-
Type:
Bug
-
Status: Open (View Workflow)
-
Priority:
Critical
-
Resolution: Unresolved
-
Affects Version/s: 1.11.0
-
Fix Version/s: None
-
Component/s: Platform
-
Labels:
-
Environment:
ONOS Version 1.11.1
Atomix Version 2.0.2
ONOS single node deployment with OVS switches
cluster.json is as follows.
sdn@sdnc-vm-1:/opt/onos/config$ cat cluster.json { "nodes": [ { "ip": "10.251.91.192", "id": "10.251.91.192", "port": 9876 } ], "name": 2974496688, "partitions": [ { "id": 1, "members": [ "10.251.91.192" ] } ] }ONOS Version 1.11.1 Atomix Version 2.0.2 ONOS single node deployment with OVS switches cluster.json is as follows. sdn@sdnc-vm-1:/opt/onos/config$ cat cluster.json { "nodes" : [ { "ip" : "10.251.91.192" , "id" : "10.251.91.192" , "port" : 9876 } ], "name" : 2974496688, "partitions" : [ { "id" : 1, "members" : [ "10.251.91.192" ] } ] }
PartitionManager activation is suspended after sudden ONOS VM power reset.
Reproduction steps are as follows.
1. Run ONOS as single node deployment in VM
2. Attaching 4 OVSes to the single node ONOS
onos> devices id=of:0afbd67200000000, available=true, local-status=connected 1m56s ago, role=MASTER, type=SWITCH, mfr=Nicira, Inc., hw=Open vSwitch, sw=2.6.90, serial=None, driver=dc-vtn, channelId=10.251.214.114:49631, managementAddress=10.251.214.114, protocol=OF_13 id=of:0afbd67200000001, available=true, local-status=connected 1m55s ago, role=MASTER, type=SWITCH, mfr=Nicira, Inc., hw=Open vSwitch, sw=2.6.90, serial=None, driver=dc-vtn, channelId=10.251.214.114:49632, managementAddress=10.251.214.114, protocol=OF_13 id=of:0afbd67400000000, available=true, local-status=connected 1m56s ago, role=MASTER, type=SWITCH, mfr=Nicira, Inc., hw=Open vSwitch, sw=2.6.90, serial=None, driver=dc-vtn, channelId=10.251.214.116:50059, managementAddress=10.251.214.116, protocol=OF_13 id=of:0afbd67400000001, available=true, local-status=connected 1m55s ago, role=MASTER, type=SWITCH, mfr=Nicira, Inc., hw=Open vSwitch, sw=2.6.90, serial=None, driver=dc-vtn, channelId=10.251.214.116:50060, managementAddress=10.251.214.116, protocol=OF_13
3. Do power reset for the VM(having ONOS) instantly
4. ONOS become alive, start re-boot but suspended in 'PartitionManage-activation' with the below log .
2017-11-29 05:31:30,844 | INFO | FelixStartLevel | NettyMessagingManager | 144 | Started
2017-11-29 05:31:31,108 | INFO | FelixStartLevel | DistributedClusterStore | 144 | Started
2017-11-29 05:31:31,131 | INFO | FelixStartLevel | ClusterManager | 142 | Started
2017-11-29 05:31:31,146 | INFO | FelixStartLevel | ClusterCommunicationManager | 144 | Started
2017-11-29 05:31:31,159 | INFO | FelixStartLevel | ComponentsMonitor | 142 | Started
2017-11-29 05:31:31,186 | INFO | FelixStartLevel | PersistenceManager | 146 | A previous database file has been found.
2017-11-29 05:31:31,271 | INFO | FelixStartLevel | PersistenceManager | 146 | Started
2017-11-29 05:31:34,334 | INFO | rver-partition-1 | RaftContext | 90 | RaftServer{partition-1} - Transitioning to FOLLOWER
2017-11-29 05:31:38,538 | INFO | rver-partition-1 | RaftContext | 90 | RaftServer{partition-1} - Transitioning to CANDIDATE
2017-11-29 05:31:38,543 | INFO | rver-partition-1 | CandidateRole | 90 | RaftServer{partition-1}{role=CANDIDATE} - Starting election
2017-11-29 05:31:38,556 | INFO | rver-partition-1 | RaftContext | 90 | RaftServer{partition-1} - Transitioning to LEADER
2017-11-29 05:31:38,595 | INFO | rver-partition-1 | RaftContext | 90 | RaftServer{partition-1} - Found leader 10.251.91.192
2017-11-28 20:33:46,845 | ERROR | rver-partition-1 | RaftServiceManager | 90 | RaftServer{partition-1} - Cannot commit index 1058
// ONOS is stopped from here
* Group Discusssion
https://groups.google.com/a/onosproject.org/forum/#!topic/onos-dev/zwgePdbWVzg