Objects overflows happen on the Observers due to bursty sources, the app user may click on the views very quickly so the Observers don’t get enough time to consume the emitted objects because of this from RxJava 2.x to have enough room for all values, programmers are allow to specify the size of their internal buffers while using the backpressure-sensitive operators with relevant parameters: bufferSize, prefetch or capacityHint

xxxxxxxxxx

PublishProcessor<Integer> source = PublishProcessor.create();

​

    source.observeOn(Schedulers.computation(), 1024 * 1024)

          .subscribe(e -> { }, Throwable::printStackTrace);

​

    for (int i = 0; i < 1_000_000; i++) {

        source.onNext(i);

    }

https://medium.com/@ajay.dewari/backpressure-in-reactivex-rxjava-and-difference-between-observable-and-flowable-6074c25234ea