Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Set your sites on tracing

Set your sites on tracing

Spring I/O presentation of "Set your sites on tracing". Afterwards there were interesting questions about performance of reactive tracing (I'm told next release of spring reactor will be massively less overhead), also a number of folks interested in where to start. Hopefully, the notes in this deck will help answer that.

Adrian Cole

May 16, 2019
Tweet

More Decks by Adrian Cole

Other Decks in Technology

Transcript

  1. Adrian Cole
    @adrianfcole
    Set your sites on tracing

    View full-size slide

  2. Set your sites on tracing
    An overview of distributed tracing practice
    @adrianfcole
    works at Pivotal
    works on Zipkin

    View full-size slide

  3. Introduction
    introduction
    a typical zipkin site
    wrapping up
    @adrianfcole
    #zipkin

    View full-size slide

  4. @adrianfcole
    • spring cloud at pivotal
    • focused on distributed tracing
    • helped open zipkin

    View full-size slide

  5. What is Distributed Tracing?
    Distributed tracing tracks production requests as they touch
    different parts of your architecture.
    Requests have a unique trace ID, which you can use to lookup a
    trace diagram, or log entries related to it.
    Causal diagrams are easier to understand than scrolling through logs.

    View full-size slide

  6. Example Trace Diagram
    Wire Send Store
    Async Store
    Wire Send
    POST /things
    POST /things

    View full-size slide

  7. Why do I care?
    - Reduce time in triage by contextualizing errors and delays
    - Visualize latency like time in my service vs waiting for other services
    - Understand complex applications like async code or microservices
    - See your architecture with live dependency diagrams built from traces

    View full-size slide

  8. Distributed Tracing Vocabulary
    A Span is an individual operation that took place. A span contains timestamped
    events and tags.
    A Trace is an end-to-end latency graph, composed of spans.
    Tracers records spans and passes context required to connect them into a trace
    Instrumentation uses a tracer to record a task such as an http request as a span

    View full-size slide

  9. wombats:10.2.3.47:8080
    A Span is an individual operation
    Server Received a Request
    POST /things
    Server Sent a Response
    Events
    Tags
    Operation
    request-i
    d
    abcd-ffe
    http.status_code 202
    http.url …&features=HD-uploads

    View full-size slide

  10. Tracing is capturing important events
    Wire Send Store
    Async Store
    Wire Send
    POST /things
    POST /things

    View full-size slide

  11. Tracers record time, duration and host
    Wire Send Store
    Async Store
    Wire Send
    POST /things
    POST /things
    Tracers don’t decide what to record, instrumentation does.. we’ll get to that

    View full-size slide

  12. Tracers send trace data out of process
    Tracers propagate IDs in-band,
    to tell the receiver there’s a trace in progress
    Completed spans are reported out-of-band,
    to reduce overhead and allow for batching

    View full-size slide

  13. How do I turn on tracing?
    A tracer is a utility library, similar to metrics or logging libraries. It is a mechanism
    uses to trace an operation.
    Instrumentation is framework-specific code that uses a tracer to collect details
    such as the http url and request timing.
    Instrumentation must be configured and pointed to a tracing system for tracing to
    work. This is often done automatically with agents or frameworks like Spring Boot.

    View full-size slide

  14. Zipkin is a distributed tracing system
    real trace from type form here in barcelona

    View full-size slide

  15. Zipkin can be as simple as one file listening on one port
    $ curl -sSL https://zipkin.apache.org/quickstart.sh | bash -s
    $ SELF_TRACING_ENABLED=true java -jar zipkin.jar
    ********
    ** **
    * *
    ** **
    ** **
    ** **
    ** **
    ********
    ****
    ****
    **** ****
    ****** **** ***
    ****************************************************************************
    ******* **** ***
    **** ****
    **
    **
    ***** ** ***** ** ** ** ** **
    ** ** ** * *** ** **** **
    ** ** ***** **** ** ** ***
    ****** ** ** ** ** ** ** **
    :: Powered by Spring Boot :: (v2.1.4.RELEASE)
    2019-05-16 14:52:44.695 INFO 30176 --- [ main] z.s.ZipkinServer : Starting ZipkinServer v2.13.0 on MacBook-Pro-7.local with PID 30176 (/private/tmp/
    zipkin.jar started by acole in /private/tmp)
    —snip—
    $ curl -s localhost:9411/api/v2/services|jq .
    [
    "api-proxy",
    "auth-api",
    "phoenix"
    ]

