100GB of cached files sounds enormous. What kinds of files are you caching? How large are they? How many do you have?
If you look at your access log what hit rate is your cache seeing?

Sent from my iPad

> On Feb 16, 2018, at 3:16 AM, Andrzej Walas <nginx-forum@forum.nginx.org> wrote:
>
> After this inactive logs I have anther logs:
> [alert] 11371#0: worker process 24870 exited on signal 9
>
> Can you explain whats happend when I have logs like this:
> [error] 9045#0: *1138267 readv() failed (104: Connection reset by peer)
> while reading upstream...
>
> I have problem like https://forum.nginx.org/read.php?2,278534 too. Max size
> is max_size=40g but offen I have more than 100 GB cache on disc and problem
> with empty space to download new files to cache.
>
> Posted at Nginx Forum: https://forum.nginx.org/read.php?2,278589,278611#msg-278611
>
> _______________________________________________
> nginx mailing list
> nginx@nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx
_______________________________________________
nginx mailing list
nginx@nginx.org
http://mailman.nginx.org/mailman/listinfo/nginx