Re: [rsyslog] Interesting behaviour of random()

2024-02-07 Thread Mariusz Kruk via rsyslog
What is even more interesting - I can't reproduce that behaviour on my laptop with 8.2402 on Fedora 38. Either there is some change in either the behaviour of rsyslog itself or the linked libraries (the 8.2302 was on SUSe). Or there is something else in my config (that random-picking code was

Re: [rsyslog] Rsyslog stop working when output destination (rabbitmq) is unavialable

2024-02-07 Thread David Lang via rsyslog
just a note that a batch size of 4000 doesn't make a lot of sense when the queue size is only 4096 :-) batch sizes are a case of diminishing returns. There is a huge advantage of going from to around 100 or so, but far less gains in going from 100 to 1000 (unless you are doing database inserts