    View full-size slide

  16. A typical Zipkin site
    introduction
    a typical zipkin site
    wrapping up
    @adrianfcole
    #zipkin

    View full-size slide

  17. What is a Zipkin site
    Site owner: End user who champions Zipkin as a part of additional roles in their
    company. Many site owners are part time, yet contribute back to open source.
    Zipkin site: Production deployment of distributed tracing, which considers Zipkin
    format, instrumentation or backends strategic to their observability function.
    For example, some sites use tools like zipkin-php or zipkin-go to collect data, but export it to Google Stackdriver for analysis and visualization. Others use our data format in their tracing pipeline which
    gathers data from Zipkin tools, alternative or legacy internal ones. Some use Zipkin backends, but other agent technology, such as SkyWalking to gather it. By sharing real life setups, you can ideally
    understand the different approaches that coexist to solve the problem of tracing.

    View full-size slide

  18. What information do we collect on Zipkin sites
    * Introduction of the company context and team on tracing
    * System overview from application until visualization/analysis
    * Site-specific data conventions such as services are named
    * Why tracing is important, goals and service level agreements
    * Status like costs adoption, ingestion and costs incurred
    System Overview

    instrumentation - approach, platforms supported

    data ingestion - formats, data pipeline, sampling

    data store and aggregation - data at rest, retention, indexing, cleansing

    realtime and batch analysis - techniques, visualizations, UI, tooling

    Site-specific data conventions

    service name - what is the source of your Zipkin service name? does it come from discovery? Is it used in other tools like metrics?

    site-specific tags - which tags do you rely on for search or aggregation? For example, do you add correlation or environment IDs to spans? Which are fixed cardinality?

    Goals

    What near, middle and long term milestones exist?

    What value are the business looking to receive?

    What improvements are you looking to further?

    What other projects relate to your tracing goals?

    View full-size slide

  19. Why bother with tracing?
    Ascend Money says: Measure latency improvements before and after
    refactoring the services. Identify non-conformant service communications that
    deviates from the design.
    Hotels.com says: helps in pointing out the worst offenders and by making it
    easier to identify performance improvements such as network calls that could
    be done in parallel.
    Netflix says: The business value is in providing operational visibility into the
    systems and enhance developer productivity.
    https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=95651348

    https://cwiki.apache.org/confluence/display/ZIPKIN/Hotels.com

    https://cwiki.apache.org/confluence/display/ZIPKIN/Netflix

    View full-size slide

  20. What kind of infrastructure is involved?
    Effective tracing matches
    the architecture and skillset
    of the site owners.
    Sites have different
    application and tracing
    infrastructures.

    View full-size slide

  21. So, a site doesn’t only run Zipkin server?
    Zipkin Server is the canonical backend which receives Zipkin
    format, and presents a UI. Some don’t run Zipkin server, or also run
    other products for various reasons.
    * SaaS preference
    * APM integration
    * Hybrid setup
    * SaaS preference: Ex. Infostellar use Google Stackdriver

    * APM integration: Ex. hotels.com use a centralised Expedia Haystack which also adds anomaly detection

    * Pipeline setup: Ex. yelp have “firehose” collectors which forward to a zipkin server

    View full-size slide

  22. And.. applications don’t always use Zipkin libraries?!
    Zipkin curates propagation and trace formats which decouple
    sites from a mandate of using our code. By producing the same
    data, applications have more flexibility and choice.
    * 3rd party libraries
    * Proxies (service mesh) integration
    * In-house custom tools
    * 3rd party libraries - ex OpenCensus and some OpenTracing libraries

    * Proxies (service mesh) integration - ex linkerd and envoy do not use zipkin libraries for tracing, but they emit our formats

    * In-house custom tools - ex utilities not yet open sourced or are too bespoke.

