| *** krow has joined #openstack-trove | 00:05 | |
| *** matsuhashi has joined #openstack-trove | 00:28 | |
| *** agliodbs has joined #openstack-trove | 00:35 | |
| agliodbs | howdy! <-- Josh Berkus | 00:35 |
|---|---|---|
| *** nosnos has joined #openstack-trove | 00:38 | |
| *** rnirmal has quit IRC | 00:39 | |
| *** krow has quit IRC | 00:42 | |
| *** krow has joined #openstack-trove | 00:46 | |
| *** krow has quit IRC | 01:01 | |
| *** agliodbs has quit IRC | 01:04 | |
| *** radez_g0n3 is now known as radez | 01:29 | |
| *** krow has joined #openstack-trove | 01:31 | |
| SlickNik | hello Josh / agliodbs! | 01:34 |
| *** erkules_ has joined #openstack-trove | 01:35 | |
| *** krow has quit IRC | 01:35 | |
| *** erkules has quit IRC | 01:37 | |
| *** jaishanker has joined #openstack-trove | 01:49 | |
| *** amcrn has quit IRC | 01:52 | |
| *** yogeshmehra has joined #openstack-trove | 02:49 | |
| *** yogeshmehra has quit IRC | 02:54 | |
| *** krow has joined #openstack-trove | 03:09 | |
| *** krow has quit IRC | 03:13 | |
| *** matsuhashi has quit IRC | 03:15 | |
| *** amytron has quit IRC | 03:18 | |
| *** jaishanker has quit IRC | 03:23 | |
| haomaiwang | ping hub_cap | 03:29 |
| *** yogeshmehra has joined #openstack-trove | 03:51 | |
| openstackgerrit | Erik Redding proposed a change to openstack/trove: Moved create_instance DNS attempt to post-prepare https://review.openstack.org/48188 | 03:51 |
| *** yogeshmehra has quit IRC | 03:56 | |
| *** matsuhashi has joined #openstack-trove | 03:56 | |
| *** radez is now known as radez_g0n3 | 04:03 | |
| *** josh__ has joined #openstack-trove | 04:05 | |
| *** josh__ is now known as agliodbs | 04:05 | |
| *** krow has joined #openstack-trove | 04:15 | |
| *** krow has quit IRC | 04:37 | |
| *** yogeshmehra has joined #openstack-trove | 04:51 | |
| *** yogeshmehra has quit IRC | 04:57 | |
| *** yogeshmehra has joined #openstack-trove | 05:00 | |
| *** SushilKM has joined #openstack-trove | 05:01 | |
| *** SushilKM has quit IRC | 05:05 | |
| *** SushilKM has joined #openstack-trove | 05:06 | |
| *** krow has joined #openstack-trove | 05:07 | |
| *** adrian_otto has joined #openstack-trove | 05:15 | |
| *** dmakogon_ has joined #openstack-trove | 05:17 | |
| *** Kapil has joined #openstack-trove | 05:22 | |
| openstackgerrit | Michael Basnight proposed a change to openstack/python-troveclient: Massive refactoring to the troveclient. - WIP until further notice. https://review.openstack.org/48576 | 05:26 |
| *** yogeshmehra has quit IRC | 05:27 | |
| *** agliodbs has quit IRC | 05:36 | |
| *** yogeshmehra has joined #openstack-trove | 05:41 | |
| *** yogeshmehra has quit IRC | 05:43 | |
| *** krow has quit IRC | 05:46 | |
| *** josh__ has joined #openstack-trove | 05:47 | |
| *** josh__ is now known as agliodbs | 05:47 | |
| *** krow has joined #openstack-trove | 05:49 | |
| *** adrian_otto has quit IRC | 06:03 | |
| *** SushilKM has quit IRC | 06:09 | |
| *** agliodbs has quit IRC | 06:09 | |
| *** SnowDust has joined #openstack-trove | 07:00 | |
| *** SnowDust has quit IRC | 07:03 | |
| *** SnowDust has joined #openstack-trove | 07:09 | |
| *** SnowDust has joined #openstack-trove | 07:09 | |
| *** krow has quit IRC | 07:15 | |
| *** yogesh has quit IRC | 07:16 | |
| SnowDust | anyone to help with heat implementation | 07:26 |
| SnowDust | ? | 07:26 |
| SnowDust | hub_cap: u thr buddy ? | 07:26 |
| SnowDust | SlickNik: can u help ? | 07:26 |
| *** ashestakov_ has joined #openstack-trove | 07:49 | |
| *** dmakogon_ has quit IRC | 07:56 | |
| openstackgerrit | Illia Khudoshyn proposed a change to openstack/trove: Extract generic part of GA code from MySQL specific modules https://review.openstack.org/48435 | 08:18 |
| *** krow has joined #openstack-trove | 08:29 | |
| *** erkules_ is now known as erkules | 08:44 | |
| openstackgerrit | Denis M. proposed a change to openstack/trove: Update statuses on GA timeout https://review.openstack.org/45723 | 08:59 |
| *** rushiagr has joined #openstack-trove | 09:17 | |
| openstackgerrit | Illia Khudoshyn proposed a change to openstack/trove: Extract generic part of GA code from MySQL specific modules https://review.openstack.org/48435 | 09:27 |
| *** rushiagr has quit IRC | 09:35 | |
| *** rushiagr has joined #openstack-trove | 09:38 | |
| *** krow has quit IRC | 09:44 | |
| *** rushiagr has quit IRC | 09:59 | |
| *** rushiagr has joined #openstack-trove | 10:01 | |
| openstackgerrit | Denis M. proposed a change to openstack/trove: Trove DNS refactoring. https://review.openstack.org/45644 | 10:02 |
| *** Kapil has quit IRC | 10:24 | |
| *** Kapil has joined #openstack-trove | 10:25 | |
| *** esmute has quit IRC | 10:37 | |
| openstackgerrit | Illia Khudoshyn proposed a change to openstack/trove: Extract generic part of GA code from MySQL specific modules https://review.openstack.org/48435 | 10:44 |
| *** esmute has joined #openstack-trove | 10:46 | |
