After long run zipkin will not reply

[mumutu]

I am using the default configuration to start the image only expose the inner port with -p 9411:9411

and in every client use AsyncReporter.builder(URLConnectionSender.create(url)).build()); and At the very beginning it worked well， but two weeks later the zipkin server would not reply though the docker instance was still running

here is the docker stats：

CONTAINER CPU % MEM USAGE / LIMIT MEM % NET I/O BLOCK I/O PIDS 17b6255c8518 0.08% 905.4 MiB / 3.703 GiB 23.87% 235 MB / 78.7 MB 142 MB / 0 B 0

it occured several times， at first I thought it was the low configuration of the pc to blame， but after move the instance to a new one this error still turn up

。unfortunately I am still new to this，so am I missing some start configuration to fit the production environment?

[codefromthecrypt]

are you running zipkin against a storage backend (like mysql) or are you leaving it default (in-memory)

if using in-memory (our testing option), zipkin will eventually run out of memory https://github.com/openzipkin/zipkin/issues/1528.

[mumutu]

Yep I am using the in-memory storage. So even the used mem not reached limit mem , there still has a chance the server will down (docker stats showed the mem usage is still within the limits) ? Whatever really thanks! I will use a mysql storage backend to fit the production environment!

[codefromthecrypt]

good point, so the memory that is in contention is that in the java process.