| *** krow has quit IRC | 00:00 | |
| *** krow has joined #openstack-trove | 00:01 | |
| *** yogeshmehra has joined #openstack-trove | 00:05 | |
| *** krow has quit IRC | 00:07 | |
| *** yogeshmehra has quit IRC | 00:12 | |
| *** yidclare has quit IRC | 00:15 | |
| *** yogeshmehra has joined #openstack-trove | 00:16 | |
| *** mmcdaris has quit IRC | 00:19 | |
| *** mmcdaris has joined #openstack-trove | 00:19 | |
| *** matsuhashi has joined #openstack-trove | 00:20 | |
| *** yogeshmehra has quit IRC | 00:22 | |
| *** mmcdaris has quit IRC | 00:32 | |
| *** mmcdaris has joined #openstack-trove | 00:35 | |
| *** mmcdaris has quit IRC | 00:44 | |
| *** grapex has joined #openstack-trove | 00:49 | |
| *** saurabhs has quit IRC | 01:01 | |
| *** nosnos has joined #openstack-trove | 01:11 | |
| *** yogeshmehra has joined #openstack-trove | 01:16 | |
| *** yogeshmehra has quit IRC | 01:21 | |
| *** grapex has quit IRC | 01:24 | |
| *** jlucci has quit IRC | 01:29 | |
| *** jmontemayor has joined #openstack-trove | 01:30 | |
| *** erkules_ has joined #openstack-trove | 01:33 | |
| *** erkules has quit IRC | 01:36 | |
| *** jmontemayor has quit IRC | 01:41 | |
| *** matsuhashi has quit IRC | 01:46 | |
| *** matsuhashi has joined #openstack-trove | 01:47 | |
| *** matsuhas_ has joined #openstack-trove | 01:49 | |
| *** matsuhashi has quit IRC | 01:49 | |
| *** tanisdl has quit IRC | 01:52 | |
| *** haomaiwang has quit IRC | 01:57 | |
| *** haomaiwang has joined #openstack-trove | 01:58 | |
| *** adrian_otto has quit IRC | 02:09 | |
| *** haomaiwa_ has joined #openstack-trove | 02:11 | |
| *** haomaiwang has quit IRC | 02:14 | |
| *** yogeshmehra has joined #openstack-trove | 02:17 | |
| *** demorris has joined #openstack-trove | 02:21 | |
| *** yogeshmehra has quit IRC | 02:23 | |
| *** amcrn has quit IRC | 02:26 | |
| *** coolsvap has joined #openstack-trove | 02:58 | |
| *** jasonb365 has joined #openstack-trove | 03:12 | |
| *** yogeshmehra has joined #openstack-trove | 03:18 | |
| *** yogeshmehra has quit IRC | 03:23 | |
| *** yogeshmehra has joined #openstack-trove | 04:19 | |
| *** KapilSaxena has joined #openstack-trove | 04:22 | |
| *** yogeshmehra has quit IRC | 04:25 | |
| *** radez_g0n3 has quit IRC | 04:40 | |
| *** SushilKM has joined #openstack-trove | 04:48 | |
| *** radez_g0n3 has joined #openstack-trove | 04:49 | |
| *** amytron has joined #openstack-trove | 04:56 | |
| *** yogeshmehra has joined #openstack-trove | 05:01 | |
| *** yogeshmehra has quit IRC | 05:01 | |
| *** dmakogon has joined #openstack-trove | 05:05 | |
| *** demorris has quit IRC | 05:09 | |
| *** ashestakov has joined #openstack-trove | 05:09 | |
| ashestakov | vipul here? | 05:10 |
|---|---|---|
| *** krow has joined #openstack-trove | 05:42 | |
| *** ashestakov has quit IRC | 05:57 | |
| *** yogeshmehra has joined #openstack-trove | 06:01 | |
| *** yogeshmehra has quit IRC | 06:06 | |
| *** yogeshmehra has joined #openstack-trove | 06:12 | |
| *** vipul has quit IRC | 06:12 | |
| *** vipul has joined #openstack-trove | 06:12 | |
| *** yogeshmehra has quit IRC | 06:17 | |
| *** yogeshmehra has joined #openstack-trove | 06:19 | |
| *** vipul is now known as vipul-away | 06:31 | |
| *** yogeshmehra has quit IRC | 06:32 | |
| *** vipul-away is now known as vipul | 06:34 | |
| *** yogeshmehra has joined #openstack-trove | 06:52 | |
| *** erkules_ is now known as erkules | 06:56 | |
| *** yogeshmehra has quit IRC | 06:56 | |
| *** adrian_otto has joined #openstack-trove | 07:03 | |
| *** yogeshmehra has joined #openstack-trove | 07:18 | |
| *** dafter has joined #openstack-trove | 07:50 | |
| *** tvb|afk has joined #openstack-trove | 07:51 | |
| *** adrian_otto1 has joined #openstack-trove | 07:52 | |
| *** adrian_otto has quit IRC | 07:52 | |
| *** dmakogon has quit IRC | 07:55 | |
| *** dafter has quit IRC | 07:55 | |
| *** adrian_otto1 has quit IRC | 07:55 | |
| *** krow1 has joined #openstack-trove | 07:56 | |
| *** krow has quit IRC | 07:59 | |
| *** yogeshmehra has quit IRC | 08:18 | |
| *** krow has joined #openstack-trove | 08:21 | |
| *** krow1 has quit IRC | 08:25 | |
| *** adrian_otto has joined #openstack-trove | 08:42 | |
| *** krow has quit IRC | 08:56 | |
| *** KapilSaxena_ has joined #openstack-trove | 09:13 | |
| *** KapilSaxena has quit IRC | 09:15 | |
