| *** Qiming has quit IRC | 00:00 | |
| *** lhcheng has quit IRC | 00:01 | |
| *** zeih has joined #tacker | 00:11 | |
| *** amotoki has joined #tacker | 00:12 | |
| *** arturt_ has quit IRC | 00:13 | |
| *** zeih has quit IRC | 00:15 | |
| *** arturt_ has joined #tacker | 00:18 | |
| openstackgerrit | Merged openstack/tacker: Fix the typo for 'VimNotFoundException' https://review.openstack.org/322296 | 00:20 |
|---|---|---|
| *** arturt_ has quit IRC | 00:24 | |
| *** amotoki has quit IRC | 00:32 | |
| *** arturt_ has joined #tacker | 00:33 | |
| *** sripriya has quit IRC | 00:41 | |
| *** Qiming has joined #tacker | 01:12 | |
| *** amotoki has joined #tacker | 01:15 | |
| *** amotoki_ has joined #tacker | 01:24 | |
| *** amotoki has quit IRC | 01:26 | |
| *** arturt_ has quit IRC | 01:38 | |
| *** zeih has joined #tacker | 01:39 | |
| *** zeih has quit IRC | 01:43 | |
| *** arturt_ has joined #tacker | 01:50 | |
| *** s3wong has quit IRC | 01:54 | |
| *** vishwanathj has joined #tacker | 02:03 | |
| *** zeih has joined #tacker | 02:06 | |
| *** zeih has quit IRC | 02:11 | |
| *** dkushwaha has quit IRC | 02:19 | |
| *** prashantD has quit IRC | 02:29 | |
| *** arturt_ has quit IRC | 02:33 | |
| *** dkushwaha has joined #tacker | 02:36 | |
| *** vishwanathj has quit IRC | 03:01 | |
| *** dkushwaha has quit IRC | 03:06 | |
| *** KanagarajM has joined #tacker | 03:10 | |
| *** sripriya has joined #tacker | 03:24 | |
| *** sripriya has quit IRC | 03:33 | |
| *** sripriya has joined #tacker | 03:34 | |
| *** zeih has joined #tacker | 03:35 | |
| *** dkushwaha has joined #tacker | 03:36 | |
| *** zeih has quit IRC | 03:39 | |
| KanagarajM | sripriya, ping | 03:40 |
| sripriya | KanagarajM: pong | 03:40 |
| KanagarajM | sripriya, for the db optimization, for making some the entry not null, would like to get your opinion | 03:41 |
| KanagarajM | sripriya, https://etherpad.openstack.org/p/tacker-db | 03:42 |
| KanagarajM | sripriya, from line 52 | 03:42 |
| * sripriya looking | 03:43 | |
| *** arturt_ has joined #tacker | 03:48 | |
| sripriya | KanagarajM: the list of changes look fine for me | 03:50 |
| sripriya | KanagarajM: my concern is for mgmt_driver being not NULL may throw error if mgmt_driver is not specified | 03:50 |
| KanagarajM | sripriya, ok in that case, should we keep mgmt_driver keeps NULL as exist now? | 03:52 |
| *** arturt_ has quit IRC | 03:53 | |
| *** arturt_ has joined #tacker | 03:53 | |
| sripriya | KanagarajM: or ensure we set it to 'noop' as default in api layer before pushing it to db | 03:53 |
| sripriya | KanagarajM: https://github.com/openstack/tacker/blob/master/tacker/vm/tosca/utils.py#L244 | 03:53 |
| KanagarajM | sripriya, only make vims.tenant_id CAN NOT NULL | 03:53 |
| KanagarajM | devicetemplates.tenant_id CAN NOT NULL | 03:53 |
| KanagarajM | servicetypes.tenant_id CAN NOT NULL | 03:53 |
| sripriya | KanagarajM: yup | 03:54 |
| KanagarajM | sripriya, ok i think its good idea, how about the infra_driver? | 03:54 |
| KanagarajM | it should be not null right | 03:54 |
| sripriya | KanagarajM: which one? | 03:55 |
| KanagarajM | devicetemplates.infra_driver CAN NOT NULL | 03:56 |
| KanagarajM | sripriya, ^^ | 03:56 |
| sripriya | KanagarajM: yes, that should be NOT NULL | 03:57 |
| KanagarajM | sripriya, thanks, i will try to put one more patch for this ones. | 03:59 |
