William Morgan

@wm

Chief Email Outputter at . Nominal creator. The smallest mime of the snow gods

Austin, TX
Liittynyt heinäkuu 2008

Twiitit

Olet estänyt käyttäjän @wm

Haluatko varmasti nähdä nämä twiitit? Twiittien näyttäminen ei poista käyttäjän @wm estoa.

  1. Kiinnitetty twiitti
    27. elok. 2021

    I wrote a guide to mTLS for Kubernetes users. A bit of a long read. Would love your feedback!

    Näytä tämä ketju
    Kumoa
  2. uudelleentwiittasi
    29. jouluk. 2021

    New blog post! The service mesh in 2022: What's next for Linkerd? by

    Kumoa
  3. 20. jouluk. 2021

    A planned outage for... your service mesh 🤔

    Kumoa
  4. 16. jouluk. 2021
    Kumoa
  5. 15. jouluk. 2021
    Kumoa
  6. 14. jouluk. 2021
    Kumoa
  7. uudelleentwiittasi
    5. jouluk. 2021

    Welcome ! "Web Summit is a proud adopter of Linkerd for all our Kubernetes clusters we use to run conferences. Also in use for the Summit Engine platform that will run CES 2022 in January." 🚀

    Kumoa
  8. 4. jouluk. 2021
    Näytä tämä ketju
    Kumoa
  9. 3. jouluk. 2021

    I always find it a bit tacky when founders brag about sales... but I'll just say that right now at everything is beautiful and nothing hurts 🎈

    Kumoa
  10. 30. marrask. 2021

    My cousin Cathy was recently diagnosed with ALS. If your hasn’t yet come to a close, please consider donating to her cause.

    Kumoa
  11. 22. marrask. 2021

    In my inbox this morning, from an engineer at Twitter. Guess those snarky code comments really paid off in the end. (Also... 10 years! 🙈)

    Kumoa
  12. 15. marrask. 2021
    Tämä twiitti ei ole saatavilla.
    Kumoa
  13. 29. lokak. 2021

    Always love hearing how people get started with . This is a great example of our core strategy: a) get people to evaluate different options; and b) be better than the alternatives. (Weirdly, (a) has been much harder than (b).)

    Kumoa
  14. 27. lokak. 2021

    And while I'm sympathetic to resource consumption, we should address that by shrinking proxies, not moving to a less secure model. IMO the future of cloud native traffic security *must* be built on the workload identity and pod-level enforcement, and not on anything less.

    Näytä tämä ketju
    Kumoa
  15. 27. lokak. 2021

    Plenty of service mesh articles that come through that I roll my eyes at and ignore. This one got my attention because I think eBPF is super cool and I have a ton of respect for the folks that are doing it. But I just don't think this is the right argument.

    Näytä tämä ketju
    Kumoa
  16. 27. lokak. 2021

    The future is also a) keeping network processing code as far away from the kernel as possible, in the safety of userspace; b) only doing network stuff in memory-safe languages; and c) minimizing sidecar resource cost through good ol' fashioned engineering.

    Näytä tämä ketju
    Kumoa
  17. 27. lokak. 2021

    Finally, the conclusion: "eBPF appears to be the natural path for the service mesh data plane." For , at least, this is not the case. The future of Linkerd's data plane is sidecars, workload identity via mTLS, authz from that identity, and keeping the PEP in the pod.

    Näytä tämä ketju
    Kumoa
  18. 27. lokak. 2021

    You also can't build rich policies (e.g. the Foo service can only do GETs on /bar). Can't do hierarchical identity, etc. mTLS is not a panacea (and workload identity is not the ultimate identity) but it is pretty amazing in controlled environments.

    Näytä tämä ketju
    Kumoa
  19. 27. lokak. 2021

    But network encryption can at best do host identity, or network identity, which has little meaning in a cloud native world. It's the difference between "I am " vs "I am someone who lives in Austin" or "I am someone wearing a blue shirt". Who would you trust?

    Näytä tämä ketju
    Kumoa
  20. 27. lokak. 2021

    Why? Encryption is meaningless without authentication. And authentication is only meaningful to the extent that the identities themselves "mean" something. The workload identities provided by mTLS are meaningful—they tell you the service on the other end.

    Näytä tämä ketju
    Kumoa
  21. 27. lokak. 2021

    The article moves on to encryption: "if your only reason for using a service mesh is to provide encryption, you may want to consider network-level encryption". This is presented as an alternative to mTLS. But network layer encryption is *not* an alternative to mTLS.

    Näytä tämä ketju
    Kumoa

Lataaminen näyttää kestävän hetken.

Twitter saattaa olla ruuhkautunut tai ongelma on muuten hetkellinen. Yritä uudelleen tai käy Twitterin tilasivulla saadaksesi lisätietoja.

    Saatat pitää myös

    ·