Profiler topology zookeeper client timed out

[Gonçalo Pedras <go...@ctd.pt>]

Hi,

I've set up a profiler of 15 minutes period, although it isn't really 15 minutes long. I've been monitoring the topology in Storm UI and I reached the conclusion that when my profiler flushes, after a few minutes the worker restarts. I checked the logs and I found that the worker restarts when the Client session times out:
"2020-03-09 17:11:02.923 o.a.s.s.o.a.z.ClientCnxn main-SendThread(XXXXXXXXXXXXXXXXXXXXXX) [INFO] Client session timed out, have not heard from server in 21341ms for sessionid 0x170aa100eba035e, closing socket connection and attempting reconnect"
And when I checked it against the zookeeper logs, it shows that the session expired:
"2020-03-09 17:11:36,003 - INFO  [SessionTracker:ZooKeeperServer@357] - Expiring session 0x170aa100eba035e, timeout of 30000ms exceeded"
"2020-03-09 17:11:36,003 - INFO  [ProcessThread(sid:3 cport:-1)::PrepRequestProcessor@491] - Processed session termination for sessionid: 0x170aa100eba035e"

Therefore my profiler doesn't flush every 15 minutes because when it restarts I have to wait 15 minutes + (plus) the time it wasted after the previous flush.

What should I do?

Thanks