| sripriya | KanagarajM: in fact there is inconsistency in API layer for infra_driver, we set infra_driver as ATTR_NOT_SPECIFIED (which set to default), but then raise an exception if infra driver is not specified at plugin level | 03:59 |
| KanagarajM | sripriya, its evident that infra_driver can't be null ! thanks for pointing out it. | 04:01 |
| KanagarajM | sripriya, do you think setting noop as default value for mgmt_driver would be nice option? | 04:01 |
| KanagarajM | sripriya, in the db schema itself | 04:01 |
| *** zeih has joined #tacker | 04:02 | |
| *** prashantD has joined #tacker | 04:02 | |
| sripriya | KanagarajM: suppose at any point in future, i want to change the default mgmt_driver, i will have to migrate my db scripts just to update this | 04:05 |
| *** prashantD_ has joined #tacker | 04:06 | |
| *** zeih has quit IRC | 04:06 | |
| KanagarajM | sripriya, we can always override from code i believe, and i assume that noop would be the default one for mgmt_driver , may be for ever :) | 04:08 |
| *** arturt_ has quit IRC | 04:09 | |
| *** prashantD has quit IRC | 04:10 | |
| sripriya | KanagarajM: may be some providers do not want to have 'noop' at all as default and have their custom driver as default :-) | 04:12 |
| sripriya | KanagarajM: i think having in db as 'noop | 04:12 |
| sripriya | KanagarajM: is fine and we can overrride in code at any time | 04:12 |
| KanagarajM | sripriya, yes. that flexibility always we have | 04:12 |
| sripriya | KanagarajM: cool. | 04:13 |
| KanagarajM | sripriya, i feel, we could do the same for the vims.type SET TO openstack | 04:13 |
| sripriya | KanagarajM: i don't know if it is a value add to have 'openstack' as default as some providers may not necessarily want 'openstack' vim as default and we force them to override it as default in code. also vim type in api layer is mandatory | 04:16 |
| *** arturt_ has joined #tacker | 04:16 | |
| KanagarajM | sripriya, got your point, that is right. thanks :) | 04:17 |
| *** arturt_ has quit IRC | 04:24 | |
| *** arturt_ has joined #tacker | 04:24 | |
| *** arturt_ has quit IRC | 04:27 | |
| *** arturt_ has joined #tacker | 04:27 | |
| openstackgerrit | Sripriya Seetharam proposed openstack/tacker: Fix keystone error handling in openstack driver https://review.openstack.org/321806 | 04:35 |
| *** prashantD_ has quit IRC | 04:38 | |
| *** arturt_ has quit IRC | 04:44 | |
| openstackgerrit | bharaththiruveedula proposed openstack/tacker: Add monitor feature to the VIM https://review.openstack.org/315475 | 04:47 |
| *** arturt_ has joined #tacker | 04:52 | |
| *** arturt_ has quit IRC | 05:04 | |
| *** arturt_ has joined #tacker | 05:05 | |
| *** arturt_ has quit IRC | 05:05 | |
| *** arturt_ has joined #tacker | 05:07 | |
| *** arturt_ has quit IRC | 05:07 | |
| *** arturt_ has joined #tacker | 05:10 | |
| *** arturt_ has quit IRC | 05:15 | |
| sridhar_ram | sripriya: ping | 05:17 |
| *** amotoki_ is now known as amotoki | 05:19 | |
| *** zeih has joined #tacker | 05:30 | |
| *** zeih has quit IRC | 05:35 | |
| openstackgerrit | Kanagaraj Manickam proposed openstack/tacker: db schema improvements: openstack entity id set to 64 length https://review.openstack.org/323694 | 05:55 |
| openstackgerrit | Kanagaraj Manickam proposed openstack/tacker: db schema improvements: set mandatory column not null https://review.openstack.org/324219 | 05:55 |
