Skip to content
Snippets Groups Projects

Repository graph

You can move around the graph by using the arrow keys.
Select Git revision
  • gh-pages
  • hot-failover
  • master default protected
  • master-green
  • master-prejava8
  • perf-test
  • release.asplus
  • release.asplus.bvs
  • revert-555-master
  • revert-559-revert-555-master
  • revert-594-master
  • v0.8
  • v0.82
  • v0.85
  • v0.90
  • v0.91
  • v1.0
  • v1.1
  • v1.2
  • wallaby
  • v1.2
  • v0.91
  • v1.1
  • v1.0
  • v0.90
  • v0.85
  • v0.82
  • v0.8
  • asplus-rc5
  • asplus-rc2
30 results
Created with Raphaël 2.2.08Apr65432131Mar30292827262423222120191716151413121198765432129Feb28272524232221201716151413121098763131Jan30292827252423212019181716151413121312111098432131Dec3022232221201916Update device AP learningMerge into master from pull request #74:Merge into master from pull request #75:Get all the OFActions working for IStaticFlowPusherService.Add try-catch clauses around all update queues.Add try-catch clauses around all update queues.Added an API to list all static flows by switch. The URL is:Change how modules are inited/started to be more deterministic.When you clear static flows via the REST API you can now specify "all" or a DPID in hex format.Make static flow pusher into a service (IStaticFlowEntryPusherService). This will allow other Floodlight modules to push static flows without using the REST API.Merge remote-tracking branch 'floodlight/master'Add 5 seconds for attachment point change from one broadcast domain to another. When the attachment points flap quickly, the controller should avoid learing the attachment point but should continue processing the packet.Resetting broadcast domain to broadcast domain attachment point change to 0 ms.Merge remote-tracking branch 'floodlight/master'Device manager will have a 5-second time gap to change an attachment point for a host from one broadcast domain to another.Merge remote-tracking branch 'floodlight/master'Merge into master from pull request #72:Reduce log level of some messagesFix an issue causing infinite broadcast loops in OpenFlow topologies that have loops.v0.82 v0.82v0.82 v0.82Merge pull request #71 from alexreimers/masterFix an issue causing infinite broadcast loops in OpenFlow topologies that have loops.Merge into master from pull request #70:Add message to debug flapping between the same portbroadcast ARP for cluster that doesn't have the dst device's APMerge remote-tracking branch 'floodlight/master'Merge remote-tracking branch 'maiergre/master'Introduce a convenience method inSameIsland, which is the same as inSameCluster. In future, they may have different functionalities.Introduce a method called isConsistent(..) that compares an old attachment and a new attachment point. If the new attachment point is not consistent with the old one, then it has to be learnt by the device manager.Merge into master from pull request #69:Fix a bug for checking BDDP packet if the packet is not unicast for it to continue down the processing chain.If the incoming node port is blocked, allow only unicast and BDDP packets to continue down the processing chain.Interface methods for getting outgoing broadcast ports from a between two switch port tuples.Removing prints not through loggerRemove obsolete ITopology method stubs from LinkDiscoveryManager. Clean up logging.Merge into master from pull request #67:Merge into master from pull request #68:Topology service can provide alternative broadcast port given a src/dst pairWhen the device manager gets cleaned up (or when during topologyChanged), it should consider the time difference between switching from broadcast to non-broadcast attachment points (as they might be previously on two different switches).Harden up HexString.fromHexString() and check for errors in the DeviceManager REST API.Check for the correct MAC address size when getting a device from DeviceManager.