https://discord.cloudflare.com logo
Join Discord
Powered by
# durable-objects
  • a
    seems to change a lot from this location
  • a
    Copy code
    FRA
OTP
PRG
DUS
ATH
ZAG
BEG
  • a
    Currently, durable objects are available in 39% of Cloudflare PoPs
    - it was 44% when I started 😆
  • j
    denominator+++++++
  • a
    Alright, I "managed" to bring the percentage down to 37. Won't spend anymore time on this 😅
  • j
  • j
    (tho should’ve handled that more gracefully)
  • h
    How'd I not noticed that? I needed the locations of the datacenters for a project, so I scraped them off the statuspage then geocoded them.
  • v
    Me too 😅
  • v
  • j
    i did the same until i found that endpoint!
  • j
    good thing workers makes scraping so easy…
  • v
    Actually no - the coords in that list are literally the airport coords which makes the map look rather weird zoomed in. I'll keep my reverse geocoder.
  • h
    Does your Latency Map test latency to your datacenters, or the actual Colos?
  • v
    it's measuring roundtrip latency client – worker – DO – worker – client using a websocket connection (as we do for our "real" apps)
  • h
    And how do you figure out which colo the DO is in?
  • v
  • h
    Should have foreseen that.
  • v
    😁
  • v
    Hmm, curious. One of my DOs moved from LAX to SJC today 🤔 Is that supposed to happen?
    Copy code
    2021-08-04 18:34:58.913 PDT DO resumed in LAX via worker in GRU
2021-08-05 01:46:26.744 PDT DO resumed in LAX via worker in FRA
2021-08-05 03:22:13.610 PDT DO resumed in SJC via worker in FRA
2021-08-05 16:55:44.912 PDT DO resumed in SJC via worker in LAX
  • h
    Might have been me...
  • h
    Sorry about that, @Vanessa🦩!
  • v
    Huh? How? Supposedly DOs are fixed to one colo, unless that colo is down, I thought
  • h
    I believe that DOs persist in one colo, until they are shut down. Once they are triggered again, they are reinitialized in the nearest DO-ready colo to the incoming request.
  • v
    The way @User described it is that it lives in one colo, but is replicated to a few near colos for backup. It would not "jump" from the EU to the US.
  • h
    According to your log above, it seems that because there was an extended period of time with no requests, it was terminated(FRA). Once someone else triggered it(Me, probably), it was reestablished in the nearest Colo(SJC) to my request.
  • h
    At least I believe so, I'm having a hard time figuring out that log.
  • v
    LAX being down for a while would make sense, but I'd have assumed it would move back to LAX once its up again.
  • h
    Not if the next request comes from somewhere else, before it can reestablish itself there.
  • v
    there were like 13 hours in between
1...139140141...567Latest