Closed i-chernyshov closed 5 months ago
i found that the second problem can be solved by turning the singleton binding 'log' into scoped:
/vendor/laravel/framework/src/Illuminate/Log/LogServiceProvider.php
PR didn't work for me
i can do it myself, but how many places in laravel should i still think about? is this production-ready? could it be the right solution to redefine all singleton bindings as scoped?
Hi there,
Thanks for reporting but it looks like this is a question which can be asked on a support channel. Please only use this issue tracker for reporting bugs with the library itself. If you have a question on how to use functionality provided by this repo you can try one of the following channels:
However, this issue will not be locked and everyone is still free to discuss solutions to your problem!
Thanks.
@driesvints
Hey, why did you close the problem? I have 3 important questions about the work of octane, could you show more respect to the consumer and give advice?
These questions can be asked on a support channel. We generally only handle bug reports on this issue tracker.
@driesvints
you just threw me over to try my luck somewhere else, even though I clearly attached two problems, two obvious errors and a potential security issue. thanks)
could you give a couple of comments about this?
Could you help me understand what the problem is?
what could be the reason for such unexpected behavior?
These are support questions. This is open source, not free support.
Octane Version
2.3.2
Laravel Version
10.42.0
PHP Version
8.1
What server type are you using?
Open Swoole
Server Version
22.1.2
Database Driver & Version
(PostgreSQL) 15.4 (Debian 15.4-2.pgdg120+1)
Description
hello! I have combined 3 of my current problems into one thread
first: security warnings, could you give a couple of comments about this?
second: in monolog, I use my custom driver and handler (inherited from Abstract ProcessingHandler), thanks to which I collect records in the terminate method of my middleware and send them to storage. they are quite simple and I have limited rights, so I can't attach the full code. so the problem is that records is filled only when processing the very first request after the start of the worker, then in all subsequent requests the write method is called, but records always turns out to be empty. commenting out FlushMonologState does not help. Could you help me understand what the problem is?
third: sometimes I just get an error. it works stably for 2/3 of the time. what could be the reason for such unexpected behavior?
Thanks!
Steps To Reproduce
first problem: use a security checker
second problem: periodically occurs on its own
third problem: create your own custom handler like this
throw several exceptions when processing a request in terminated middleware, try to get GetRecords, they will only be available when processing the first request after the start of the worker