| *** tvb|afk has quit IRC | 09:18 | |
| *** yogeshmehra has joined #openstack-trove | 09:18 | |
| *** matsuhas_ has quit IRC | 09:19 | |
| *** matsuhas_ has joined #openstack-trove | 09:19 | |
| *** yogeshmehra has quit IRC | 09:23 | |
| openstackgerrit | A change was merged to openstack/python-troveclient: change mode to octal for os.makedirs() https://review.openstack.org/51952 | 09:27 |
| openstackgerrit | Illia Khudoshyn proposed a change to openstack/trove: Initial support for single instance MongoDB support https://review.openstack.org/50597 | 09:28 |
| *** yogeshmehra has joined #openstack-trove | 09:29 | |
| *** SnowDust has joined #openstack-trove | 09:33 | |
| *** yogeshmehra has quit IRC | 09:33 | |
| openstackgerrit | A change was merged to openstack/trove-integration: Adds conductor to ENABLED_SERVICES in localrc. https://review.openstack.org/51888 | 09:35 |
| *** dafter has joined #openstack-trove | 09:36 | |
| *** dafter has quit IRC | 09:36 | |
| *** dafter has joined #openstack-trove | 09:36 | |
| *** KapilSaxena_ has quit IRC | 09:37 | |
| *** KapilSaxena_ has joined #openstack-trove | 09:38 | |
| openstackgerrit | A change was merged to openstack/trove: Add GA related parameters to trove-taskmanager.conf.sample https://review.openstack.org/50624 | 09:46 |
| openstackgerrit | A change was merged to openstack/trove: Replace deprecated method aliases in tests https://review.openstack.org/50286 | 09:47 |
| openstackgerrit | Denis M. proposed a change to openstack/trove: Provide service type management code structure https://review.openstack.org/50686 | 10:17 |
| SnowDust | dmakogon_work | 10:24 |
| SnowDust | around ? | 10:24 |
| dmakogon_work | yes | 10:24 |
| SnowDust | lets talk security group through heat template | 10:24 |
| SnowDust | u made changes too but were asked questions .. | 10:25 |
| dmakogon_work | ok | 10:25 |
| SnowDust | the problem to flexibility is because of two alternate routes of instance creation | 10:25 |
| SnowDust | if we create a template with security group | 10:25 |
| SnowDust | the security group is dependent to service type and required number of ports .. | 10:26 |
| dmakogon_work | i'd like to keep default ports for now, than use ConfigurationParameters for changing database config parameters | 10:26 |
| SnowDust | and can be configured directly from the template. | 10:26 |
| dmakogon_work | we could define port range or single port | 10:26 |
| SnowDust | but .. then .. we miss when the code attaches security group for non heat instance creation | 10:26 |
| dmakogon_work | we would need it in future | 10:27 |
| dmakogon_work | *would not | 10:27 |
| SnowDust | yeah ... | 10:28 |
| SnowDust | rt now i dont have clarity on this .. | 10:28 |
| SnowDust | still thinking :-) | 10:28 |
| dmakogon_work | heat would assign SG to instance | 10:28 |
| SnowDust | rt now .. | 10:28 |
| SnowDust | SG is not attached to the heat instance | 10:28 |
| SnowDust | or is it ? | 10:28 |
| dmakogon_work | yes | 10:29 |
| SnowDust | even in case of mysql ? | 10:29 |
| dmakogon_work | missing nova call | 10:29 |
| dmakogon_work | but current heat support is a mess | 10:29 |
| SnowDust | we see the SG created .. for 3306 .. but none attached to the heat insance | 10:29 |
| dmakogon_work | yes | 10:29 |
| SnowDust | yeah .. so .. as am changing around it | 10:29 |
| SnowDust | thinking what changes we can go with | 10:29 |
| SnowDust | or we are tied to your checkin blocked by vipul's question :) | 10:30 |
| dmakogon_work | just call novaclient.servers.add_security group(name) | 10:30 |
| dmakogon_work | no we are not tied | 10:30 |
| SnowDust | understand .. but in that case we always add 3306 security group .. | 10:30 |
| dmakogon_work | i must go, will available in 1 hour | 10:30 |
| dmakogon_work | yes | 10:31 |
| SnowDust | that _create_server_volume_heat receives as argument | 10:31 |
| dmakogon_work | i know | 10:31 |
| SnowDust | then ? | 10:31 |
| dmakogon_work | we need flexibility | 10:31 |
| SnowDust | as the implementation is refactoring around service_type | 10:31 |
| SnowDust | we need to do it around service_type | 10:31 |
| dmakogon_work | yes | 10:31 |
| SnowDust | similar to the service_registry_ext | 10:31 |
| dmakogon_work | SnowDust: must go, i'll ping you later | 10:31 |
| SnowDust | sure | 10:31 |