    View full-size slide

  23. Let’s look at a site that once used Zipkin server
    Hotels.com started with a Zipkin backend, but are transitioning to Expedia
    Haystack, which provides more features like adaptive alerting.
    https://github.com/ExpediaDotCom/haystack
    Applications still emit data in Zipkin v2 format, which is forwarded to Haystack
    with a tool they created called Pitchfork. Developers still use Zipkin on their
    laptops for local troubleshooting, as it is easy to run.
    Hotels.com is part of the Expedia Group and is a website for booking hotel rooms online. It's tracing team operates from London.

    https://cwiki.apache.org/confluence/display/ZIPKIN/Hotels.com

    View full-size slide

  24. Let’s look at a site that didn’t initially use Zipkin server
    Netflix created a Dapper-based tracing system to trace RPC calls involved in video
    streaming. This included framework libraries to produce trace headers and data.
    As Spring Boot became prevalent, Zipkin became more useful as it is built-into the
    tracing library Spring Cloud Sleuth.
    Netflix convert legacy spans into Zipkin v2 format in their Kafka/Flink pipeline. This
    allows traces to stitch together for query and analysis.
    Netflix is a video streaming service. Its tracing team operates from the Silicon Valley.

    https://cwiki.apache.org/confluence/display/ZIPKIN/Netflix

    View full-size slide

  25. Let’s look at a site that never used Zipkin server
    Infostellar architecture runs in Google Cloud, except ground station software
    that runs locally at an antennae site.
    Many components trace with Zipkin libraries, some with OpenTracing, some
    homegrown. All use Zipkin’s B3 format for propagation.
    Even when using Zipkin libraries, data sends directly to Google Stackdriver for
    query and analysis. There’s no Zipkin server footprint at Infostellar.
    Infostellar (StellarStation) - is a space communications infrastructure firm. Its tracing team operates from Tokyo.

    https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=95655004

    View full-size slide

  26. Let’s look at a site that uses stock Zipkin server
    Medidata is an entirely AWS architecture, using the zipkin-aws
    image will allows http and SQS span collection. They collect 100%
    data into AWS-managed Elasticsearch storage.
    While the zipkin service is standard, Medidata has a service that
    reads trace data from Elasticsearch, comparing it with performance
    objectives in APIs and issuing alerts when performance degrades.
    Medidata is the largest provider of software for clinical trials.

    https://cwiki.apache.org/confluence/display/ZIPKIN/Medidata

    View full-size slide

  27. ok it is stock++ Medidata wrote SLAP

    View full-size slide

  28. Besides architecture, what’s different across sites?
    Data collection policy: Typeform always provision request IDs. Infostellar use
    antenna, satellite and plan tags for business context. LINE add company-
    specific tags like phase and instance ID. Expedia Haystack scrubs secrets.
    Data retention policy: Medidata retain 100% for 100 days. Netflix sample
    100% of FIT experiments, 0.1% otherwise. SoundCloud retain a very low sample
    rate for 7 days.
    Tracing adoption rate: LINE is only one team’s services, Ascend <50%, Tyro is
    over 90%

    View full-size slide

  29. How do sites get started with tracing
    Proxy: starting traces at a proxy can raise visibility of upstream and
    downstream. Typeform initialise a trace and request ID in their custom proxy.
    Single service: hotels.com recognised even though tracing is a team sport, starting
    with a single service can still add value.
    New Framework: Sites like Ascend rolled out tracing in new applications as it was
    out-of-box supported with Spring Boot (via Spring Cloud Sleuth).
    Green Field: Infostellar engineers had previous experience with tracing, and built
    their platform with tracing in mind.
    https://cwiki.apache.org/confluence/display/ZIPKIN/Hotels.com

    https://cwiki.apache.org/confluence/display/ZIPKIN/Typeform

    View full-size slide

  30. Wrapping Up
    @adrianfcole
    #zipkin
    introduction
    a typical zipkin site
    wrapping up

    View full-size slide

  31. Wrapping up
    Contribute https://cwiki.apache.org/confluence/display/ZIPKIN/Sites
    Check our “Last Month In Zipkin”
    Chat any time on Gitter
    @adrianfcole
    #zipkin
    gitter.im/openzipkin/zipkin
    github.com/openzipkin/zipkin

    View full-size slide