> Pattern-matching website in body, pattern-matching website in title ---------- Title - Position 1-20: supportnumberusa.us Body - Position 1-20: supportnumberusa.us
the only time I've ever had to use more than one worker is when someone told me I had to scale to 1k simultaneous users
just keep speed optimization in the back of your mind while writing the thing, and it'll be fast enough to run on one worker for any conceivable workload that it's going to get
Rather than looking up various configuration values in the DB whenever we're sending something off to the bot or posting a message I'm having it build a function with the logic hardcoded
But since these are functions I was going to keep it in a class variable
@ArtOfCode Actually I just remembered that I was going to have the requests done with a pool of processes anyways. So I can message them to rebuild. The real trouble now is how to get the workers to share a pool...
@micsthepick That pattern looks like it's already caught by Bad keyword in body and Bad keyword in answer; append -force if you really want to do that.
> Pattern-matching website in body, pattern-matching website in title, pattern-matching website in username ---------- Title - Position 17-36: upportaustralia.com Body - Position 17-36: upportaustralia.com Username - Position 17-36: upportaustralia.com
@TetsuyaYamamoto That pattern looks like it's already caught by Pattern-matching website in body and Pattern-matching website in answer; append -force if you really want to do that.
> Bad keyword in body, bad keyword in title, bad keyword in username ---------- Title - Position 1-15: Support Number Body - Position 1-15: Support Number Username - Position 1-15: Support Number
@ArtOfCode bigger problem: I can't even figure out a way to start a thread before the workers fork off (if I do it afterwards, it'll spawn one per worker) that works everywhere. Puma and Unicorn have before_fork, but not Passenger
it's weird how something so fundamental isn't even given the time of day; I googled and found people asking about this and everyone said to have a separate application that runs ActiveRecord standalone which is kinda absurd
This is just about spawning one thread to listen to the SE websocket
Only certain servers let you run code before it forks off a bunch of workers, and if you have the thread spawn in a piece of code every worker runs (like say boot.rb) then you have one thread per worker
@quartata so either (a) assume you're running on Puma; (b) don't use multiple worker processes, use multiple worker threads, and set a variable somewhere when you've initialized the listener from one thread.
DB is for data persistence, using it for application-level signaling feels wrong
Python and Ruby have what's called a global interpreter lock. They use kernel threads, but they all try to acquire this one reentrant lock so only one thread can run at a time.
That's why they use processes
At any rate I don't see a whole lot of difference between using the DB and Redis or shared memory. It's not so bad
Yeah, so AC has different adapters. By default it uses async in dev and test, and redis on prod. async has never actually worked for me; I've always had to set dev to redis to make it actually work.
Claims to be a Java programmer, but I see no Java in his 4 non-empty repos on front page. Three of them contains a (probably) templated code and the last one contains a .tar.gz
Anyone got the SmokeDetector swag mail? has a google form link. But for some reason giving me a warning popup "This link has been flagged as suspicious."...
[ SmokeDetector | MS ] URL-only title, bad keyword in body, blacklisted website in body, blacklisted website in title, link at end of body, +1 more: mumybear.com/luxury-lean/ by Luxury Lean on askubuntu.com