| SnowDust | bye 4 now | 10:31 |
| *** dmakogon_work is now known as dmakogon_away | 10:32 | |
| *** jasonb365 has quit IRC | 10:44 | |
| *** yogeshmehra has joined #openstack-trove | 11:30 | |
| *** KapilSaxena_ has quit IRC | 11:33 | |
| *** yogeshmehra has quit IRC | 11:37 | |
| dmakogon_away | SnowDust: guess who's back | 11:40 |
| SnowDust | guess crazymac | 11:41 |
| dmakogon_away | maybe | 11:41 |
| *** dmakogon_away is now known as dmakogon_work | 11:41 | |
| dmakogon_work | what do we need to discuss ? | 11:41 |
| SnowDust | security group | 11:42 |
| SnowDust | during instance creation | 11:42 |
| SnowDust | how to make it flexible | 11:43 |
| SnowDust | to handle all kinds of it | 11:43 |
| SnowDust | if we keep it in templates .. heat side .. goes fine | 11:43 |
| SnowDust | but non-heat instance creation is still .. 3306 bound :) | 11:43 |
| SnowDust | ? | 11:47 |
| dmakogon_work | current API v1 fully supports only nova provisioning | 11:51 |
| dmakogon_work | heat support as a part of new API | 11:51 |
| dmakogon_work | v2 | 11:51 |
| dmakogon_work | in v2 we would support only heat | 11:51 |
| dmakogon_work | and all resources would be as defaults(no flags turn-on/off) and they would be parts of heat template | 11:53 |
| *** pdmars has joined #openstack-trove | 12:02 | |
| *** KapilSaxena has joined #openstack-trove | 12:08 | |
| SnowDust | that says .. | 12:11 |
| SnowDust | i can go ahead with security group in templates .. | 12:11 |
| SnowDust | dmakogon_work: thats agreed ? | 12:12 |
| dmakogon_work | yes | 12:12 |
| dmakogon_work | that is what i was doing | 12:13 |
| dmakogon_work | you can do that, even copy it from my review | 12:13 |
| *** adrian_otto has quit IRC | 12:14 | |
| openstackgerrit | Jenkins proposed a change to openstack/trove: Updated from global requirements https://review.openstack.org/51042 | 12:22 |
| *** adrian_otto has joined #openstack-trove | 12:33 | |
| *** yogeshmehra has joined #openstack-trove | 12:33 | |
| *** SnowDust has quit IRC | 12:37 | |
| *** yogeshmehra has quit IRC | 12:38 | |
| openstackgerrit | Denis M. proposed a change to openstack/trove: Initial support for single instance Cassandra Database https://review.openstack.org/51884 | 12:41 |
| *** coolsvap has quit IRC | 12:44 | |
| *** nosnos has quit IRC | 12:55 | |
| *** nosnos has joined #openstack-trove | 12:55 | |
| *** KapilSaxena has quit IRC | 12:57 | |
| *** KapilSaxena has joined #openstack-trove | 12:59 | |
| *** nosnos has quit IRC | 13:00 | |
| *** radez_g0n3 is now known as radez | 13:04 | |
| *** robertmyers has joined #openstack-trove | 13:17 | |
| *** shakayumi has joined #openstack-trove | 13:32 | |
| *** amytron has quit IRC | 13:32 | |
| *** SnowDust has joined #openstack-trove | 13:33 | |
| *** yogeshmehra has joined #openstack-trove | 13:34 | |
| *** rnirmal has joined #openstack-trove | 13:37 | |
| *** yogeshmehra has quit IRC | 13:39 | |
| *** ashestakov has joined #openstack-trove | 13:40 | |
| *** SnowDust has quit IRC | 13:44 | |
| *** jlucci has joined #openstack-trove | 13:55 | |
| *** SushilKM has quit IRC | 13:58 | |
| *** jcru has joined #openstack-trove | 14:01 | |
| *** amytron has joined #openstack-trove | 14:02 | |
| *** robertmyers has quit IRC | 14:04 | |
| *** robertmyers has joined #openstack-trove | 14:07 | |
| *** Nate1 has joined #openstack-trove | 14:18 | |
| openstackgerrit | Andrey Shestakov proposed a change to openstack/python-troveclient: Support datastore types https://review.openstack.org/47936 | 14:25 |
| openstackgerrit | Andrey Shestakov proposed a change to openstack/python-troveclient: Support datastore types https://review.openstack.org/47936 | 14:32 |
| *** demorris has joined #openstack-trove | 14:36 | |
| *** cweidenk1ller has joined #openstack-trove | 14:36 | |
| *** matsuhas_ has quit IRC | 14:38 | |
| *** matsuhashi has joined #openstack-trove | 14:39 | |
| *** matsuhashi has quit IRC | 14:39 | |
| *** matsuhashi has joined #openstack-trove | 14:40 | |
| *** matsuhashi has quit IRC | 14:42 | |
| *** matsuhashi has joined #openstack-trove | 14:42 | |
| *** matsuhashi has quit IRC | 14:47 | |
| *** tanisdl has joined #openstack-trove | 14:52 | |
| *** KapilSaxena has left #openstack-trove | 15:04 | |
| *** rnirmal has quit IRC | 15:13 | |
| *** grapex has joined #openstack-trove | 15:15 | |