| *** matsuhashi has quit IRC | 10:52 | |
| *** ikhudoshyn has quit IRC | 11:02 | |
| *** dmakogon has quit IRC | 11:03 | |
| *** ikhudoshyn has joined #openstack-trove | 11:08 | |
| *** ikhudoshyn has quit IRC | 11:11 | |
| *** ikhudoshyn has joined #openstack-trove | 11:12 | |
| *** ashestakov_ has quit IRC | 11:21 | |
| *** ashestakov_ has joined #openstack-trove | 11:24 | |
| *** matsuhashi has joined #openstack-trove | 11:44 | |
| *** rushiagr has quit IRC | 11:52 | |
| *** ashestakov_ has quit IRC | 12:00 | |
| *** matsuhashi has quit IRC | 12:01 | |
| *** pdmars has joined #openstack-trove | 12:03 | |
| *** matsuhashi has joined #openstack-trove | 12:10 | |
| *** matsuhashi has quit IRC | 12:11 | |
| *** radez_g0n3 is now known as radez | 12:34 | |
| *** matsuhashi has joined #openstack-trove | 12:35 | |
| *** Kapil has quit IRC | 12:41 | |
| *** SushilKM has joined #openstack-trove | 12:43 | |
| *** jrodom has joined #openstack-trove | 12:51 | |
| *** jrodom has joined #openstack-trove | 12:51 | |
| *** haomaiwang has quit IRC | 13:04 | |
| *** haomaiwang has joined #openstack-trove | 13:05 | |
| openstackgerrit | Illia Khudoshyn proposed a change to openstack/trove: Extract generic part of GA code from MySQL specific modules https://review.openstack.org/48435 | 13:12 |
| *** SushilKM has quit IRC | 13:15 | |
| *** nosnos has quit IRC | 13:15 | |
| *** robertmyers has joined #openstack-trove | 13:19 | |
| *** adrian_otto has joined #openstack-trove | 13:24 | |
| *** jcru has joined #openstack-trove | 13:28 | |
| *** robertmyers has quit IRC | 13:40 | |
| *** djohnstone has joined #openstack-trove | 13:44 | |
| *** robertmyers has joined #openstack-trove | 13:50 | |
| *** amytron has joined #openstack-trove | 13:52 | |
| *** adrian_otto has quit IRC | 14:03 | |
| *** SnowDust has quit IRC | 14:08 | |
| *** ashestakov_ has joined #openstack-trove | 14:13 | |
| *** rnirmal has joined #openstack-trove | 14:13 | |
| *** shakayumi has joined #openstack-trove | 14:15 | |
| *** dmakogon has joined #openstack-trove | 14:16 | |
| *** KennethWilke has joined #openstack-trove | 14:20 | |
| *** datsun180b has joined #openstack-trove | 14:23 | |
| *** adrian_otto has joined #openstack-trove | 14:24 | |
| *** shakayumi has quit IRC | 14:28 | |
| *** Barker has joined #openstack-trove | 14:30 | |
| dmakogon | anybody home ? | 14:36 |
| *** SushilKM has joined #openstack-trove | 14:37 | |
| *** aj_ has joined #openstack-trove | 14:41 | |
| *** adrian_otto has quit IRC | 14:49 | |
| *** matsuhashi has quit IRC | 14:51 | |
| *** Nate2 has joined #openstack-trove | 14:53 | |
| datsun180b | a few of us | 14:55 |
| dmakogon | jenkins doesn't working at all | 14:56 |
| *** adrian_otto has joined #openstack-trove | 15:05 | |
| *** tanisdl has joined #openstack-trove | 15:07 | |
| *** adrian_otto has quit IRC | 15:08 | |
| datsun180b | i haven't looked at -infra lately, maybe they're talking about it | 15:17 |
| *** SushilKM has quit IRC | 15:26 | |
| *** jmontemayor has joined #openstack-trove | 15:33 | |
| *** SushilKM has joined #openstack-trove | 15:37 | |
| *** adrian_otto has joined #openstack-trove | 15:53 | |
| *** jodom has joined #openstack-trove | 16:04 | |
| *** Barker has quit IRC | 16:06 | |
| *** jrodom has quit IRC | 16:08 | |
| *** jodom has quit IRC | 16:18 | |
| *** jrodom has joined #openstack-trove | 16:18 | |
| *** josh_513 has joined #openstack-trove | 16:22 | |
| *** josh_513 is now known as agliodbs | 16:22 | |
| *** tanisdl has quit IRC | 16:22 | |
| *** dmakogon_ has joined #openstack-trove | 16:36 | |
| *** SushilKM has quit IRC | 16:39 | |
| *** Barker has joined #openstack-trove | 16:53 | |
| *** yogeshmehra has joined #openstack-trove | 16:58 | |
| yogeshmehra | ping hub_cap | 16:58 |
| *** dmakogon__ has joined #openstack-trove | 16:59 | |
| *** paul_lodronio has joined #openstack-trove | 17:08 | |
| openstackgerrit | Dan Nguyen proposed a change to openstack/trove: Added server_id to my.cnf template logic https://review.openstack.org/47206 | 17:19 |
| aj_ | hi. I am new to openstack. I have worked with drizzle database as a part of gsoc. I would like to contribute to trove. Can anybody suggest some beginner level tasks? | 17:26 |
| *** SnowDust has joined #openstack-trove | 17:28 | |
| esp | hello aj_ you probably want to talk to hub_cap for tasks but have had a chance to look through the wiki and set up your environment yet? | 17:31 |
| esp | aj_: https://wiki.openstack.org/wiki/Trove https://wiki.openstack.org/wiki/Trove#Installation_and_Deployment | 17:34 |
| juice | ashestakov_: what is the latest with the many flavors of flavor | 17:35 |
| ashestakov_ | juice: mean? | 17:35 |
| juice | just wondering where you are at with getting some consistency on our use of flavor | 17:36 |
| juice | ref vs id vs uuid | 17:36 |
| ashestakov_ | as i checked, trove gets id from flavorRef and pushes to nova, i not sure why trove have deals with refs | 17:37 |