| *** zeih has joined #tacker | 05:57 | |
| *** zeih has quit IRC | 06:02 | |
| *** dkushwaha has quit IRC | 06:04 | |
| *** vishnoianil has quit IRC | 06:10 | |
| *** dkushwaha has joined #tacker | 06:17 | |
| sripriya | sridhar_ram: pong | 06:22 |
| *** vishwanathj has joined #tacker | 07:15 | |
| *** vishwana_ has joined #tacker | 07:15 | |
| *** vishwanathj has quit IRC | 07:19 | |
| openstackgerrit | gongysh proposed openstack/tacker: Change default port into 9898 https://review.openstack.org/324261 | 07:19 |
| *** vishwana_ is now known as vishwanathj | 07:26 | |
| *** vishwanathj has quit IRC | 07:34 | |
| *** zeih has joined #tacker | 07:35 | |
| openstackgerrit | Santosh Kodicherla proposed openstack/tacker: Adding vnf testcase with monitoring with new tosca template https://review.openstack.org/300248 | 07:47 |
| *** zeih has quit IRC | 08:01 | |
| *** dkushwaha has quit IRC | 08:06 | |
| *** mbound has joined #tacker | 08:08 | |
| *** vishwanathj has joined #tacker | 08:14 | |
| *** dkushwaha has joined #tacker | 08:28 | |
| *** zeih has joined #tacker | 08:40 | |
| *** mbound has quit IRC | 08:49 | |
| *** zeih has quit IRC | 08:52 | |
| *** dkushwaha has quit IRC | 08:59 | |
| openstackgerrit | Kanagaraj Manickam proposed openstack/tacker: nova infra driver failed to load https://review.openstack.org/324319 | 09:15 |
| openstackgerrit | gongysh proposed openstack/tacker: Use oslo log and oslo concurrency https://review.openstack.org/324329 | 09:41 |
| openstackgerrit | gongysh proposed openstack/tacker: Use oslo log and oslo concurrency https://review.openstack.org/324329 | 09:44 |
| openstackgerrit | gongysh proposed openstack/tacker: Use oslo log and oslo concurrency https://review.openstack.org/324329 | 09:46 |
| *** zeih has joined #tacker | 09:51 | |
| *** zeih has quit IRC | 09:56 | |
| *** KanagarajM has quit IRC | 09:57 | |
| *** vishwanathj has quit IRC | 09:57 | |
| *** Qiming has quit IRC | 09:59 | |
| *** amotoki has quit IRC | 10:01 | |
| *** amotoki has joined #tacker | 10:04 | |
| *** vishwanathj has joined #tacker | 10:07 | |
| *** vishwanathj has quit IRC | 10:21 | |
| *** amotoki has quit IRC | 10:37 | |
| *** amotoki has joined #tacker | 10:39 | |
| *** fxpester has quit IRC | 10:46 | |
| *** fxpester has joined #tacker | 10:46 | |
| *** Qiming has joined #tacker | 10:56 | |
| *** zeih has joined #tacker | 10:59 | |
| *** amotoki has quit IRC | 11:07 | |
| *** amotoki has joined #tacker | 11:15 | |
| *** amotoki has quit IRC | 12:11 | |
| *** amotoki has joined #tacker | 12:18 | |
| *** amotoki has quit IRC | 12:18 | |
| *** aasmith has joined #tacker | 12:25 | |
| *** tbh has joined #tacker | 12:34 | |
| *** bobh has joined #tacker | 12:36 | |
| *** amotoki has joined #tacker | 12:36 | |
| *** bobh has quit IRC | 12:41 | |
| *** amotoki has quit IRC | 12:53 | |
| *** KanagarajM has joined #tacker | 13:03 | |
| *** mbound has joined #tacker | 13:11 | |
| *** mbound has quit IRC | 13:13 | |
| *** mbound has joined #tacker | 13:13 | |
| *** tbh has quit IRC | 13:16 | |
| *** zeih has quit IRC | 13:22 | |
| *** zeih has joined #tacker | 13:34 | |
| *** twm2016 has joined #tacker | 14:05 | |
| *** bobh has joined #tacker | 14:13 | |
| *** aasmith has quit IRC | 14:30 | |
| *** amotoki has joined #tacker | 14:35 | |
| *** aasmith has joined #tacker | 14:46 | |
| *** zeih has quit IRC | 14:58 | |
