
  • Reschedule or cancel June 30th
• W3C Distributed Tracing - Chris
• Discussion

5/18/22

• Quick Items (5 minutes)
  • Slack follow-up - Free workspace has been established. How many members have Slack licenses?
  • Prior work call for experts - OpenTelemetry, Trace Context
• SVA Presentation quick review - Chris, Josh (10 minutes)
• CMSD Tracing Design - Will (40 minutes)

4/28/22

• Quick Topics
  • Slack
  • New working document location
  • Next Session - Prior Work: CMSD Tracing Discussions & Design - Will Law
  • Proposal: May 19 at 9 am PT
• Scope & Scenarios - document status
• Prior Work
  • Proxy Status - Chris Lemmons
  • Identify other prior work - working draft (Open Telemetry, W3C Tracing) - what else?
• Debate Topic List Creation
  • Example: Player timeouts & the need for independent telemetry channels (log streaming, real-time data transfer, etc)
  • Example: Live stream real-time publication of chunk & manifest delivery

4/14/22

• Recurring day/time decision
• Working notes document
• Scope & Scenarios
• Identify prior work - working draft (CMSD, Proxy Status, Open Telemetry, W3C Tracing)