about install etcd on DCOS 1.8.7

[wenhongqiang]

I am trying install kubernetes on DCOS 1.8.7. and follow https://docs.mesosphere.com/wp-content/uploads/pdf/manage-service/kubernetes.pdf guide to do that.

but when I installed etcd. It's showing idle status. and I checked the etcd services, it always restarting. below is part log

2016-11-30 06:56:24.433316 I | Connected to master.mesos:2181 2016-11-30 06:56:24.434559 I | Authenticated: id=97013182613227278, timeout=5000 2016-11-30 06:56:24.436730 I | Recv loop terminated: err=EOF I1130 06:56:24.436764 12 app.go:226] Found stored framework ID in Zookeeper, attempting to re-use: 8205f618-9c84-48b9-87d8-388f1b065ab6-0016 2016-11-30 06:56:24.438333 I | Send loop terminated: err= I1130 06:56:24.440442 12 scheduler.go:212] found failover_timeout = 168h0m0s I1130 06:56:24.440516 12 scheduler.go:334] Initializing mesos scheduler driver I1130 06:56:24.440615 12 scheduler.go:833] Starting the scheduler driver... I1130 06:56:24.440726 12 http_transporter.go:383] listening on 10.0.3.150 port 8380 I1130 06:56:24.440772 12 scheduler.go:850] Mesos scheduler driver started with PID=scheduler(1)@10.0.3.150:8380 I1130 06:56:24.440791 12 scheduler.go:862] starting master detector zoo.MasterDetector: &{cancel:0x79dba0 client: done:0xc20805a7e0 ignoreInstalled:0 leaderNode: minDetectorCyclePeriod:1000000000 bootstrapLock:{w:{state:0 sema:0} writerSem:0 readerSem:0 readerCount:0 readerWait:0} bootstrapFunc:0x79dbb0} I1130 06:56:24.440871 12 scheduler.go:1053] Scheduler driver running. Waiting to be stopped. I1130 06:56:24.440886 12 scheduler.go:677] running instances: 0 desired: 3 offers: 0 I1130 06:56:24.440899 12 scheduler.go:685] PeriodicLaunchRequestor skipping due to Immutable scheduler state. I1130 06:56:24.440910 12 scheduler.go:1050] Admin HTTP interface Listening on port 8379 2016-11-30 06:56:24.443066 I | Connected to master.mesos:2181 2016-11-30 06:56:24.444182 I | Authenticated: id=97013182613227279, timeout=40000 I1130 06:56:24.445568 12 scheduler.go:419] New master master@10.0.7.189:5050 detected I1130 06:56:24.445599 12 scheduler.go:483] No credentials were provided. Attempting to register scheduler without authentication. I1130 06:56:24.445644 12 scheduler.go:976] Reregistering with master: master@10.0.7.189:5050 I1130 06:56:24.445809 12 scheduler.go:933] will retry registration in 888.186136ms if necessary I1130 06:56:24.447773 12 scheduler.go:810] Handling framework error event. I1130 06:56:24.447810 12 scheduler.go:1141] Aborting framework [&FrameworkID{Value:8205f618-9c84-48b9-87d8-388f1b065ab6-0016,XXX_unrecognized:[],}] I1130 06:56:24.447865 12 scheduler.go:1122] stopping messenger I1130 06:56:24.447887 12 messenger.go:295] stopping messenger scheduler(1)@10.0.3.150:8380.. I1130 06:56:24.447910 12 http_transporter.go:447] stopping HTTP transport I1130 06:56:24.447921 12 scheduler.go:1125] Stop() complete with status DRIVER_ABORTED error I1130 06:56:24.447983 12 http_transporter.go:421] HTTP server stopped because of shutdown 2016-11-30 06:56:24.448906 I | Recv loop terminated: err=EOF 2016-11-30 06:56:24.448932 I | Send loop terminated: err= I1130 06:56:24.448996 12 scheduler.go:1109] Sending error via withScheduler: Framework has been removed I1130 06:56:24.449061 12 scheduler.go:301] stopping scheduler event queue.. I1130 06:56:24.449079 12 scheduler.go:457] Scheduler received error: Framework has been removed I1130 06:56:24.449144 12 scheduler.go:253] finished processing scheduler events

Anyone has same issues or any correct docs can be refered? Thanks a lot.