If you can manage more than 20,000 builds a month I'd be impressed
If you can manage more than 20,000 builds a month I'd be impressed
Jthis would be avoiding pages and just using a worker from my understanding, so wouldn't count towards those limits
KO - just saw the word Pages and "tons of unique environments" on the same line haha
KMakes sense
WIt should be most specific route wins yes but I'd need to check. You can also add middleware with Functions with is cool.
KNada - Workers will follow the usual anycast Cloudflare IP's and get spun up in the 'nearest' colo to the user
IThat's holds major abuse potential though.
IOne of the main reasons Workers work so well is that they are distributed
IIf something like that were to be added then attackers could selectively take down CF datacenters by just sending waves of Worker requests.
IOr a customer, expecting a Worker request to be in one DC, might have their app fail because that DC is busy or rerouted.
Ais the option to change the worker metrics time period in the dashboard gone?
IGitHub
A library to easily send request events and traces to https://honeycomb.io
- GitHub - cloudflare/workers-honeycomb-logger: A library to easily send request events and traces to https://honeycomb.io
Odoes the websocket pair automatically ping? im using aiohttp from python to connect with it and its reporting that (client side) its getting an abnormal closure
ZHm, I'm finding in workers logs errors like this:
"exceptions": [
{
"name": "TypeError",
"message": "Cannot read properties of undefined (reading 'get')",
"timestamp": 1647996540375
}
],
event request shows:
{
"url": "http://domain/",
"method": "HEAD",
"headers": {
"accept": "text/html,application/xhtml+xml,application/xml;q=0.9,/;q=0.8",
"accept-charset": "ISO-8859-1,UTF-8;q=0.7,*;q=0.7",
"accept-encoding": "gzip",
"accept-language": "en-US,en;q=0.8",
"cache-control": "no-cache",
"cf-connecting-ip": "69.162.124.235",
"cf-ipcountry": "US",
"cf-ray": "6f032a281b2dfed2",
"cf-visitor": "{"scheme":"http"}",
"connection": "Keep-Alive",
"content-length": "0",
"host": "domain",
"referer": "http://domain,
"user-agent": "Mozilla/5.0+(compatible; UptimeRobot/2.0; http://www.uptimerobot.com/)",
"x-forwarded-proto": "http",
"x-real-ip": "69.162.124.235"
},
"exceptions": [
{
"name": "TypeError",
"message": "Cannot read properties of undefined (reading 'get')",
"timestamp": 1647996540375
}
],
event request shows:
{
"url": "http://domain/",
"method": "HEAD",
"headers": {
"accept": "text/html,application/xhtml+xml,application/xml;q=0.9,/;q=0.8",
"accept-charset": "ISO-8859-1,UTF-8;q=0.7,*;q=0.7",
"accept-encoding": "gzip",
"accept-language": "en-US,en;q=0.8",
"cache-control": "no-cache",
"cf-connecting-ip": "69.162.124.235",
"cf-ipcountry": "US",
"cf-ray": "6f032a281b2dfed2",
"cf-visitor": "{"scheme":"http"}",
"connection": "Keep-Alive",
"content-length": "0",
"host": "domain",
"referer": "http://domain,
"user-agent": "Mozilla/5.0+(compatible; UptimeRobot/2.0; http://www.uptimerobot.com/)",
"x-forwarded-proto": "http",
"x-real-ip": "69.162.124.235"
},
Zbut can't repeat that, tried with curl -I/X HEAD, tried request with copied all the same headers - when I execute that request locally it works
Z(other thing is I have sentry attached so I would expect exceptions to be catched to sentry, not only CF worker logs)
Zcould headers be empty? I'll try changing few lines headers.get to headers?.get and see then, thanks for idea @ptim !
Zincomming, I'm having proxy worker that proxies some request to legacy platform, other to new platform
Zand it works like a charm, except these HEAD requests from uptimerbot.com 
Zpublished handy pattern, will see in few minutes if this error still happen
Zstill a problem. I'm mostly suprised I can't repeat that by copying request from CF logs
ZIf I could repeat that in some curl or postman request it would be much easier to fight, now I have to wait 5 minutes for next uptimerbot.com request 
Zhmm, I have try,catch in functions that are called from listener, I'll move it up
Nhi everyone. I have a question regarding wrangler and worker environments. My wrangler.toml looks like this:
however, when we do it creates a new worker in the cloudflare UI called "worker-staging" with an environment of "production" . Are the "environment" settings on the workers within the clouflare UI not the same as the environments defined in wrangler?
however, when we do it creates a new worker in the cloudflare UI called "worker-staging" with an environment of "production" . Are the "environment" settings on the workers within the clouflare UI not the same as the environments defined in wrangler?
Nfor reference - this is the UI component I'm referring to
Zyeah I also discovered environment dropdown in CF web panel and was wondering how it works - maybe name of worker should be the same?
N@zapalona_swieczka wrangler wont allow that 
Zah
Zthen I'm curious too
Nah - is there a wrangler 2 release date or a link to said API services 
Znote - not production recommended
Zgreat, happy to give it a go whenever it's marked as more less stable
N@a16g is there any reference for what the new wrangler.toml config should look like in order to take advantage of the new environment service