16:05:00 <edwarnicke> #startmeeting MD-SAL meeting 16:05:00 <odl_meetbot> Meeting started Tue Jun 24 16:05:00 2014 UTC. The chair is edwarnicke. Information about MeetBot at http://ci.openstack.org/meetbot.html. 16:05:00 <odl_meetbot> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 16:05:00 <odl_meetbot> The meeting name has been set to 'md_sal_meeting' 16:05:04 <edwarnicke> #info https://meetings.webex.com/collabs/#/meetings/detail?uuid=MCQK23GTHZNO7LDV6UWT6KLYPV-9VIB&rnd=497132.35684 16:05:13 <devinavery> #info https://wiki.opendaylight.org/view/MD-SAL_Weekly_Call 16:05:21 <edwarnicke> #link https://wiki.opendaylight.org/view/MD-SAL_Weekly_Call 16:05:27 <edwarnicke> #link https://meetings.webex.com/collabs/#/meetings/detail?uuid=MCQK23GTHZNO7LDV6UWT6KLYPV-9VIB&rnd=497132.35684 16:05:36 <edwarnicke> #topic Weekly Status 16:07:02 <edwarnicke> devinavery: Could we get a link to the change you are speaking of? 16:09:47 <edwarnicke> #link https://git.opendaylight.org/gerrit/#/c/8046/ - Updates to DataBroker javadocs 16:11:01 <devinavery> #link https://git.opendaylight.org/gerrit/#/c/8168 - Updates to the toaster example to just the new DataBroker as documented 16:11:36 <edwarnicke> #link https://git.opendaylight.org/gerrit/#/q/project:controller+branch:master+topic:clustering,n,z - clustering patches 16:15:06 <devinavery> #link https://git.opendaylight.org/gerrit/#/c/8255/ Adding documentation to the opendaylight-invetory.yang file. 16:16:30 <edwarnicke> #chair devinavery 16:16:30 <odl_meetbot> Current chairs: devinavery edwarnicke 16:17:57 <edwarnicke> #info Colin raised possible future discussion about what you can understand from yang vs java 16:18:04 <colindixon> thanks 16:18:16 <edwarnicke> #info alagalah raised (potentially separable) question about what you can *do* from yang vs java 16:18:23 <alagalah> Thank you sir 16:18:51 <colindixon> #info I think alagalah’s point was actually what *should be done* in Yang vs. Java, but all 3 are worth talking about 16:18:57 <alagalah> Not so much what you CAN do, as to what you SHOULD do 16:19:06 <colindixon> :p 16:19:09 <alagalah> colindixon: ack 16:20:11 <kw_> re: yang vs java, sounds like an interesting discussion, too bad i can't make next week's meeting 16:20:49 <edwarnicke> #link https://git.opendaylight.org/gerrit/#/q/message:XSQL,n,z - XSQL patches Sharon would like to present next week 16:20:53 <alagalah> kw_: Its quite interesting 16:21:22 <edwarnicke> kw_: You will be missed 16:22:06 <colindixon> ugh, tony was *so* audible yesterday 16:22:21 <colindixon> and he’s back to so quiet it hurts :-( 16:22:24 <alagalah> colindixon: welcome to the interwebs 16:23:36 <devinavery> #topic Operational vs Config Datastores 16:23:48 <devinavery> #info Operational datastore is always a superset of the configuration datastore. 16:26:03 <devinavery> #info - apx 22 minutes in Ed is sharing screen which provides example of this 16:26:04 <rgowrishankar> #info if you add the line "config false" to a property, it will exist only in the operational datastore 16:30:59 <devinavery> #info Q: if you have a config attr called "port" how would you keep them in sync? Does MD-SAL provide this? Or is it the responsibility of the app? 16:31:49 <devinavery> #info A from Ed: It is the responsibility of the app - there is no way for MD-SAL to understand the operational attributes and how they are used. 16:32:54 <devinavery> #info A from Ed continue: The operational and config can be out of sync for a valid reason - you want something configured, but it couldn't be applied to operational correctly 16:33:11 <devinavery> #info Q from Colin: could we have good defaults like sync config to operational by default? 16:33:32 <devinavery> #info A from Ed: We need to explore the use-cases where sync from config to operational is desired. 16:34:37 <devinavery> #info Colin points out that config is the intent, and operational is the application of the intent. Colin suggests improving documentation to make this easier to understand and use. 16:35:06 <colindixon> yes! 16:35:07 <colindixon> that! 16:35:56 <devinavery> #info Recommendation from Ramkumar - good improvement would be an easy way to correlate between the config and operational trees. For example, given node 1 in the config tree, get me the same node in the operational 16:36:02 <colindixon> +10 to that 16:36:54 <devinavery> #info From Tony - use the same instance identifier in the operational store to get the same information that you did from the config store (i.e. opendaylight-inventory:nodes/node/myCustomNode is the ID in both datastores 16:37:09 <colindixon> devinavery: I should stop trying to take notes 16:37:14 <colindixon> you’re faster :p 16:37:17 <devinavery> :P 16:37:46 <devinavery> #info "A node has an address - It should have the same address in config and operational." 