20:34:01 #startmeeting better host tracker interns 20:34:01 Meeting started Thu Aug 7 20:34:01 2014 UTC. The chair is colindixon. Information about MeetBot at http://ci.openstack.org/meetbot.html. 20:34:01 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 20:34:01 The meeting name has been set to 'better_host_tracker_interns' 20:34:10 #topic status update 20:34:10 https://git.opendaylight.org/gerrit/#/c/9796/ 20:34:25 #info we presented to some folks from Ericsson and the SFC project this morning 20:34:36 #link https://git.opendaylight.org/gerrit/#/c/9796/ the current code 20:34:49 #topic current work 20:36:17 evanz: can you merge this is you like it: https://git.opendaylight.org/gerrit/#/c/9722/ 20:36:47 gotcha 20:36:50 I meant to do that 20:36:51 forgot 20:37:12 so, right now, I see there being 4 things we need to do: 20:37:26 1.) merge the two topology nodes into one 20:37:33 2.) add links to the topology 20:37:46 3.) make sure we have a way to turn on host tracker without programming l2switch rules into the network 20:37:59 4.) provide (some of) the IfIptoHost API 20:38:14 1.) and 2.) I have some problems with that need to be discussed 20:38:21 and 0.) fix the l2switch 20:38:35 do those make sense? we can take them one at time 20:38:45 yep let's start with 1) 20:38:57 do you have the the draft code? 20:39:00 yeah I do 20:39:08 so actually, I wanted to start with fixing the l2swich 20:39:10 what’s going on there? 20:39:24 not sure 20:39:42 amit_mandke says hostnodes are working for him 20:39:46 let me pull completely fresh and try 20:39:57 that is what I did evanz 20:40:12 #info we have 4 (or 5) things on our plate: 0.) fix l2switch, 1.) Merge the two topology nodes for a host into one, 2.) Add links into the topology for attachment points, 3.) Make sure there's a karaf feature that disables any *active* control, but gives host tracker, 4.) IfIptoHost API 20:40:13 I wasn't able to see any log messages from datachange listener 20:40:26 #topic fixing the l2switch 20:41:08 #info https://git.opendaylight.org/gerrit/#/c/9766/ djx says he hasn’t been able to get data change messages after this commit was merged 20:41:13 #undo 20:41:13 Removing item from minutes: 20:41:21 #link https://git.opendaylight.org/gerrit/#/c/9766/ djx says he hasn’t been able to get data change messages after this commit was merged 20:41:28 does anyone knows if I'm a commiter of l2switch? 20:41:36 djx: no 20:41:44 #info amit_mandke says he does see host nodes after it 20:41:56 #action djx, evanz and colindixon to look into this more later if it’s not fixed 20:42:10 #topic merging the two topology nodes for each host into one 20:42:16 djx: you had things you wanted to go over here? 20:42:20 http://paste.fedoraproject.org/124004/43967140 20:42:36 this is what we see if both node and hostnode have the same NodeID 20:43:17 MD-SAL don't *merge* them 20:43:37 #info as of this morning we were adding two nodes for each host, one which had the augmentation (and thus the host-tracer-service:id and the attachment points) and another that had the termination points 20:43:59 so, two things 20:44:45 #link http://paste.fedoraproject.org/124004/43967140 djx says if you push the two nodes with the same ID, they don’t get merged, the second just replaces the first 20:44:51 #info this is the expected behavior 20:47:04 #info in addition to put, there should be a merge operation that will merge the two nodes 20:48:21 the only errors I'm getting are from statistics manager 20:48:22 #info alternately, you can build them as a single node by using NodeBuilder and HostNodeBuilder and then calling addAugmentation on the NodeBuilder with the built node 20:48:26 but the hostnodes aren't showing up 20:48:37 djx: do those make sense? 20:49:30 evanz: d'oh 20:50:42 alternately, you can build them as a single node by using NodeBuilder and HostNodeBuilder-> that is what I'm doing 20:50:48 both of them with the same ID 20:51:09 and it doesn’t work? 20:51:33 nop not even with merge 20:51:41 * colindixon looks 20:51:44 HostTrackerImpl line 185 20:51:59 I substitute a put with a merge 20:52:29 no, no 20:52:40 let me check out your patch 20:52:42 hold on 20:52:57 do you want hangout? 20:53:11 sure 20:53:27 * colindixon starts one up 20:53:41 send me an invitation to my email 20:53:54 https://plus.google.com/hangouts/_/g2zp6d7w73hcb5mmlhcweetlx4a 21:08:00 #info problem fixed 21:12:45 #topic back to fixing l2switch 21:12:56 #Info evanz says that he eventually sees host nodes, but it takes a while 21:13:24 colindixon: you're cutting out in hangout 21:13:26 #info because the change makes so we only get Addresses from ARP packets, the hypothesis is that we’re missing the creation of the first few Addresses 21:14:06 #info we would fix this by querying for all Addresses after we subscribe and calling the relevant functions 21:16:15 #info colindixon suggests running mininet with --arp will pre populate the ARP tables and thus will cause no ARPs to be sent and would verity this 21:18:21 #topic adding links 21:18:39 #action djx is going to take care of this 21:18:44 #undo 21:18:44 Removing item from minutes: 21:18:52 #action djx to write code to add links 21:19:10 #info you have to add two links one in each direction 21:46:44 #info djx asks how we can get the node ID for the switch we’re connected to, very long discussion leads to an answer 21:46:54 #info InstanceIdentifier iin = iinc.firstIdentifierOf(Node.class); 21:47:00 #info ListenableFuture> nodeFuture = readTx.read(LogicalDatastoreType.OPERATIONAL, iin); 21:47:10 #info Node n = nodeFuture.get().get(); 21:47:16 #info all in the the onDataChanged() 21:47:39 #info then we can assume that the inventory node ID is the same as the topology node ID 21:50:29 #action djx to make it so that when a NodeConnector goes away, we remove the link and the attachment point 21:51:09 #topic karaf feature 21:51:48 #info this is a task for later, but we need to make sure there's a karaf feature that disables any *active* control, but gives host tracker 21:52:24 #info if we don’t use karaf in l2swich, instead, this needs to be a list of the bundles to get only hostracker with no rule programming 21:52:55 #topic IfIptoHost API 21:56:16 #info the only hard part is generate all the AD-SAL types 21:59:44 #info there are helper functions inside the compatibility-parent project in the NodeMapping class 21:59:56 #info toADNode and toADNodeConnector are your friends 22:02:04 #undo 22:02:04 Removing item from minutes: 22:02:06 #undo 22:02:06 Removing item from minutes: 22:02:38 #info the things you want to make this easy are in the the TopologyMapping class in org.opendaylight.controller.md.compatibility.topology in the controller 22:02:59 #info you want toAdTopologyNode(), toAdTopologyNodeConnector() 22:03:29 #link https://jenkins.opendaylight.org/controller/job/controller-merge/lastSuccessfulBuild/artifact/target/apidocs/org/opendaylight/controller/sal/compatibility/topology/TopologyMapping.html here 22:04:38 #info once you get this working you might want to make a toADHost() method and find a place for it in the controller compatibility 22:06:56 #topic fixing assumptions 22:07:06 #info we currently assume the topology key is “flow:1” 22:08:41 #undo 22:08:41 Removing item from minutes: 22:09:14 #info we currently don’t actually get an instance identifier for a a termination point 22:15:00 #undo 22:15:00 Removing item from minutes: 22:15:25 #undo 22:15:25 Removing item from minutes: 22:15:28 #endmeeting