diff --git a/etcd.go b/etcd.go
index c08de232..c8fce194 100644
--- a/etcd.go
+++ b/etcd.go
@@ -1618,6 +1618,7 @@ func (obj *EmbdEtcd) StartServer(newCluster bool, peerURLsMap etcdtypes.URLsMap)
 	cfg.APUrls = peerURLs
 	cfg.LCUrls = obj.clientURLs
 	cfg.LPUrls = peerURLs
+	cfg.StrictReconfigCheck = false // XXX: workaround https://github.com/coreos/etcd/issues/6305
 
 	cfg.InitialCluster = initialPeerURLsMap.String() // including myself!
 	if newCluster {
diff --git a/test/shell/t8.sh b/test/shell/t8.sh
index 6db30e3b..d159fb72 100755
--- a/test/shell/t8.sh
+++ b/test/shell/t8.sh
@@ -3,20 +3,20 @@
 # run empty graphs, we're just testing etcd clustering
 timeout --kill-after=120s 90s ./mgmt run --hostname h1 --tmp-prefix &
 pid1=$!
-sleep 5s	# let it startup
+sleep 10s	# let it startup
 
 timeout --kill-after=120s 90s ./mgmt run --hostname h2 --seeds http://127.0.0.1:2379 --client-urls http://127.0.0.1:2381 --server-urls http://127.0.0.1:2382 --tmp-prefix &
 pid2=$!
-sleep 5s
+sleep 10s
 
-$(sleep 5s && kill -SIGINT $pid2)&	# send ^C to exit 2nd mgmt
+$(sleep 10s && kill -SIGINT $pid2)&	# send ^C to exit 2nd mgmt
 wait $pid2
 e=$?
 if [ $e -ne 0 ]; then
 	exit $e
 fi
 
-$(sleep 5s && kill -SIGINT $pid1)&	# send ^C to exit 1st mgmt
+$(sleep 10s && kill -SIGINT $pid1)&	# send ^C to exit 1st mgmt
 wait $pid1	# get exit status
 # if pid1 exits because of a timeout, then it blocked, and this is a bug!
 exit $?