16:40:28 <devinavery> #info If lead Foo { config: false } is defined, then you can not put foo=5 into the config store (since it is an operational attribute) 16:40:55 <edwarnicke> #info examples of how I tend to represent nodes tersely 16:41:26 <edwarnicke> #info nodes->node[nodeId]- the node with nodeId 16:41:37 <devinavery> #info Are the stores enforcing the config false? 16:42:12 <edwarnicke> #info nodes->node[nodeId](flowcapable) -> tables -> table[tableId] - the table with tableId in a flowcapable augmented node with nodeId 16:44:25 <devinavery> #info no, The stores are not enforcing the config = false because the data stores for the binding aware enforce that. It complicates the case where people read from one store and writing to other and would complicate the binding aware logic 16:45:45 <devinavery> #info Question - what is a use case to copy from operational to config? A: Example - you query the switch for the current values and then write to the current config (which is stored in the operational) to the config 16:46:46 <colindixon> devinavery: let me know if you want me to take over scribing 16:46:54 <devinavery> #info Q: If we are not going to enforce this flag, why do we even have the flag and confuse people? 16:47:17 <devinavery> colindixon: sorry :) When I am not talking I am happy to type. :) Will let you know if I stop 16:47:42 <colindixon> scribing always makes sure you understand what’s going on 16:49:28 <devinavery> #info A: Having two datastores makes sense to separate what the user wants for a config and what it actually comes as (example, MTU size) 16:50:36 <devinavery> #info Suggestion (POV) - ONLY the data read from the network should be written to operational 16:51:13 <devinavery> #info Further clarification. Operational is a feedback loop, providing information from the real world, and the config is the intent. 16:51:47 <colindixon> I am now even more confused about things 16:52:19 <colindixon> I have (at least) two different ideas about how to do things should be separated between operational and config 16:52:40 <colindixon> one is that config is intent and operational is expression 16:52:57 <colindixon> the other is that config is for writing and opeartaional is for reading (loosely) 16:53:14 <devinavery> #info Some comments on phone suggest that we should have a config attribute and a separate operational attribute 16:53:23 <colindixon> this is sounding like we have a powerful mechanism and it can be used in a variety of patterns 16:53:53 <uchau> Question: The diagram labels some nodes as flowcapable and others as vswitchcapable, are they bound to an underlying protocol plugin to for operational and config data processing? 16:54:08 <colindixon> and we need to demonstrate those patterns and possibly build helper code to make common patterns easily 16:54:10 <colindixon> esier 16:54:12 <colindixon> easier 16:56:20 <devinavery> #info Q: Should we enforce the config: false in the MD-SAL? some thoughts that we should enforce it 16:57:27 <devinavery> #info A: Sounds like general concensus that we should enforce this at the config datastore level 16:58:00 <devinavery> #info recommendation - suggestion that you just ignore the config attributes, or possible create an object which has a copy constructor in the java class 17:00:44 <devinavery> #info Need an enhancement to have the config store enforce this. 17:01:28 <devinavery> #info Suggestion that if someone writes operational data to config store it should be a hard fail 17:01:30 <colindixon> #info colindixon just wants to point out that he’s now confused about how we should use operational vs. config—some people are saying it’s intent vs. expression and others are saying that it’s user expressed data vs. measured data—and it’s sounding like this separation may more of a mechanism that can be used for different things and we should define common patters 17:03:48 <edwarnicke> #endmeeting