[OPNFLWPLUG-617] RPC Exceptions observed with openflowplugin He design (tested with VTN) Created: 09/Feb/16 Updated: 27/Sep/21 Resolved: 24/Feb/16 |
|
| Status: | Resolved |
| Project: | OpenFlowPlugin |
| Component/s: | General |
| Affects Version/s: | None |
| Fix Version/s: | None |
| Type: | Bug | ||
| Reporter: | Venkatrangan Govindarajan | Assignee: | Anil Vishnoi |
| Resolution: | Cannot Reproduce | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Environment: |
Operating System: All |
||
| Attachments: |
|
| External issue ID: | 5276 |
| Description |
|
Tested the VTN Openstack support. the instances were not reachable. When analyzing the log, Found exceptions when VTN Manager tries to invoke rpc's of openflowplgin. :54:14,874 | ERROR | TN Flow Thread-0 | ClearNodeFlowsTask | 163 - org.opendaylight.vtn.manager.implementation - 0.4.0.SNAPSHOT | remove-flow: Caught an exception: canceled=false, input=RemoveFlowInput [_flowTable=FlowTableRef [_value=KeyedInstanceIdentifier {targetType=interface org.opendaylight.yang.gen.v1.urn.opendaylight.flow.inventory.rev130819.tables.Table, path=[org.opendaylight.yang.gen.v1.urn.opendaylight.inventory.rev130819.Nodes, org.opendaylight.yang.gen.v1.urn.opendaylight.inventory.rev130819.nodes.Node[key=NodeKey [_id=Uri [_value=openflow:121713027126083]]], org.opendaylight.yang.gen.v1.urn.opendaylight.flow.inventory.rev130819.FlowCapableNode, org.opendaylight.yang.gen.v1.urn.opendaylight.flow.inventory.rev130819.tables.Table[key=TableKey [_id=0]]]}], _node=NodeRef [_value=KeyedInstanceIdentifier {targetType=interface org.opendaylight.yang.gen.v1.urn.opendaylight.inventory.rev130819.nodes.Node, path=[org.opendaylight.yang.gen.v1.urn.opendaylight.inventory.rev130819.Nodes, org.opendaylight.yang.gen.v1.urn.opendaylight.inventory.rev130819.nodes.Node[key=NodeKey [_id=Uri [_value=openflow:121713027126083]]]]}], _tableId=0, _transactionUri=Uri [_value=clean-up:openflow:121713027126083], _barrier=true, _strict=false, augmentation=[]] at com.google.common.util.concurrent.AbstractFuture$Sync.getValue(AbstractFuture.java:299)[39:com.google.guava:18.0.0] at org.opendaylight.controller.remote.rpc.RemoteRpcImplementation$1.onComplete(RemoteRpcImplementation.java:65)[138:org.opendaylight.controller.sal-remoterpc-connector:1.3.0.SNAPSHOT] at akka.dispatch.japi$CallbackBridge.apply(Future.scala:172)[124:com.typesafe.akka.actor:2.3.14] |
| Comments |
| Comment by Venkatrangan Govindarajan [ 09/Feb/16 ] |
|
As discussed enabled the vtn log in karaf. the following sequence created the issue ODL + VTN Manager, Stacked one openstack node, Createed Network and instances and there was no issue, Stacked an additional compute node, Created instances , error occured in - 0.2.0.SNAPSHOT | unregisterEntityOwnershipRequest: Unregister controller entity ownership request for Uri [_value=openflow:68962386306631] , but unable to find it in nodeCollectorMap ]} |
| Comment by Venkatrangan Govindarajan [ 09/Feb/16 ] |
|
Attachment vtn_karaf_log.tgz has been added with description: VTN Karaf Log with TRACE enabled |
| Comment by A H [ 10/Feb/16 ] |
|
Raising the severity of this to blocker because of impact to Beryllium release with respect to VTN project. |
| Comment by Abhijit Kumbhare [ 10/Feb/16 ] |
|
Can you please look at this Anil? |
| Comment by Anil Vishnoi [ 10/Feb/16 ] |
|
Looking at it. |
| Comment by Anil Vishnoi [ 10/Feb/16 ] |
|
Looking at the log, it looks like by the time VTN calls removeFlow() rpc to the bridge (br-int), bridge gets disconnected from the controller. So once bridge disconnected from the controller, openflowplugin will deregister the rpcs. So this is expected behavior from plugin. Also looks like the second compute node connected to the controller already had br-int bridge and that connects to the controller as a openflow switch and disconnected immediately and never connects back. This is where we see this rpc behavior. I see another bridge pops up and connected to controller in <1 second. So to me it looks like compute node stacking issue. Discussed this with venkat, he will be re-testing it with some sanity check on the compute node stacking and will update the bug with the current status. |
| Comment by Venkatrangan Govindarajan [ 10/Feb/16 ] |
|
Tested the scenario again with OpenStack. The exceptions and errors were not observed. I have modified the importance as Normal. The integration has issues but seems to be problem due to https://bugs.opendaylight.org/show_bug.cgi?id=4678, which we are debugging now. |
| Comment by Venkatrangan Govindarajan [ 24/Feb/16 ] |
|
The issue was initially thought to be a problem with openflowplugin as some exceptions were observed in karaf.log when tested. But, After further investigation, it was identified that VTN Bug#4678 was observed again. The issues have been addressed now. Although there was an observation for openflowplugin that sometimes a switch that was not found in inventory features in the notifications. We are not able to reproduce the problem now. So closing this bug for now. |