| ashestakov_ | i propose to change apischema to allow uuid instad/addition to refs | 17:38 |
| ashestakov_ | and remove validation on list | 17:38 |
| juice | for backwards compatibility it would need to be "in addition to" | 17:38 |
| cweid | Hey are any of you having troubles using the default redstack setup? All of my instances keep falling into ERROR status. It seems to be related to a volume prov error. Here is the gist https://gist.github.com/cweidenkeller/6732148 | 17:39 |
| cweid | Not sure if any of you have seen that before I am just stumped.. | 17:39 |
| ashestakov_ | cweid: can you look to cinder logs? | 17:41 |
| ashestakov_ | juice: flavors get method is not implemented in cli, why? | 17:41 |
| juice | oversite | 17:42 |
| juice | or is that oversight | 17:43 |
| ashestakov_ | juice: so we can skip validation in flavors list | 17:43 |
| juice | are there any parameters for flavor list | 17:43 |
| ashestakov_ | no | 17:44 |
| ashestakov_ | it gets flavors from nova as is | 17:44 |
| juice | so there would be no validation on that | 17:44 |
| juice | ashestakov_: so i was just reviewing nova's api and at least through the cli there is no exposure of the UUID for the flavor | 17:49 |
| ashestakov_ | https://github.com/openstack/python-novaclient/blob/master/novaclient/v1_1/flavors.py#L133 | 17:51 |
| ashestakov_ | hub_cap vipul SlickNik around? | 17:53 |
| vipul | ashestakov_: what's up | 17:54 |
| ashestakov_ | vipul: can you start to review my changes for service_types? | 17:54 |
| vipul | ashestakov_: Yes I can take a look today | 17:54 |
| juice | ashestakov_: there is also not a uuid field in the "instance_types" (flavor) table in nova | 17:54 |
| juice | ashestakov_: have you tried passing in "auto" to see if this indeed does what the documentation says it will do? | 17:56 |
| ashestakov_ | juice: yep, you should specify flavor object, which can be retrived here https://github.com/openstack/python-novaclient/blob/master/novaclient/v1_1/flavors.py#L112 | 17:56 |
| ashestakov_ | here is my flavors | 17:58 |
| ashestakov_ | nova flavor-list | 17:58 |
| ashestakov_ | +--------------------------------------+---------------+-----------+------+-----------+------+-------+-------------+-----------+ | 17:58 |
| ashestakov_ | | ID | Name | Memory_MB | Disk | Ephemeral | Swap | VCPUs | RXTX_Factor | Is_Public | | 17:58 |
| ashestakov_ | +--------------------------------------+---------------+-----------+------+-----------+------+-------+-------------+-----------+ | 17:58 |
| ashestakov_ | | 1 | m1.tiny | 512 | 0 | 0 | | 1 | 1.0 | True | | 17:58 |
| ashestakov_ | | 2 | m1.small | 2048 | 20 | 0 | | 1 | 1.0 | True | | 17:58 |
| ashestakov_ | | 3 | m1.medium | 4096 | 40 | 0 | | 2 | 1.0 | True | | 17:58 |
| ashestakov_ | | 5 | m1.xlarge | 16384 | 160 | 0 | | 8 | 1.0 | True | | 17:58 |
| ashestakov_ | | 7b67e99a-4289-48d7-819f-e55875343e05 | m1.small_disk | 1024 | 10 | 1 | | 2 | 1.0 | True | | 17:58 |
| ashestakov_ | | 91961aa7-ab31-429c-8d8f-8578d0cc213c | m1.large | 8192 | 80 | 1 | | 4 | 1.0 | True | | 17:58 |
| ashestakov_ | +--------------------------------------+---------------+-----------+------+-----------+------+-------+-------------+-----------+ | 17:58 |
| juice | ashestakov_: there's the proof :) | 17:58 |
| juice | ok well it seems that you do INDEED need to open up the flavor show to allow for uuid type | 17:59 |
| ashestakov_ | i just want to use it, i can specify href on instance create, but why i cannot specify just uuid | 17:59 |
| juice | change it in two places - change it in instance create AND flavor show | 17:59 |
| ashestakov_ | and as i said before trove-cli flavor list fails on my env | 18:00 |
| juice | as it does on mine | 18:01 |
| juice | and I do not have UUIDs for flavor IDs - so there must be something else going on | 18:01 |
| juice | looks like you may have found yourself a bug to fix :) | 18:02 |
| juice | ashestakov_: hold that last comment - my trove/nova env is not in a stable state. | 18:03 |
| *** krow has joined #openstack-trove | 18:03 | |
| ashestakov_ | juice: when you do edit flavor in horizon, do it change id to uuid on you env? | 18:04 |
| juice | ashestakov_: i only use the cli or curl directly. I haven't used horizon in some time | 18:05 |
| juice | also after restarting my trove/nova stack, I can successfully get the flavors from trove cli | 18:05 |
| hub_cap | ashestakov_ hey im around, and plz gist stuff like that lol | 18:06 |
| ashestakov_ | hey, sure :) | 18:07 |
| hub_cap | so yes horizon is putting flavors as uuid's | 18:07 |
| hub_cap | its not necessary persay, but horizon is | 18:07 |
| hub_cap | so as long as it doesnt break compatibility we sholud be able to lift the "ban" on string (uuid) for flavor id | 18:07 |
| juice | ashestakov_: confirmed that flavor with UUID ID will cause flavor list to choke | 18:07 |
| ashestakov_ | there is issue with flavor tests... | 18:10 |
