Tweetovi

Blokirali ste korisnika/cu @willsewell_

Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @willsewell_

  1. 30. sij

    Importing existing infrastructure into Terraform is a painful process, but also a great way of uncovering all the weird inconsistencies and surprises that have probably but up over time.

    Poništi
  2. 29. sij

    Just updated the Rust SDK for Pusher to use hyper 0.13 and switched to an async interface. I really like the idea of the borrow checker, but damn that learning curve is steep for any "real world" code.

    Poništi
  3. 23. sij

    It’s a really specific slice of London I’ve experienced over the years

    Poništi
  4. 19. sij

    We desperately need to replace cars with bicycles, and this seems like an easy win.

    Prikaži ovu nit
    Poništi
  5. 19. sij

    There's a real "market failure" with bike hangers in Waltham Forest. Apparently the waiting list is 3,500. I would happily pay more than the asking price for one. Please build more , or outsource the problem if you are unable to!

    Prikaži ovu nit
    Poništi
  6. 7. sij
    Poništi
  7. 3. sij

    it would awesome if you could work with Prometheus on getting first class support. There has been some interest in this: .

    Prikaži ovu nit
    Poništi
  8. 3. sij

    Either move your apps onto a more flexible platform, or embrace the push based monitoring systems that are already widely used in the Heroku ecosystem.

    Prikaži ovu nit
    Poništi
  9. 3. sij

    This thread makes it clear that Prometheus is a poor fit for the Heroku use-case, but I’m concerned this is not as obvious from the outset as it could be!

    Prikaži ovu nit
    Poništi
  10. 3. sij

    With Prometheus Aggregation Gateway you cannot parameterise the metrics by dyno or process, so you lose the ability to drill down.

    Prikaži ovu nit
    Poništi
  11. 3. sij

    Pushgateway will require you to implement a job to clean up old time series from dead processes. You cannot set a TTL.

    Prikaži ovu nit
    Poništi
  12. 3. sij

    You might be tempted to push metrics to either or even , which Prometheus can scrape. I would not recommend this.

    Prikaži ovu nit
    Poništi
  13. 3. sij

    A fundamental limitation of Heroku is you cannot individually address dynos. The only way of reaching them is through a LB. This makes it a poor fit for a pull-based monitoring system.

    Prikaži ovu nit
    Poništi
  14. 3. sij

    If you’re thinking about monitoring apps with : don’t.

    Prikaži ovu nit
    Poništi
  15. 12. stu 2019.

    Great overview of probabilistic data structures in RedisBloom by at (it’s not just Bloom fillers!) – mjesto: Park Plaza Riverbank Restaurant

    Poništi
  16. 12. stu 2019.

    Great to see the schedule on the back of the lanyard at – mjesto: Park Plaza Riverbank Restaurant

    Poništi
  17. 12. stu 2019.
    Poništi
  18. 12. stu 2019.

    ⁩ introducing redis 6 at London – mjesto: Park Plaza Riverbank Restaurant

    Poništi
  19. 31. lis 2019.
    Poništi
  20. proslijedio/la je Tweet
    11. lis 2019.

    The mathematical modeling parts of my job are the relatively easy parts. The hard part is getting people to agree on the business requirements that the mathematical modeling is being designed to support.

    Prikaži ovu nit
    Poništi

Čini se da učitavanje traje već neko vrijeme.

Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.

    Možda bi vam se svidjelo i ovo:

    ·