Though I would note too that Vercel Edge is Workers too
Though I would note too that Vercel Edge is Workers too
HSo for any non-platform-bound applications, they should perform about the same
✦nop it's aws afaik
HThat’s their core(I.E. non-Edge). That’s also why they are so particular about the APIs that the Edge Runtime does and doesn’t support
HBecause the Edge Runtime(WorkerD) is V8, but not Node
Hleerob
Vercel's Edge Functions are built on top of Cloudflare Workers. The runtime is designed for any similar provider, but we've chosen Cloudflare for now because they have an amazing product.
Hacker News
•10/26/21, 5:41 PM
✦thx for sharing, reading now
✦okay let me try the worker with vercel kv in this case and see if it's better
✦how to get around this? Uncaught ReferenceError: process is not defined
cloudflare workers don't use node right?
cloudflare workers don't use node right?
✦nvm, followed upstash cloudfare quickstart
✦you're right, switched to upstach. Cloudflare KV is very very slow
✦still aw seems to be faster for now;
Uyou show 2 responses for the other platform one of which is faster one of which is slower so 
CWhat are you measuring? Puts or Gets?
https://developers.cloudflare.com/kv/learning/how-kv-works/
You can signal for it to cache results longer as well, and stream back responses, ex
(It is eventually consistent by design, but by getting it to cache for longer you're not going to see updates until cache expires or if you write in same location. (KV 2.0 which was delayed aimed to solve this by checking every minute in the background for updates, but it was delayed due to issues))
It's pretty fast, once its cached locally ~20ms or less
should've gone for foundation dbThe magic with KV is its unlimited scale and getting faster the more requests you give it, it's made for read-heavy cases, like for example Cloudflare Pages uses it
https://developers.cloudflare.com/kv/learning/how-kv-works/
You can signal for it to cache results longer as well, and stream back responses, ex
(It is eventually consistent by design, but by getting it to cache for longer you're not going to see updates until cache expires or if you write in same location. (KV 2.0 which was delayed aimed to solve this by checking every minute in the background for updates, but it was delayed due to issues))
It's pretty fast, once its cached locally ~20ms or less
✦understood thx chaika
DWhat are your preferred ways of securely accessing a Worker route from an AWS Lambda? I’ve done the reverse before, undecided on what makes the most sense for the other way around.
Sgood old secret key in the headers 
AHey,
I'm using Workers to push some logic to the request and proxying the request to the origin backend
When combining that with the Dynamic URL Rewrite Transform Rule it seems like the Transform Rule is triggered two times - from the client to the worker and when proxying the request
In my Dynamic Transform Rule, I used -
I tried to adjust the Custom Filter Expression with no success (skipping the rule every time the URL starts with
Any idea how to overcome this behavior?
Thanks
(also posted to the community - https://community.cloudflare.com/t/proxyrequest-in-workers-triggers-transform-rules-more-than-once/557777)
I'm using Workers to push some logic to the request and proxying the request to the origin backend
When combining that with the Dynamic URL Rewrite Transform Rule it seems like the Transform Rule is triggered two times - from the client to the worker and when proxying the request
In my Dynamic Transform Rule, I used -
concat(“/v1”, http.request.uri.path) and ended up with /v1/v1/<original_req_path>I tried to adjust the Custom Filter Expression with no success (skipping the rule every time the URL starts with
/v1 works but doesn't feel robust enough)Any idea how to overcome this behavior?
Thanks
(also posted to the community - https://community.cloudflare.com/t/proxyrequest-in-workers-triggers-transform-rules-more-than-once/557777)
JHi folks,
I have worker route in my site example (miton.com)::
route value us wildcard / and it is pointed to some worker
Here all the requested from any domain or wildcard domain connected with this site are coming to worker route
I need to by pass some of domain example "javi.milton.com" not to go to worker route
Because i have pointed "javi.militon.com" to R2 bucket
I have worker route in my site example (miton.com)::
route value us wildcard / and it is pointed to some worker
Here all the requested from any domain or wildcard domain connected with this site are coming to worker route
I need to by pass some of domain example "javi.milton.com" not to go to worker route
Because i have pointed "javi.militon.com" to R2 bucket
JI have create new route for "javi.milton.com" and disabled route withour choosing any workers
Jbut still no luck, it is keep pointing to / --> workers
KYou can use Unbound, which gives you up to 30 seconds of CPU time.
Dif you only do hashing on the client, you're essentially storing the "raw" password, just transformed, which I don't recommend
Dall hashing algorithms transform the password, and the theory behind it is interesting, but it's different
Da malicious actor that knows the clienthash can get access, vs if you hash on the server, you still can't do anything if you know the hash
D(if set up properly, secure algorithm, salt, etc etc)
Dwhat if your db is a file on a linux computer, like /etc/passwd
Dno
Dthanks, although this is nothing about simplicity or implementation, it's just how cryptography works, I recommend the Katz and Lindell "Introduction to modern cryptography" book if you'd like to see why hashing on the client does nothing for security
ZHi, just purchase domain through cloudflare. But now my website have error
Sbad take
ZThe issue is that if you only hash client side the attackers will immediately be able to log in to your client's accounts once they have access to the login credentials, while if you hash server side the attacker still has to brute force the hash first.