| *** aasmith has quit IRC | 15:00 | |
| *** zeih has joined #tacker | 15:07 | |
| *** aasmith has joined #tacker | 15:08 | |
| *** zeih has quit IRC | 15:11 | |
| *** aasmith is now known as tsa-la-gi | 15:23 | |
| *** zeih has joined #tacker | 15:30 | |
| *** zeih has quit IRC | 15:35 | |
| *** prashantD has joined #tacker | 15:37 | |
| *** KanagarajM has quit IRC | 15:41 | |
| *** vishnoianil has joined #tacker | 15:48 | |
| *** tsa-la-gi has quit IRC | 15:50 | |
| *** zeih has joined #tacker | 15:58 | |
| *** uck has joined #tacker | 16:02 | |
| *** zeih has quit IRC | 16:03 | |
| *** tsa-la-gi has joined #tacker | 16:05 | |
| *** KanagarajM has joined #tacker | 16:07 | |
| *** tbh has joined #tacker | 16:08 | |
| *** aasmith has joined #tacker | 16:09 | |
| *** Qiming has quit IRC | 16:11 | |
| *** tsa-la-gi has quit IRC | 16:11 | |
| *** lhcheng has joined #tacker | 16:13 | |
| *** amotoki has quit IRC | 16:20 | |
| *** amotoki has joined #tacker | 16:22 | |
| *** amotoki has quit IRC | 16:32 | |
| *** tbh has quit IRC | 16:37 | |
| *** tbh has joined #tacker | 16:39 | |
| *** s3wong has joined #tacker | 17:09 | |
| *** tbh has quit IRC | 17:24 | |
| *** tbh has joined #tacker | 17:27 | |
| *** sripriya has quit IRC | 17:28 | |
| *** amotoki has joined #tacker | 17:32 | |
| *** amotoki has quit IRC | 17:38 | |
| sridhar_ram | tbh: ping | 17:57 |
| tbh | sridhar_ram, Hi | 17:57 |
| sridhar_ram | quick follow up on https://review.openstack.org/#/c/315475/ | 17:58 |
| sridhar_ram | tbh: new patchset looks reasonable, only concern is when we have lots of unreachable VIM how does tacker server would perform | 17:59 |
| sridhar_ram | tbh: particularly when a new vim-register happen to arrive in this "window" it will "hang" for a while | 17:59 |
| tbh | sridhar_ram, yes I agree, in the worst case, all the CRUD operations will hold | 18:00 |
| sridhar_ram | tbh: can you characterize the impact by adding few "bad" VIMs and see how long it takes for monitor to come out of the critical section ? | 18:01 |
| *** sripriya has joined #tacker | 18:01 | |
| sridhar_ram | tbh: say, adding 10 invalid VIMs and see the impact | 18:01 |
| sridhar_ram | sripriya: ^^^ | 18:01 |
| tbh | sridhar_ram, so I was thinking, I am just reading the values of created_vims and not writing anything | 18:02 |
| * sripriya catching up | 18:02 | |
| tbh | sridhar_ram, sure I will check that | 18:02 |
| sridhar_ram | tbh: bottom line, i'm looking to characterize the situation, so that we can leave a TODO in the code to revisit / enhance this piece | 18:03 |
| sridhar_ram | tbh: .. in a follow on | 18:03 |
| sridhar_ram | tbh: thanks! | 18:03 |
| sripriya | tbh: sridhar_ram: will take a look at the updated patchset | 18:08 |
| sripriya | tbh: i see in code that you are still creating an url connection request to server | 18:09 |
| sripriya | tbh: do you plan to change that? | 18:09 |
| tbh | sripriya, yes | 18:10 |
| sripriya | tbh: ok, just wanted to confirm. thanks. | 18:10 |
| sridhar_ram | tbh: sripriya: fyi, left just more comment as well | 18:11 |
| sripriya | sridhar_ram: ack | 18:11 |
| tbh | sripriya, sure | 18:12 |
| *** uck has quit IRC | 18:16 | |
| tbh | sridhar_ram, actually as in the current version we are taking values from tacker.conf | 18:16 |
| tbh | sripriya suggested to take monitor details from the client | 18:16 |
