to test how would the restarting process would work. As a test environment I'm using a docker-compose file, based on this docker-compose file:
. I've tried changing the envoy command to the following
This is the output of the logs I'm seing (I don't see much difference by changing the level to trace)
envoy_1 | [2018-03-17 13:24:16.315][5][info][main] source/server/server.cc:178] initializing epoch 0 (hot restart version=9.200.16384.127.options=capacity=16384, num_slots=8209 hash=228984379728933363)
envoy_1 | [2018-03-17 13:24:16.322][5][info][config] source/server/configuration_impl.cc:52] loading 1 listener(s)
envoy_1 | [2018-03-17 13:24:16.323][5][info][config] source/server/configuration_impl.cc:92] loading tracing configuration
envoy_1 | [2018-03-17 13:24:16.323][5][info][config] source/server/configuration_impl.cc:119] loading stats sink configuration
envoy_1 | [2018-03-17 13:24:16.324][5][info][main] source/server/server.cc:353] starting main dispatch loop
envoy_1 | [2018-03-17 13:24:16.326][5][info][upstream] source/common/upstream/cluster_manager_impl.cc:131] cm init: all clusters initialized
envoy_1 | [2018-03-17 13:24:16.326][5][info][main] source/server/server.cc:337] all clusters initialized. initializing init manager
envoy_1 | [2018-03-17 13:24:16.326][5][info][config] source/server/listener_manager_impl.cc:583] all dependencies initialized. starting workers
envoy_1 | starting hot-restarter with target: /usr/local/bin/start_envoy.sh
envoy_1 | forking and execing new child process at epoch 0
envoy_1 | forked new child process with PID=5
envoy_1 | got SIGHUP
envoy_1 | forking and execing new child process at epoch 1
envoy_1 | forked new child process with PID=18
envoy_1 | got SIGCHLD
envoy_1 | PID=18 exited with code=1
envoy_1 | Due to abnormal exit, force killing all child processes and exiting
envoy_1 | force killing PID=5
envoy_1 | exiting due to lack of child processes
envoytest_envoy_1 exited with code 1