Description
For the test (not for production), a cluster with one node on single host is deployed.
With the periodicity of 10-20 minutes, all the containers are restarted.
In dockerd log:
Jan 16 21:54:36 swarm-test1.jopa.xxx dockerd[22095]: time="2018-01-16T21:54:36.595246162+03:00" level=info msg="NetworkDB stats swarm-test1.jopa.xxx(03e47b90a3c4) - netID:nr31u59rzjus967xxj9og4hsx leaving:false netPeers:1 entries:12 Queue qLen:0 netMsg/s:0"
Jan 16 21:54:36 swarm-test1.jopa.xxx dockerd[22095]: time="2018-01-16T21:54:36.595403674+03:00" level=info msg="NetworkDB stats swarm-test1.jopa.xxx(03e47b90a3c4) - netID:vujhpe4ba7kasg5xrr64gknd7 leaving:false netPeers:1 entries:16 Queue qLen:0 netMsg/s:0"
Jan 16 21:54:36 swarm-test1.jopa.xxx dockerd[22095]: time="2018-01-16T21:54:36.597752278+03:00" level=debug msg="heartbeat expiration"
Jan 16 21:54:36 swarm-test1.jopa.xxx dockerd[22095]: time="2018-01-16T21:54:36.635619893+03:00" level=error msg="agent: session failed" error="node not registered" module=node/agent node.id=entvk2h3x7tkb7uqwhu5bebx4
Jan 16 21:54:36 swarm-test1.jopa.xxx dockerd[22095]: time="2018-01-16T21:54:36.635847522+03:00" level=debug msg="agent: rebuild session" module=node/agent node.id=entvk2h3x7tkb7uqwhu5bebx4
Jan 16 21:54:36 swarm-test1.jopa.xxx dockerd[22095]: time="2018-01-16T21:54:36.696474740+03:00" level=debug msg="node status updated" method="(*Dispatcher).processUpdates" module=dispatcher node.id=entvk2h3x7tkb7uqwhu5bebx4
Jan 16 21:54:36 swarm-test1.jopa.xxx dockerd[22095]: time="2018-01-16T21:54:36.700151162+03:00" level=debug msg="global orchestrator: node entvk2h3x7tkb7uqwhu5bebx4 is down, shutting down its tasks" module=node node.id=entvk2h3x7tkb7uqwhu5bebx4
Jan 16 21:54:36 swarm-test1.jopa.xxx dockerd[22095]: time="2018-01-16T21:54:36.707573026+03:00" level=debug msg="(*session).start" module=node/agent node.id=entvk2h3x7tkb7uqwhu5bebx4
Issue here https://github.com/moby/moby/issues/36031