| hub_cap | there are probably multiple issues w/ the app | 18:10 |
| hub_cap | weve always behaved like the flavors are ints | 18:10 |
| hub_cap | if horizon is inserting them as uuid's i think its safe to assume thats the "expected behavior" in openstack now | 18:10 |
| ashestakov_ | it creating fake flavor in fake nove with id of int, but when it gets id from flavorRef="https://localhost:8779/v1.0/2500/flavors/1" - value is string, and couldnt be found | 18:11 |
| juice | ashestakov_: look at flavor/views.py line 34 - that's not good | 18:11 |
| juice | hub_cap: ashestakov_: it is all stored as varchar but it either expects and int or "auto" which it then generates a UUID | 18:12 |
| ashestakov_ | juice: i fixed that one in my change | 18:13 |
| *** dmakogon__ has quit IRC | 18:14 | |
| *** dmakogon_ has quit IRC | 18:14 | |
| hub_cap | auto is a trove thing right? | 18:15 |
| *** ppenjoy has joined #openstack-trove | 18:15 | |
| *** dmakogon_ has joined #openstack-trove | 18:15 | |
| ppenjoy | hi guys I have volume that is stock in Deleted in horizon , how can I delete it ? here my database : http://paste.openstack.org/show/47611/ the record is : c4b1212b-0257-492b-b6bb-9cec09562e21 | 18:15 |
| *** dmakogon__ has joined #openstack-trove | 18:15 | |
| ashestakov_ | looks like flavor id should be always string | 18:16 |
| ashestakov_ | hub_cap: auto? | 18:16 |
| hub_cap | "auto" being the option to store the varchar (what juice said) | 18:17 |
| ashestakov_ | do you mean in novaclient on flavor create? | 18:18 |
| juice | yes | 18:20 |
| juice | i just tried creating a flavor with an ID value = "bobcat" and nova didn't complain | 18:20 |
| juice | so basically it CAN be anything | 18:20 |
| juice | do we want to ALLOW anything is the question | 18:21 |
| ashestakov_ | not anything, only id, uuid, refs | 18:21 |
| juice | having said that, if nova does not place any restriction and we do, we will caused impedance mismatches | 18:21 |
| juice | ./redstack nova-client flavor-create jrh.xl bobcat 512 1 1 | 18:22 |
| juice | nova created the flavor for me just fine | 18:22 |
| cweid | Yeah so if you disable trove_volume_support everything works ok. | 18:22 |
| cweid | Gonna dig into the cinder logs to see what is shakin' | 18:22 |
| cweid | I assume this is a problem with devstack. | 18:23 |
| *** dmakogon__ has quit IRC | 18:24 | |
| SnowDust | hub_cap: can u help me with the concept of "heatkey" in /trove/trove/taskmanager/models.py | 18:24 |
| *** dmakogon_ has quit IRC | 18:24 | |
| ashestakov_ | there is question, should we support id as int(), maybe just string with number? | 18:24 |
| *** krow has quit IRC | 18:24 | |
| *** dmakogon_ has joined #openstack-trove | 18:24 | |
| SnowDust | or anyone else there ? | 18:25 |
| SnowDust | its been a FRY DAY since i started looking at HEAT ..in trove ! | 18:26 |
| SnowDust | anyone thr to help me ..save my skin :) | 18:27 |
| yogeshmehra | Snowdust: i can help you | 18:27 |
| SnowDust | :) omg ! | 18:27 |
| yogeshmehra | Got through it... | 18:27 |
| SnowDust | kewl | 18:27 |
| SnowDust | lets take it private then | 18:28 |
| yogeshmehra | yup | 18:28 |
| juice | ashestakov_: my opinion is that we should not restrict it at all but document it as either a UUID or Int (without validation). | 18:29 |
| ashestakov_ | juice: can you link me this doc? | 18:30 |
| hub_cap | SnowDust: hi | 18:30 |
| *** tanisdl has joined #openstack-trove | 18:30 | |
| dmakogon_ | hi 2 all | 18:31 |
| dmakogon_ | guys, does anyone knows what happaning with jenkins ? | 18:31 |
| dmakogon_ | today it failed a lot time | 18:32 |
| hub_cap | dmakogon_ yes the gate is failing pretty steadily | 18:32 |
| hub_cap | there are 2 critical bugs causing almost every gate to fail | 18:32 |
| hub_cap | http://status.openstack.org/rechecks/ | 18:32 |
| dmakogon_ | gates are not failing | 18:32 |
| openstackgerrit | Dan Nguyen proposed a change to openstack/trove: Fixed method signature _create_server_volume_heat https://review.openstack.org/48535 | 18:32 |
| ashestakov_ | hub_cap: can you start to review service types? | 18:32 |
| hub_cap | and they are asking to not recheck (as per a email) | 18:32 |
| hub_cap | ashestakov_ sure | 18:32 |
| hub_cap | can u link it to me? | 18:32 |
| juice | ashestakov_: https://github.com/openstack/database-api/tree/master/apidocs | 18:33 |
| juice | there is also a dbapi repo somewhere | 18:33 |
| ashestakov_ | hub_cap: i just want to follow fail-fast strategy :) | 18:33 |
| dmakogon_ | hub_cap: gates are not starting | 18:33 |
| hub_cap | +1 ashestakov_ | 18:33 |
| hub_cap | :) | 18:33 |
| dmakogon_ | hub_cap: almost all day | 18:33 |
| hub_cap | they are overloaded | 18:33 |
| juice | ashestakov_: here https://github.com/openstack/database-api/blob/master/openstack-database-api/src/markdown/database-api-v1.md | 18:33 |
| hub_cap | http://status.openstack.org/zuul/ | 18:33 |
| ashestakov_ | juice: so, there id is string | 18:34 |