| tbh | sridhar_ram, so I am trying to have another monitor_drivers folder in nfvo/ | 18:17 |
| sridhar_ram | tbh: sorry, which value are we grabbing from tacker.conf ? | 18:18 |
| tbh | and take monitor_type and monitor_action value from the tacker client | 18:18 |
| tbh | sridhar_ram, retry and timeout currently | 18:19 |
| tbh | and we are not taking any monitoring_type value( I mean no modularity) | 18:19 |
| sridhar_ram | tbh: sripriya: why ? I don't think VIM health mon needs to be such a customizable feature .. at least at this point. I'd suggest something simple, as i mentioned earlier a lightweight, lazy health check | 18:19 |
| tbh | sridhar_ram, yeah that's true, but most of the comments raised on this concern, | 18:21 |
| sridhar_ram | tbh: sripriya: we have a TOSCA template policies to describe complicated, customizable monitoring for VNFs & VDUs | 18:21 |
| tbh | i.e, not to just check check network reachability | 18:21 |
| sripriya | sridhar_ram: this won't allow to bring in external agents for vim monitoring | 18:21 |
| sridhar_ram | sripriya: this is something we can delegate to individual VIM type to take inputs from tacker.conf... | 18:22 |
| sridhar_ram | sripriya: we shouldn't push that complexity into vim-register API... we will have only a handful of VIM types.. | 18:22 |
| sridhar_ram | sripriya: custom VIM monitoring for a given VIM can still be configured using [vim_foo_monitor_type] in tacker.conf | 18:23 |
| sripriya | sridhar_ram: but this will require us to restart tacker sever in case we want to change the monitor_type for a given VIM | 18:24 |
| sridhar_ram | sripriya: or rather have a [vim_foo] section in the tacker.conf and have monitor_type and monitor_action as settings | 18:24 |
| sridhar_ram | sripriya: yes, the assumption here is introducing new VIMs is usually a s/w upgrade and needs a tacker-server restart | 18:25 |
| sripriya | sridhar_ram: that also implies we will have the same monitoring type for all VIMs | 18:26 |
| sridhar_ram | sripriya: introducing new VIM type is a major event in to a tacker runtime system.. opposed to introducing new VNF moniroting types | 18:26 |
| sripriya | sridhar_ram: i saw 'vim' and 'vnf' as resources that can be configured run time dynamically | 18:27 |
| sridhar_ram | sripriya: no necessarily, openstack VIM might not do more than a simple auth_url ping .. and if it does offer few flavors of it, we can still have this selected using tacker.conf | 18:27 |
| sripriya | sridhar_ram: and that would apply to all VIMs right? the same auth_url ping. | 18:28 |
| sridhar_ram | sripriya: we can let that actual vim driver to implement the "monitor_vim" call .. and they can do what ever they want.. | 18:29 |
| sridhar_ram | sripriya: if a particular VIM type has multuple VIM monitoring types.. lets them read that from tacker.conf | 18:29 |
| sridhar_ram | again, we will *not* have tons of VIM types at any moment in tacker... | 18:30 |
| sridhar_ram | since this is the first time we are introducing this health mon for VIM, lets do the basic first and see if we get some operator inputs | 18:31 |
| *** amotoki has joined #tacker | 18:34 | |
| sripriya | sridhar_ram: ok, so we need the implementation of reading from tacker.conf and having a custom monitor_vim call which can either call default monitor type or write their own which can listen to some external agent specific for that VIM | 18:35 |