| *** paul_lodronio has joined #openstack-trove | 15:18 | |
| *** yidclare has joined #openstack-trove | 15:26 | |
| *** haomaiwa_ has quit IRC | 15:29 | |
| *** haomaiwang has joined #openstack-trove | 15:30 | |
| dmakogon_work | haomaiwang: hi | 15:33 |
| haomaiwang | here | 15:33 |
| dmakogon_work | haomaiwang: have you read my comment ? | 15:33 |
| haomaiwang | sorry, going | 15:33 |
| dmakogon_work | about review and renaming/moving ? | 15:34 |
| *** jasonb365 has joined #openstack-trove | 15:34 | |
| haomaiwang | dmakogon_work: Oh, yes. I find your patch https://review.openstack.org/#/c/51884/ rename mysql relevant files | 15:35 |
| dmakogon_work | haomaiwang: yes | 15:35 |
| haomaiwang | but you did it another too | 15:35 |
| dmakogon_work | haomaiwang: yes | 15:35 |
| dmakogon_work | it would be better to merge into already prepared file structure | 15:36 |
| haomaiwang | And the previous is dependent to latter | 15:36 |
| *** yogeshmehra has joined #openstack-trove | 15:36 | |
| dmakogon_work | haomaiwang: historically refactoring review was first one | 15:36 |
| dmakogon_work | haomaiwang: it would prepare guestagent for integration with other services | 15:37 |
| haomaiwang | dmakogon_work: Hmm, seemed no real problem. Thanks. | 15:37 |
| haomaiwang | Another question, cloudinit is integrated into trove-integration? | 15:37 |
| haomaiwang | dmakogon_work: I find your review is the first | 15:38 |
| dmakogon_work | haomaiwang: cloudinit was already in trove | 15:38 |
| haomaiwang | dmakogon_work: yes | 15:38 |
| dmakogon_work | haomaiwang: i just used it for cassandra, because it needs special preparations | 15:39 |
| haomaiwang | wget https://bitbucket.org/pypa/setuptools/raw/bootstrap/ez_setup.py | 15:39 |
| haomaiwang | 37python ez_setup.py | 15:39 |
| haomaiwang | 38wget https://raw.github.com/pypa/pip/master/contrib/get-pip.py | 15:39 |
| haomaiwang | 39python get-pip.py | 15:39 |
| haomaiwang | I don't think cloudinit should contain source codes from github or others | 15:40 |
| dmakogon_work | haomaiwang: cloud-init script would be polished | 15:41 |
| *** yogeshmehra has quit IRC | 15:41 | |
| dmakogon_work | haomaiwang: this one is for my own tests | 15:41 |
| haomaiwang | dmakogon_work: OK | 15:41 |
| *** jasonb365 has quit IRC | 15:49 | |
| *** jlucci has quit IRC | 16:02 | |
| *** Barker has joined #openstack-trove | 16:03 | |
| *** dmakogon_work is now known as me_gone | 16:03 | |
| *** Barker has quit IRC | 16:07 | |
| *** Barker has joined #openstack-trove | 16:08 | |
| *** saurabhs has joined #openstack-trove | 16:11 | |
| *** Barker has quit IRC | 16:13 | |
| *** Barker has joined #openstack-trove | 16:16 | |
| *** dafter has quit IRC | 16:16 | |
| *** grapex has quit IRC | 16:20 | |
| *** jmontemayor has joined #openstack-trove | 16:20 | |
| *** SushilKM has joined #openstack-trove | 16:23 | |
| *** romainh has joined #openstack-trove | 16:27 | |
| *** ashestakov has quit IRC | 16:27 | |
| *** shakayumi has quit IRC | 16:33 | |
| *** rnirmal has joined #openstack-trove | 16:38 | |
| *** adrian_otto has quit IRC | 16:43 | |
| openstackgerrit | Illia Khudoshyn proposed a change to openstack/trove: Initial support for single instance MongoDB support https://review.openstack.org/50597 | 16:45 |
| *** demorris has quit IRC | 16:46 | |
| *** demorris has joined #openstack-trove | 17:02 | |
| *** demorris_ has joined #openstack-trove | 17:06 | |
| *** demorris has quit IRC | 17:07 | |
| *** demorris_ is now known as demorris | 17:07 | |
| *** arborism has joined #openstack-trove | 17:08 | |
| *** arborism is now known as amcrn | 17:08 | |
| *** yogeshmehra has joined #openstack-trove | 17:11 | |
| *** ashestakov has joined #openstack-trove | 17:17 | |
| *** jasonb365 has joined #openstack-trove | 17:30 | |
| *** jasonb365 has quit IRC | 17:30 | |
| *** amytron_ has joined #openstack-trove | 17:37 | |
| *** amytron has quit IRC | 17:38 | |
| *** amytron_ is now known as amytron | 17:38 | |
| *** SushilKM has quit IRC | 17:38 | |
| *** yogesh has joined #openstack-trove | 17:40 | |
| *** romainh has quit IRC | 17:48 | |
| *** SushilKM has joined #openstack-trove | 17:51 | |
| *** yogesh has quit IRC | 17:55 | |
| *** SnowDust has joined #openstack-trove | 17:59 | |
| *** demorris has quit IRC | 18:00 | |
