> On 1 Nov 2019, at 20:12, Pavel Rappo wrote:
>
> Chris,
>
> Which commit is this update based on? I think this is worth mentioning for
> the future maintainers.
> We supplied this information when we made the initial push:
"commit 7207b1b4fb618c0cf17f3391a5e551648c75769d
Author: Viktor Kla
Hi Martin,
On 02/11/2019 16:40, Martin Buchholz wrote:
Hi Julia,
I think this is the wrong fix.
I disagree :-), but I understand why you could think that:
see below.
Looking at the Cache class - all the methods are synchronized, so it
looks like any failure is not actually due to concurrent
I'm still not a network engineer so feel free to disregard my comments.
Only doing local reasoning with the Cache class:
It's a bad sign that it uses two different mechanisms to deal with
concurrency - Cache is synchronized and it also uses a concurrent
collection.
It seems there's a simple bug in