| ashestakov_ | and we should not use int() | 18:35 |
| SnowDust | hub_cap : :) good to see u back .. ! | 18:35 |
| SnowDust | i wanted to ask / share a bug in trove/trove/taskmanager/models.py | 18:35 |
| hub_cap | SnowDust: i know ive been gone for a while geeeeeez :P | 18:36 |
| hub_cap | SnowDust: what is it? | 18:36 |
| *** djohnstone has quit IRC | 18:36 | |
| *** djohnstone has joined #openstack-trove | 18:36 | |
| SlickNik | hey guys. | 18:36 |
| hub_cap | ashestakov_ i think we had a reason to do that a while ago (but im not sure if its a valid reason... i think it was a rackspace reason based on bugs in the QE depbartment) | 18:37 |
| hub_cap | *department | 18:37 |
| SnowDust | _create_server_volume_heat called with 6 arguments | 18:37 |
| SnowDust | but defined with 5 | 18:37 |
| SnowDust | AZ in call is extra .. | 18:37 |
| ashestakov_ | hub_cap: can we change it to string now? | 18:37 |
| juice | hey slicknik | 18:37 |
| SnowDust | i didnt know the intention so could not file / fix that .. | 18:37 |
| hub_cap | SnowDust: thats a bug | 18:37 |
| hub_cap | i saw a review for it | 18:37 |
| hub_cap | https://review.openstack.org/#/c/48535/ | 18:38 |
| hub_cap | that one right? | 18:38 |
| hub_cap | https://review.openstack.org/#/c/48535/2/trove/taskmanager/models.py SnowDust | 18:38 |
| SlickNik | Wanted to mention that I'm gonna be out most of the day today. | 18:38 |
| hub_cap | ok i have to go get lunch.. ill be back in ~30 min | 18:38 |
| hub_cap | cool SlickNik im just getting back lol (after lunch) | 18:39 |
| SnowDust | hmm .. | 18:39 |
| SnowDust | so AZ got included .. | 18:39 |
| ashestakov_ | hub_cap: https://review.openstack.org/#/c/47934/ | 18:39 |
| ashestakov_ | hub_cap: https://review.openstack.org/#/c/47936/ | 18:39 |
| hub_cap | thx ashestakov | 18:39 |
| ashestakov_ | i added tests to client, but pypy and py33 fails | 18:40 |
| hub_cap | SnowDust: its a new review from esp it just got uploaded last night | 18:40 |
| ashestakov_ | not sure coz my changes | 18:40 |
| hub_cap | ashestakov_ check other reviews | 18:40 |
| hub_cap | i dont think we gate pypy and py33 yet | 18:40 |
| hub_cap | ok im going | 18:40 |
| hub_cap | to lunch!!!! | 18:40 |
| hub_cap | ill be back in ~30 | 18:40 |
| SnowDust | yeah .. i shared that 24 hours ago :) | 18:40 |
| hub_cap | ahhhhhhhhhhh SnowDust nice thank you! | 18:40 |
| esp | hub_cap: eat something healthy | 18:40 |
| *** ppenjoy has left #openstack-trove | 18:41 | |
| dmakogon_ | guys, please, take a look https://review.openstack.org/#/c/45723 | 18:41 |
| dmakogon_ | and this https://review.openstack.org/#/c/45708/ | 18:41 |
| dmakogon_ | https://review.openstack.org/#/c/47541/ | 18:41 |
| dmakogon_ | https://review.openstack.org/#/c/47543/ | 18:42 |
| SnowDust | yes .. eat healthy :) | 18:42 |
| dmakogon_ | https://review.openstack.org/#/c/45075/ | 18:42 |
| esp | SnowDust: sorry about that bug, that was bad. I missed it original patch. | 18:42 |
| SnowDust | esp: welcome .. sweat and "heat" .. both worthy :) | 18:44 |
| esp | SnowDust: lol | 18:44 |
| *** adrian_otto has quit IRC | 18:44 | |
| SnowDust | esp and hub_cap .. we havent documented the "heatkey" thing anywhere .. | 18:46 |
| SnowDust | even not in the code .. | 18:46 |
| SnowDust | no one knows by reading code that there should be a keypair with name "heatkey" to make HEAT work | 18:47 |
| SnowDust | pls correct me if i am wrong | 18:47 |
| dmakogon_ | guys, have you ever though about replacing exec-style code in guestagent/pkg.py to python-apt, python-yum code ? | 18:47 |
| esp | SnowDust: yeah I didn't know much about it. but got help from Ilya Sviridov | 18:48 |
| SnowDust | yeah .. can we NOW .. do some comments in code atleast :) | 18:48 |
| SnowDust | it was a good joyride ..4 me .. around heatkey | 18:48 |
| esp | dmakogon: not sure anyone has looked into that as far as I knw | 18:49 |
| yogeshmehra | snowdust: http://docs.openstack.org/developer/heat/getting_started/on_devstack.html | 18:50 |
| esp | yogeshmehra: nice! | 18:50 |
| yogeshmehra | this describes heatkey but within heat standalone, out of trove | 18:50 |
| dmakogon_ | esp: https://blueprints.launchpad.net/trove/+spec/provide-improved-guestagent-packaging-install-system-without-execs | 18:50 |
| dmakogon_ | but i've got some issues with user permissions | 18:51 |
| dmakogon_ | if anyone would like to help, it would be appreciable | 18:52 |
| esp | probably a good topic for the next meeting or mailing list :) | 18:55 |
| dmakogon_ | esp: probably | 18:59 |
| esp | in generally seems like a good improvement though. I just don't know enough about that part of the code. | 19:00 |
| dmakogon_ | esp: it is enough easy | 19:03 |
| *** vipul is now known as vipul-away | 19:03 | |
| dmakogon_ | esp: just package managing execs | 19:03 |
| dmakogon_ | sudo apt-get install, sudo yum install | 19:03 |
| esp | yep | 19:04 |