| *** jasonb365 has joined #openstack-trove | 18:01 | |
| *** yogesh has joined #openstack-trove | 18:11 | |
| *** yogesh has quit IRC | 18:13 | |
| *** jasonb365 has quit IRC | 18:28 | |
| *** datsun180b has joined #openstack-trove | 18:30 | |
| *** jlucci has joined #openstack-trove | 18:31 | |
| *** datsun180b has quit IRC | 18:35 | |
| *** datsun180b has joined #openstack-trove | 18:36 | |
| *** yidclare has quit IRC | 18:37 | |
| *** SushilKM has quit IRC | 18:37 | |
| *** jasonb365 has joined #openstack-trove | 18:46 | |
| openstackgerrit | Tim Simpson proposed a change to openstack/trove: Fixes trove-api daemon fake mode functionality. https://review.openstack.org/51262 | 18:47 |
| *** jasonb365 has quit IRC | 18:50 | |
| *** demorris has joined #openstack-trove | 18:55 | |
| *** vipul is now known as vipul-away | 18:56 | |
| *** vipul-away is now known as vipul | 18:56 | |
| *** jasonb365 has joined #openstack-trove | 18:59 | |
| *** yogesh has joined #openstack-trove | 19:00 | |
| *** jasonb365 has quit IRC | 19:01 | |
| *** demorris has quit IRC | 19:05 | |
| *** jasonb365 has joined #openstack-trove | 19:06 | |
| *** demorris has joined #openstack-trove | 19:09 | |
| *** romainh has joined #openstack-trove | 19:11 | |
| *** vipul is now known as vipul-away | 19:11 | |
| *** kevinconway has quit IRC | 19:16 | |
| *** mmcdaris has joined #openstack-trove | 19:19 | |
| openstackgerrit | Ed Cranford proposed a change to openstack/trove: Conductor proxies host db access for guests https://review.openstack.org/45116 | 19:19 |
| *** mmcdaris has quit IRC | 19:20 | |
| *** romainh1 has joined #openstack-trove | 19:20 | |
| *** romainh has quit IRC | 19:20 | |
| *** kevinconway has joined #openstack-trove | 19:21 | |
| SnowDust | hello guys | 19:22 |
| SnowDust | any news on tempest ? | 19:22 |
| *** grapex has joined #openstack-trove | 19:29 | |
| *** yogesh has quit IRC | 19:32 | |
| *** romainh1 has quit IRC | 19:33 | |
| *** yogesh has joined #openstack-trove | 19:34 | |
| grapex | Hi everyone- the agenda for this week is a tad sparse if anyone has something they'd like to add: https://wiki.openstack.org/wiki/Meetings/TroveMeeting | 19:39 |
| *** yogesh has quit IRC | 19:40 | |
| SnowDust | grapex tempest ? | 19:40 |
| grapex | SnowDust: Sure | 19:41 |
| grapex | Wasn't hub_cap on that? He's away for the day unfortunately. | 19:41 |
| SnowDust | yeah .. thats why i put it here | 19:41 |
| SnowDust | am curious :) | 19:41 |
| SnowDust | i heard few more to join him in tempest effort | 19:42 |
| grapex | SnowDust: Ok, then feel free to add it. There may be no response today unless someone's heard something recently from hub_cap. | 19:42 |
| *** Barker has quit IRC | 19:43 | |
| SnowDust | ok added | 19:45 |
| grapex | SnowDust: Thanks | 19:50 |
| *** jlucci has left #openstack-trove | 19:55 | |
| *** ashestakov has quit IRC | 19:55 | |
| datsun180b | https://review.openstack.org/#/c/45116/ i think i've appeased the beast appropriately | 19:58 |
| datsun180b | thank you by the way for +2ing and merging the necessary plumbing already | 19:58 |
| *** vipul-away is now known as vipul | 19:59 | |
| *** dmakogon has joined #openstack-trove | 20:01 | |
| openstackgerrit | Steve Leon proposed a change to openstack/trove: Handling security group error when booting an instance https://review.openstack.org/52200 | 20:02 |
| *** demorris has quit IRC | 20:02 | |
| *** yogeshmehra has quit IRC | 20:05 | |
| *** yogesh has joined #openstack-trove | 20:05 | |
| *** jmontemayor has quit IRC | 20:06 | |
| *** Barker has joined #openstack-trove | 20:06 | |
| *** romainh has joined #openstack-trove | 20:08 | |
| esp | fyi, I have a few fixes for trove-int regarding rd-client | 20:08 |
| esp | anyone know if we have bugs for this yet? | 20:09 |
| *** aignatov has quit IRC | 20:11 | |
| *** aignatov has joined #openstack-trove | 20:11 | |
| SlickNik | esp: I haven't seen any bugs for trove-int. | 20:12 |
| SlickNik | Feel free to file them as you see fit. | 20:12 |
| esp | yeah I just encounter them today. you gotta build a new redstack (trove-integration) I'll create a bug and submit a patch. | 20:13 |
| amcrn | esp: i believe hub_cap said earlier yesterday that the plan was to rip out rd-client? | 20:17 |
| esp | ah, maybe I should hold off then. thx amcrn :) | 20:17 |
