I mean that would already be passed to your origin in the cf-worker header anyway if that's a reques
I mean that would already be passed to your origin in the cf-worker header anyway if that's a request header rule
DI do the
x-x-
CI think you have most of the useful stuff though, you could grab the edge client port for fun if you wanted too
cf.edge.client_portI'll make them lower case though, good point
Dgood idea, added!
D
Cmetal and colo id can be genuinely helpful at identifying cache stuff
Cwell not really metal, but colo
Dyeah, so for context this is for long term error logging
Dsince Azure support sucks
Kcf.verified_bot_categoryweird, I got
"x-client-port": "0",
HIs there a
verified_bot_name
Cdid you already have a connection established?
Knop
DD^ example of part of the stuff I'm logging
DI opened one with curl
HDarn
HDo you have to have a dedicated UA, or just not spoof someone else's?
HBecause I feel like I've seen some crawlers with pretty generic UAs
Dyeah, already logging
verifiedBotCategorycfYeah, so if you have IP verify + regex, then you may have two crawlers with similar/same UA, that are differentiated by their IPs, no?
Danything else? 
Dhang on, isn't this pointless since I already log the request headers in my Worker?
CDoh derp
Dis there any situation where it could change or be helpful?
DI'll remove it for now
Deasier to parse without concat'ing, as I'd be worried that that field might be missing.
DI think it's just parsing that I was worried about
Dis there a way to pad http.request.timestamp.msec to include the the first two leading zeros?
Dyeah
DI feel like I got annoyed when I thought of this last time lol
Dehhhhhhhhhhhhhh
DI know we have some Workers that already use
timestamp-msec-only
BGoogle reCAPTCHA no longer provides 1 million free assessments.
Turnstile time.
Turnstile time.
Uthat is until u hit over 1m and recaptcha is cheaper 
is interesting though very large downgrade in the free
is interesting though very large downgrade in the freeBJokes on you, I'd never hit that! 
BAnd yea, that was my only reason for using it))
UI guess actually if u use managed mode turnstile is unlimited free forgot they did that
Byeah, was just about to say that! i also remembered them being free & unlimited so I checked
Bcloudflare is awesome
UCan still use other modes limited at 1 million I think but I assume managed mode is fine for 99% of people so 
Dah,
client_port is 0
Dwhat's wrong with managed mode? Can't you sorta use it like non-interactive if you really want to anyway?
UI assume managed mode is fine for 99% of people so
Dyeah but even if it isn't, can't you just put a managed in an offscreen area of the screen, and then use the callbacks?
II have a very simple worker that takes in a url, fetches it, adds a 5 second TTL, and then returns a response. when I spam refresh in my browser, I only get 1 request to the origin per 5 seconds (correct). when I do the same request spam using curl, multiple origin hits happen. anybody know what's going on?
UCache is per datacenter and your browser will stick generally to 1 datacenter (though it can switch) and curl will not. A single colo can also have multiple datacenters. So just because you only hit "DFW" doesnt mean you will hit the same cache.
Ihuh interesting. that makes sense, but wouldn't copying the browser request as curl be as close to a 1:1 as possible?