| sridhar_ram | sripriya: exactly ! | 18:36 |
| *** vishnoianil has quit IRC | 18:38 | |
| *** amotoki has quit IRC | 18:39 | |
| sridhar_ram | sripriya: monitor_vim call is already VIM specific, correct? | 18:39 |
| sridhar_ram | tbh: ^^ | 18:39 |
| sridhar_ram | tbh: sripriya : openstack-driver will have an implementation and another VIM type can have its own ? | 18:40 |
| tbh | sridhar_ram, yup, following the conversation | 18:43 |
| sridhar_ram | bobh: can you please review https://review.openstack.org/321806 ? This is the last one holding a point release off stable/mitaka | 18:43 |
| sripriya | sridhar_ram: yes | 18:45 |
| sridhar_ram | sripriya: tbh: okay, we can defer this flexibility to the actual VIM driver implementation.. | 18:46 |
| sridhar_ram | sripriya: tbh: do we need to support different monitor_type for openstack VIM driver ? | 18:47 |
| tbh | sridhar_ram, I think we can continue with checking network reachability for now | 18:51 |
| sridhar_ram | tbh: okay.. i'll let you and sripriya make the right call on that. however, IMO, we should keep this initial vim health implementation simple and let it evolve in iterations in the future. | 18:53 |
| sripriya | sridhar_ram: tbh: agree. we can enhance it in follow ons | 19:02 |
| *** KanagarajM has quit IRC | 19:20 | |
| *** amotoki has joined #tacker | 19:34 | |
| *** zeih has joined #tacker | 19:35 | |
| *** amotoki has quit IRC | 19:39 | |
| *** openstack has joined #tacker | 19:43 | |
| *** uck has joined #tacker | 19:48 | |
| *** mbound has quit IRC | 19:53 | |
| *** uck has quit IRC | 19:53 | |
| *** uck_ has joined #tacker | 19:53 | |
| *** zeih has joined #tacker | 20:03 | |
| openstackgerrit | Merged openstack/tacker: Fix keystone error handling in openstack driver https://review.openstack.org/321806 | 20:04 |
| *** zeih has quit IRC | 20:07 | |
| *** tbh has quit IRC | 20:13 | |
| *** uck has joined #tacker | 20:32 | |
| *** uck_ has quit IRC | 20:32 | |
| *** amotoki has joined #tacker | 20:35 | |
| *** vishnoianil has joined #tacker | 20:36 | |
| *** amotoki has quit IRC | 20:40 | |
| *** santoshk has joined #tacker | 20:43 | |
| *** mbound has joined #tacker | 20:53 | |
| *** mbound has quit IRC | 20:58 | |
| *** aasmith has quit IRC | 21:04 | |
| *** prashantD_ has joined #tacker | 21:13 | |
| *** prashantD has quit IRC | 21:16 | |
| *** prashantD has joined #tacker | 21:17 | |
| *** prashantD_ has quit IRC | 21:19 | |
| *** amotoki has joined #tacker | 21:36 | |
| *** amotoki has quit IRC | 21:42 | |
| *** prashantD_ has joined #tacker | 21:56 | |
| *** prashantD has quit IRC | 21:59 | |
| *** twm2016 has quit IRC | 22:06 | |
| *** openstackgerrit has quit IRC | 22:19 | |
| *** openstackgerrit has joined #tacker | 22:20 | |
| *** bobh has quit IRC | 22:22 | |
| *** arturt_ has joined #tacker | 22:36 | |
| *** arturt_ has quit IRC | 22:37 | |
| *** arturt_ has joined #tacker | 22:50 | |
| *** vishnoianil has quit IRC | 23:07 | |
| *** Qiming has joined #tacker | 23:09 | |
| *** lhcheng has quit IRC | 23:22 | |
| *** sripriya_ has joined #tacker | 23:23 | |
| *** sripriya has quit IRC | 23:26 | |
| *** aasmith has joined #tacker | 23:30 | |
| *** amotoki has joined #tacker | 23:39 | |
| *** Qiming has quit IRC | 23:40 | |
| *** amotoki has quit IRC | 23:44 | |
| *** vishwanathj has joined #tacker | 23:49 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!