| amcrn | np | 20:17 |
| *** ashestakov has joined #openstack-trove | 20:19 | |
| dmakogon | amcrn, i had fixed patch according you comments | 20:19 |
| dmakogon | amcrn, https://review.openstack.org/#/c/50686/ | 20:20 |
| amcrn | i saw, i'll re-review by eod | 20:20 |
| amcrn | stuck doing some good ol' powerpoint at the moment | 20:20 |
| amcrn | :| | 20:20 |
| ashestakov | amcrn: hey, https://review.openstack.org/#/c/47936/ | 20:21 |
| datsun180b | more like just "ol' powerpoint" | 20:21 |
| amcrn | preaching to the choir ;) | 20:21 |
| datsun180b | https://review.openstack.org/#/c/45116/ the robots are happy | 20:25 |
| *** demorris has joined #openstack-trove | 20:31 | |
| redthrux | i meant to ask - I had to miss last meeting - are we moving ancilliary services after instances come active? | 20:37 |
| *** demorris has quit IRC | 20:39 | |
| redthrux | vipul: ^ i think we were going to talk about it and I had to give my talking points to someone else | 20:39 |
| vipul | redthrux: we decided against it because it meant that the service status potentially wouldn't be accurately reflected | 20:40 |
| vipul | redthrux: I was just revisiting your DNS patch.. | 20:41 |
| redthrux | ah - okay - | 20:41 |
| dmakogon | lol, fastest meeting ever | 20:43 |
| kevinconway | dmakogon: you can get to sleep early tonight | 20:43 |
| dmakogon | kevinconway, would i ? | 20:43 |
| redthrux | :D he never sleeps! | 20:43 |
| redthrux | he's a machine! | 20:43 |
| dmakogon | redthrux, with cold heart | 20:44 |
| SlickNik | dmakogon: so that you're ready for the 6 a.m. meeting that kevinconway scheduled. | 20:44 |
| redthrux | lol dmakogon | 20:44 |
| dmakogon | 6 AM PST ? | 20:44 |
| kevinconway | CST | 20:44 |
| dmakogon | codereviewing process http://500px.com/photo/49219126?from=popular | 20:44 |
| openstackgerrit | Ed Cranford proposed a change to openstack/trove: Conductor proxies host db access for guests https://review.openstack.org/45116 | 20:45 |
| datsun180b | dmakogon: ^^ thanks | 20:45 |
| *** ashestakov has quit IRC | 20:46 | |
| vipul | redthrux: rebase and i'll reapprove | 20:46 |
| redthrux | cool that's what i figured - rebase time | 20:46 |
| dmakogon | datsun180b: np, i'll take a look at it closer tomorrow | 20:46 |
| dmakogon | vipul: what about reviewing my refactoring ? | 20:47 |
| vipul | dmakogon: yep will get there | 20:47 |
| dmakogon | vipul, thanks, very | 20:47 |
| dmakogon | http://500px.com/DenisM - photos made by mine | 20:49 |
| kevinconway | dmakogon: NSFW? | 20:49 |
| dmakogon | kevinconway, you could not see photos with NSFW tah | 20:50 |
| dmakogon | *tag | 20:50 |
| dmakogon | that is my favorite http://500px.com/photo/4234857 | 20:51 |
| *** radez is now known as radez_g0n3 | 20:52 | |
| *** SnowDust has quit IRC | 20:52 | |
| openstackgerrit | Erik Redding proposed a change to openstack/trove: Moved create_instance DNS attempt to post-prepare https://review.openstack.org/48188 | 21:04 |
| *** pdmars has quit IRC | 21:04 | |
| *** yogesh has quit IRC | 21:07 | |
| *** romainh has left #openstack-trove | 21:09 | |
| *** yogesh has joined #openstack-trove | 21:12 | |
| *** yogesh has quit IRC | 21:13 | |
| openstackgerrit | Erik Redding proposed a change to openstack/trove: Moved create_instance DNS attempt to post-prepare https://review.openstack.org/48188 | 21:13 |
| *** yogesh has joined #openstack-trove | 21:23 | |
| *** yogesh has quit IRC | 21:36 | |
| *** yidclare has joined #openstack-trove | 21:37 | |
| *** jmontemayor has joined #openstack-trove | 21:38 | |
| *** Barker has quit IRC | 21:45 | |
| datsun180b | so here's a thing that is a thing | 21:45 |
| datsun180b | https://github.com/openstack/trove/blob/master/trove/guestagent/backup/backupagent.py#L98 | 21:45 |
| datsun180b | that bkup reference is outside the scope of the with block it was created in | 21:45 |
| datsun180b | that looks like a problem | 21:46 |
| robertmyers | datsun180b: tis magic! | 21:49 |
| datsun180b | robertmyers: black magic | 21:49 |
| *** Barker has joined #openstack-trove | 21:49 | |
| robertmyers | datsun180b: https://github.com/openstack/trove/blob/master/trove/guestagent/backup/backupagent.py#L78 | 21:51 |
| robertmyers | that too | 21:51 |
| datsun180b | yeah, i'm trying to pull those into conductor functions and saw them both | 21:51 |