| esp | doesn't seem too bad | 19:04 |
| dmakogon_ | execs in code is not so good | 19:04 |
| dmakogon_ | non-production style | 19:04 |
| datsun180b | iirc it's rsynced in or baked into the image, but packaging opens the door for upgrades for example | 19:13 |
| datsun180b | stress on "iirc" | 19:16 |
| SnowDust | bye@buddies | 19:20 |
| SnowDust | thnx all | 19:20 |
| dmakogon_ | datsun180b: +1 for upgrades | 19:22 |
| *** amcrn has joined #openstack-trove | 19:29 | |
| *** SnowDust has quit IRC | 19:33 | |
| *** amytron has quit IRC | 19:33 | |
| *** jmontemayor has quit IRC | 19:45 | |
| *** djohnstone1 has joined #openstack-trove | 19:51 | |
| openstackgerrit | Andrey Shestakov proposed a change to openstack/trove: Add support of UUID in flavors Removed int() conversion from flavors https://review.openstack.org/48735 | 19:51 |
| *** djohnstone2 has joined #openstack-trove | 19:53 | |
| *** djohnstone2 has left #openstack-trove | 19:53 | |
| *** djohnstone has quit IRC | 19:53 | |
| *** djohnstone1 has quit IRC | 19:55 | |
| *** Barker has quit IRC | 19:57 | |
| cp16net | hrm i dont understand why changing the flavors to strings | 20:00 |
| cp16net | does not compute to me | 20:00 |
| cp16net | ashestakov_: can you explain? | 20:00 |
| cp16net | i caught part of the convo i think about horizon | 20:00 |
| *** vipul-away is now known as vipul | 20:01 | |
| cp16net | vipul: hello | 20:01 |
| ashestakov_ | cp16net: we discussed before | 20:01 |
| ashestakov_ | cp16net: nova uses strings for id/uuid, and is not possible to convert uuid to int() | 20:02 |
| cp16net | ashestakov_: so this is specifically around flavors right? | 20:03 |
| cp16net | from the change i see | 20:03 |
| ashestakov_ | cp16net: yes | 20:03 |
| cp16net | but trove handles its own flavors | 20:03 |
| cp16net | do flavors have uuid's that nova expects? | 20:04 |
| ashestakov_ | cp16net: own flavors? | 20:04 |
| ashestakov_ | cp16net: https://github.com/openstack/trove/blob/master/trove/flavor/models.py#L40 it gets flavors from nova | 20:05 |
| cp16net | did something change? | 20:06 |
| cp16net | on the nova side? | 20:07 |
| ashestakov_ | cp16net: nova uses uuid in some cases, but trove doesnt supports it | 20:07 |
| ashestakov_ | please read chat log about this | 20:08 |
| hub_cap | ya we need to make sure the flavor/service_type code is also updated to accept this | 20:08 |
| cp16net | i see that nova flavors are "strings" | 20:08 |
| cp16net | of integers | 20:08 |
| amcrn | ashestakov_: Either the bug and/or the patch-set commit message should explain this, that way you don't have to refer people to chat logs :) | 20:09 |
| juice | cp16net: there is no restriction on what nova accepts for flavors | 20:10 |
| juice | it is by CONVENTION only that we have seen only integer-compatible values and uuids | 20:10 |
| cp16net | so i recall reading something about setting any flavor id you want? | 20:12 |
| juice | yes you can specify any value for a flavor id | 20:12 |
| cp16net | hmm is that considered the expected behavior of nova flavors? | 20:13 |
| juice | I was suggesting that since flavors are managed by nova AND that nova place zero restriction on the flavors that we (trove) as consumers should not enforce validations on it | 20:13 |
| yogeshmehra | juice: +1 | 20:13 |
| cp16net | juice: right i agree | 20:13 |
| cp16net | i am not wondering if that is expected from nova though | 20:14 |
| cp16net | i've just not heard of that behavior before and curious about the motivation | 20:14 |
| cp16net | for having id='goofy werid flavor name that is really an id' | 20:15 |
| juice | cp16net: can't say for certain whether this is by design in nova. | 20:18 |
| cp16net | juice: how did you create your own id for a flavor? | 20:18 |
| juice | nova flavor-create <name> <id> <ram> <disk> <?> | 20:19 |
| juice | id is required - the only special feature is that if you specify the string "auto" it will generate a UUID for that value | 20:20 |
| juice | other than that, it's all fair game | 20:20 |
| cp16net | hmmm weird | 20:20 |
| *** dmakogon_ has quit IRC | 20:20 | |
| cp16net | ok | 20:21 |
| cp16net | so i think that makes sense to me | 20:22 |
| juice | if they ever put validation on the flavor then we can do likewise or just pass through to their validation. | 20:23 |
| juice | also, I was suggesting that we don't advertise "crazy ass flavor id" value in the documentation and keep all examples with ints or uuid | 20:24 |
| ashestakov_ | i think we should make validation easy, just type: string, len:... | 20:24 |
| hub_cap | cp16net: fwiw, horizon ui is, by default, creating uuid backed flavors | 20:26 |
| cp16net | oh that makes sense why its showing up there | 20:27 |
| cp16net | i found this... https://github.com/openstack/trove/commit/b93ea02cf9de6a9b206d4116f6ba7c243001e775 | 20:28 |
| cp16net | i was trying to cross reference the reason we did this | 20:29 |
