Describe the bug
BasicCrawler:AutoscaledPool: state {"currentConcurrency":0,"desiredConcurrency":100,"systemStatus":{"isSystemIdle":true,"memInfo":{"isOverloaded":false,"limitRatio":0.2,"actualRatio":0},"eventLoopInfo":{"isOverloaded":false,"limitRatio":0.6,"actualRatio":0},"cpuInfo":{"isOverloaded":false,"limitRatio":0.4,"actualRatio":0},"clientInfo":{"isOverloaded":false,"limitRatio":0.3,"actualRatio":0}}}
I'm checking if there is any possibility of this "deadlock" in our code but seems that not. Also if the code would lock, it would not have concurrency 0.
The scraper was between 50-100 concurrency, then it dropped to 2, to 1 to 0 and died there.
Reading the super-complex autoscaled pool code gives me impression that it is really hard to prove there isn't a race condition. Maybe that is something we have to live with with unless someone magically refactors it
To Reproduce
Not easily reproducible, the same scraper doesn't do this.
System information:
SDK 1.1.2 on the platform
Describe the bug
BasicCrawler:AutoscaledPool: state {"currentConcurrency":0,"desiredConcurrency":100,"systemStatus":{"isSystemIdle":true,"memInfo":{"isOverloaded":false,"limitRatio":0.2,"actualRatio":0},"eventLoopInfo":{"isOverloaded":false,"limitRatio":0.6,"actualRatio":0},"cpuInfo":{"isOverloaded":false,"limitRatio":0.4,"actualRatio":0},"clientInfo":{"isOverloaded":false,"limitRatio":0.3,"actualRatio":0}}}
I'm checking if there is any possibility of this "deadlock" in our code but seems that not. Also if the code would lock, it would not have concurrency 0.
The scraper was between 50-100 concurrency, then it dropped to 2, to 1 to 0 and died there.
Reading the super-complex autoscaled pool code gives me impression that it is really hard to prove there isn't a race condition. Maybe that is something we have to live with with unless someone magically refactors it
To Reproduce
Not easily reproducible, the same scraper doesn't do this.
System information:
SDK 1.1.2 on the platform