| datsun180b | so is my concern valid here or what | 21:51 |
| robertmyers | yes, that should only be used in the with statement | 21:52 |
| datsun180b | i'll replace magic with provable code | 21:53 |
| *** amytron has quit IRC | 21:55 | |
| vipul | it does seem to be getting set correctly.. possibly by something else | 21:55 |
| datsun180b | i'll try swapping the with and the try | 21:56 |
| datsun180b | that else needs the backup type | 21:57 |
| *** yogesh has joined #openstack-trove | 21:57 | |
| *** robertmyers has quit IRC | 22:03 | |
| *** Barker has quit IRC | 22:05 | |
| *** Barker has joined #openstack-trove | 22:07 | |
| *** Nate2 has joined #openstack-trove | 22:09 | |
| *** yogesh has quit IRC | 22:10 | |
| dmakogon | grapex | 22:10 |
| grapex | dmakogon: Hi | 22:10 |
| grapex | Quick question to the room | 22:10 |
| grapex | vipul SlickNik others: ^ | 22:10 |
| vipul | shoot | 22:10 |
| dmakogon | hi, again | 22:10 |
| grapex | dmakogon currently have a disagreement on what should happen if a Trove instance is provisioned but the security group cannot be created in Nova. | 22:10 |
| grapex | The code that determines something went wrong in Nova lives in the trove-api | 22:11 |
| grapex | https://review.openstack.org/#/c/45708/39/trove/instance/models.py | 22:11 |
| *** Nate1 has quit IRC | 22:11 | |
| grapex | dmakogon thinks a SimpleInstance should be returned as its easier for the user to see | 22:11 |
| *** zigo_ has quit IRC | 22:11 | |
| grapex | I think an exception should be raised which means the user will see a 500 status code or some other status code. | 22:11 |
| grapex | Thoughts? | 22:11 |
| *** zigo has joined #openstack-trove | 22:11 | |
| dmakogon | grapex, not only me, esmute also | 22:11 |
| grapex | dmakogon: I believe esmute's concern was that quotas would be broken. | 22:12 |
| vipul | what is the behavior if DNS provisioning fails | 22:12 |
| vipul | we throw | 22:12 |
| grapex | That's in task manager so by that point the instance has to be set to an error state and the user must poll to find out it failed. | 22:12 |
| grapex | vipul: And I believe we do raise for DNS failures. | 22:12 |
| vipul | oh yes.. we don't do DNS here yet | 22:13 |
| datsun180b | if an instance asked for a secgroup and didn't get one, that's a paddlin' | 22:13 |
| esmute | grapex: If an exception is raised, the quota engine will think that the resource provisioning failed and it will not consume. | 22:13 |
| datsun180b | which is HTTP 500 Paddlin' | 22:13 |
| esmute | we can remove that quota roll back but that may open a different can of worm | 22:13 |
| vipul | So by this time, we have an DB instance record created... if we fail to create secgroup.. just mark the DB instance record as failed | 22:13 |
| vipul | and throw | 22:13 |
| dmakogon | vipul, +1 | 22:14 |
| esmute | throwing causes the quota to rollback the reservation | 22:14 |
| dmakogon | and if we would not stop process of instance provisioning, we would get an exception | 22:14 |
| dmakogon | at create server call | 22:14 |
| esmute | so the user will see an instance in error state but the quota will not reflect that instance | 22:14 |
| datsun180b | changing vote to mark-as-failed | 22:15 |
| dmakogon | datsun180b, +1 | 22:15 |
| esmute | what is the issue with marking as failed and returning? | 22:15 |
| grapex | Maybe the issue here is security groups should be provisioned not in the API but in task manager | 22:15 |
| vipul | Yes definitely.. | 22:15 |
| esmute | grapex: that will fix things too. | 22:15 |
| dmakogon | grapex, we shouldn't move it | 22:16 |
| grapex | Because the API does some very quick and limited book keeping stuff, like storing it in the database. Calling Nova may be out of scope since it can take longer. | 22:16 |
| grapex | dmakogon: How come? | 22:16 |
| dmakogon | grapex, when heat comes all this code would be deleted as well | 22:16 |
| *** rnirmal has quit IRC | 22:16 | |
| dmakogon | as i remember, we already decided not to move code | 22:16 |
| dmakogon | only make it more safe | 22:17 |
| esmute | dmakogon: When heat comes out, we would have to remove the secgroup provsioning from the api anyways right? | 22:17 |
| dmakogon | esmute, yes | 22:17 |