| hub_cap | right cp16net and i think i had to revert something w/ it | 20:29 |
| cp16net | https://github.com/openstack/trove/commit/c8c788e92dfd0d781d5bcfb506b92462b3ef9ae0 | 20:30 |
| hub_cap | hrm maybe it wasnt this | 20:30 |
| cp16net | yeah i dont want to cause a contract change | 20:31 |
| *** jodom has joined #openstack-trove | 20:31 | |
| cp16net | thats all i can see | 20:32 |
| cp16net | it seems like we should support ints as well as the strings | 20:33 |
| hub_cap | ya well its not a contract change | 20:33 |
| cp16net | just the comm between nova and trove should use strings | 20:33 |
| cp16net | for flavors | 20:33 |
| hub_cap | if it was "dissalowing" it, id see it as a contract change | 20:33 |
| cp16net | https://review.openstack.org/#/c/48735/1/trove/common/apischema.py | 20:33 |
| hub_cap | but it is, technically removing validation for a int | 20:33 |
| cp16net | that is removing the integrations | 20:33 |
| cp16net | errr... | 20:33 |
| cp16net | integers | 20:33 |
| *** adrian_otto has joined #openstack-trove | 20:34 | |
| hub_cap | juice: can we do int|uuid in jsonschema? | 20:34 |
| juice | we can do that but keep in mind this isn't a json request its just a url GET | 20:34 |
| cp16net | maybe even just cast the input to str to be sure it is throughout the rest of the system | 20:34 |
| cp16net | i see in the tests all the api inputs changed to strings | 20:35 |
| hub_cap | juice: im talking about passing it in to the create | 20:35 |
| juice | instance create? | 20:35 |
| hub_cap | im asuming json schema is validating its an int ya? | 20:35 |
| hub_cap | sure flavor_id=some-uuid | 20:35 |
| *** jrodom has quit IRC | 20:36 | |
| juice | we can def open up the validation in instance create | 20:36 |
| hub_cap | so to me, thats the only validation we should have | 20:36 |
| hub_cap | the rest is just, as juice said, url's | 20:36 |
| cp16net | yeah | 20:36 |
| cp16net | that makes sense | 20:36 |
| juice | hub_cap: ashestakov_ and I were mostly looking at the flavor controller which handles the flavor list/show commands | 20:36 |
| juice | we have validations in there that cast the value to int | 20:37 |
| juice | we also have a conversion in the flavor/view which takes all ids and converts them to ints | 20:37 |
| ashestakov_ | juice: i fixed it in my change | 20:37 |
| ashestakov_ | https://review.openstack.org/#/c/48735/ | 20:38 |
| *** aj_ has quit IRC | 20:38 | |
| ashestakov_ | but i forgot about sql schema | 20:38 |
| juice | ashestakov_: i understand but I was just giving hub_cap the complete footprint of flavor ids in our API | 20:38 |
| juice | ashestakov_: where do we keep flavor_id in our db (not nova's db)? | 20:38 |
| ashestakov_ | juice: in instance table | 20:39 |
| ashestakov_ | its int(11) | 20:39 |
| juice | is that ID for the nova.instance_type.id (which is also int) or is that supposed to be nova.instance_type.flavorid | 20:41 |
| juice | the latter being varchar | 20:41 |
| juice | looks like it is supposed to be mapped to instance_types.flavorid (varchar) | 20:42 |
| juice | which means that field would need to change as well | 20:42 |
| cp16net | yeah looks like it | 20:43 |
| hub_cap | ashestakov_ also there is another change for service_types/flavors, im sure its an int too | 20:45 |
| yogeshmehra | servicetype/flavor mapping takes it as int right now | 20:45 |
| ashestakov_ | hub_cap: where? | 20:45 |
| hub_cap | its not merged | 20:46 |
| hub_cap | we will jsut have to notify the developer | 20:46 |
| yogeshmehra | no, it is abanodoned anyways coz we redefined the "Registration" based strategy | 20:46 |
| yogeshmehra | i'll let sushilKM know | 20:46 |
| hub_cap | yogeshmehra: cool | 20:46 |
| hub_cap | yogeshmehra: i also got a ping from jaishanker, who is offline now | 20:46 |
| hub_cap | to talk about https://review.openstack.org/#/c/41024/ | 20:47 |
| hub_cap | can you tell him that we should maybe clean up the versions code instead of testing stuff that we may not be using? | 20:47 |
| yogeshmehra | yeah, he wanted some details around it...lemme check | 20:47 |
| hub_cap | https://github.com/openstack/cinder/blob/master/cinder/version.py | 20:47 |
| yogeshmehra | true, | 20:47 |
| hub_cap | its considerably different | 20:47 |
| yogeshmehra | ok... | 20:48 |
| yogeshmehra | i'll see to it | 20:48 |
| hub_cap | <3 | 20:48 |
| yogeshmehra | hub_cap: slight digression... | 20:48 |
| yogeshmehra | the heat based instance creation... | 20:49 |
| yogeshmehra | my instance never comes ACTIVE... :-) | 20:50 |
| yogeshmehra | stuck in BUILD | 20:50 |
| datsun180b | yogeshmehra: shot in the dark, wonder if your instance can't get a line out | 20:50 |
| datsun180b | sudo iptables -t nat -A POSTROUTING -s 10.0.0.0/24 -o eth0 -j MASQUERADE look familiar? | 20:51 |
| ashestakov_ | just for clarity, ill replace field type in db and remove validation from flafors except string(255) ? | 20:51 |