| grapex | dmakogon: I believe we decided Heat was a reason to not write a bunch of code to automatically roll back all provisioned resources when code in the task manager detected a problem. | 22:17 |
| dmakogon | esmute, heat would manage all resources | 22:17 |
| grapex | This isn't the same thing- its just existing code that should live in taskmanager to be more consistent with current semantics. | 22:18 |
| esmute | so why not move it to the TM. That way, when the heat work comes, the api would not be affected | 22:18 |
| grapex | esmute: Agreed. If we don't move it now, it will mean when we move to Heat the API might change behavior. | 22:18 |
| dmakogon | grapex, what the difference between removing code from TM and API ? | 22:19 |
| grapex | dmakogon: Sorry, I should've realized this earlier | 22:19 |
| grapex | dmakogon: Not removing it, just moving it. | 22:19 |
| vipul | you should get the code in place where we can repalce the create_instance method of Taskmanager and make that Heat | 22:19 |
| dmakogon | grapex, code should stay where it is | 22:19 |
| vipul | the way the code is right now, you can't do that | 22:19 |
| esmute | whether with have Heat or not, the provisioning of secgroup should not be in the api | 22:20 |
| vipul | esmute: agreed | 22:20 |
| vipul | You could also do it post-nova boot if you want to | 22:21 |
| dmakogon | what is the differense where to create SG ? | 22:21 |
| redthrux | vipul++ | 22:21 |
| vipul | dmakogon: run_with_quotas | 22:22 |
| dmakogon | we can avoid it | 22:22 |
| dmakogon | as we discussed earlier | 22:22 |
| *** jmontemayor has quit IRC | 22:24 | |
| vipul | so the quota framework will handle committing / rolling quota back if you provision things within the run_with_quotas method | 22:24 |
| vipul | If you dont'.. then you have to do things like not throw an exception | 22:24 |
| *** jasonb365 has quit IRC | 22:25 | |
| esmute | im still not clear why we cant have the secgroup provision in the TM? At least for now until heat comes into the picture | 22:26 |
| *** jmontemayor has joined #openstack-trove | 22:26 | |
| grapex | An observation- looking at the create method, I think the code is a bit misleading, because on lines 477-481 we do some DNS stuff so it looks like we're provisioning resources in the API. But actually, that code just sets things in the database for task manager to do later (this was a request from operations who said it made it easier to track instance provisioning). So actually DNS is not provisioned anywhere but taskmanager. | 22:27 |
| redthrux | grapex thats correct | 22:27 |
| redthrux | dns casts are just in taskmanager | 22:27 |
| redthrux | i have to drop off - feel free to cast questions; I'm async | 22:28 |
| dmakogon | you are offering to mess up TM create method code | 22:29 |
| grapex | dmakogon: I think if you move it you'll see it won't be that bad. It'll actually make it somewhat easier to deal with exceptions from Nova. | 22:30 |
| dmakogon | it seems not so critical to move code | 22:30 |
| dmakogon | i could do that, but in other patch | 22:37 |
| dmakogon | ok wit that ? | 22:38 |
| datsun180b | i'd submit this conductor-backup change if the unittests weren't frozen in time | 22:38 |
| esmute | dmakogon: Why not do it in this patch? | 22:39 |
| grapex | dmakogon: I disagree. I think returning a SimpleInstance introduces a problem with this patch in that its so nonstandard. Since the goal is to handle the error from Nova correctly moving it to TM makes sense. | 22:39 |
| *** paul_lodronio has left #openstack-trove | 22:42 | |
| *** vipul is now known as vipul-away | 22:45 | |
| openstackgerrit | Ed Cranford proposed a change to openstack/trove: Use conductor to update backup status https://review.openstack.org/52241 | 22:50 |
| datsun180b | neat, that worked | 22:50 |
| datsun180b | the review, not the unittests | 22:50 |
| *** jcru has quit IRC | 22:52 | |
| *** datsun180b has quit IRC | 22:58 | |
| *** grapex has quit IRC | 22:58 | |
| *** tanisdl has quit IRC | 23:08 | |
| *** vipul-away is now known as vipul | 23:15 | |
| *** dmakogon has quit IRC | 23:30 | |
| *** jmontemayor has quit IRC | 23:30 | |
| *** Barker has quit IRC | 23:31 | |
| *** saurabhs1 has joined #openstack-trove | 23:34 | |
| *** saurabhs has quit IRC | 23:35 | |
| *** adrian_otto has joined #openstack-trove | 23:38 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!