| amcrn | yogeshmehra: what does first-boot.log show on the guest? I had the same issue, and it was because the guest image was old and didn't have the heat-client dependency | 20:51 |
| yogeshmehra | amcrn: this is a complete fresh build | 20:52 |
| amcrn | ah, disregard then :) | 20:52 |
| yogeshmehra | sudo iptables..yeah all done | 20:52 |
| datsun180b | that's from the host and not the instance, right | 20:53 |
| ashestakov_ | hub_cap: juice cp16net ? | 20:54 |
| datsun180b | the telltale sign in the guestagent log would be that apt-get update failed and timed out | 20:54 |
| juice | yes ashestakov_? | 20:54 |
| juice | oh I see | 20:54 |
| juice | yes but it also sounds like hub_cap wants to put UUID validation on the instance create for flavor id | 20:55 |
| juice | currently we have int and ref | 20:55 |
| ashestakov_ | its done now in current change | 20:55 |
| juice | but I don't know if it is worth having any validation on flavor list/show since it is pass through and read-only | 20:55 |
| juice | but if you create a flavor in nova that does not conform to uuid or int, then you cannot use it to create an instance in trove | 20:56 |
| ashestakov_ | yep | 20:56 |
| juice | kinda quirky but this is probably a corner case | 20:56 |
| ashestakov_ | i think if nova going this way, we should support it | 20:57 |
| juice | oh and ashestakov_ you are also removing the conversions in the views for flavor yes? | 20:57 |
| ashestakov_ | yes | 20:57 |
| ashestakov_ | hub_cap? | 21:01 |
| juice | ashestakov_ | 21:02 |
| juice | looks like hub_cap is busy | 21:03 |
| juice | go ahead and make the changes | 21:03 |
| cp16net | juice: ashestakov_ i am on board with the change as long as it doesnt stop the instance create/resize from working with ints if passed in from the api | 21:03 |
| juice | worst case is we have to remove the validations around flavor_id in instance create. adding one validation for UUID to the existing set is straightforward | 21:03 |
| cp16net | i notcied that you removed the integer schema and that raised a flag to me | 21:04 |
| juice | nope it won't do that cp16net so we are good there | 21:04 |
| cp16net | juice: ok | 21:04 |
| cp16net | then i guess i was all concerned over nothing... | 21:05 |
| cp16net | :-P | 21:05 |
| cp16net | not the first time... | 21:05 |
| ashestakov_ | hm, how it will upgrade table field? | 21:05 |
| juice | it's friday - your nerves are probably frazzled | 21:05 |
| cp16net | hahah | 21:05 |
| juice | you need to create a db migration for that | 21:05 |
| ashestakov_ | yep, but should it convert existing rows to varchar? | 21:06 |
| juice | it will need to since this can be applied to an existing db | 21:08 |
| juice | can you do an alter table and change the datatype for a column? | 21:09 |
| ashestakov_ | i think it can, but is it possible to convert int data to varchar? | 21:10 |
| *** pdmars has quit IRC | 21:10 | |
| *** rnirmal has quit IRC | 21:11 | |
| *** robertmyers has quit IRC | 21:11 | |
| juice | sure just not the other way around … in all cases | 21:11 |
| ashestakov_ | ok, will research and do tomorrow or monday, i falling sleep now | 21:13 |
| ashestakov_ | have good day and weekend :) | 21:13 |
| *** ashestakov_ has quit IRC | 21:14 | |
| *** vipul is now known as vipul-away | 21:15 | |
| *** jcru has quit IRC | 21:17 | |
| *** vipul-away is now known as vipul | 21:28 | |
| *** yogeshmehra has quit IRC | 21:53 | |
| *** yogesh has joined #openstack-trove | 21:54 | |
| *** dmakogon has quit IRC | 21:56 | |
| *** dmakogon has joined #openstack-trove | 21:56 | |
| *** yogesh has quit IRC | 21:58 | |
| *** yogesh has joined #openstack-trove | 21:58 | |
| *** yogesh has quit IRC | 22:03 | |
| hub_cap | whew back | 22:05 |
| datsun180b | hub_cap: just in time for all of the austin office to be gone | 22:11 |
| hub_cap | LOL | 22:11 |
| hub_cap | took niece to the airport | 22:11 |
| datsun180b | think i'll take off now too, got conductor just about where i want it now | 22:12 |
| hub_cap | noice | 22:12 |
| hub_cap | juice: yall finished talking about the stuff above? | 22:13 |
| datsun180b | got the rpc stuff nailed down, but 1. it appears the hb are only happening every 3 minutes and 2. can't seem to connect to the db, have to track that down | 22:13 |
| datsun180b | at least next week i'll have an update worth looking at so i can get some help | 22:14 |
| *** adrian_otto has quit IRC | 22:14 | |
| *** paul_lodronio has left #openstack-trove | 22:15 | |
| hub_cap | sweet | 22:18 |
| hub_cap | great news datsun180b | 22:18 |
| datsun180b | i think i'm just missing the db creds, i've got the connection details in the conf so i'll have to figure out what other parts of the rain dance i'm missing | 22:18 |
| esp | datsun180b: nice! | 22:19 |
| datsun180b | and i need to figure out why conductor's only acting every 3 minutes | 22:19 |
| datsun180b | but that's work for some sucker with my name come monday morning! have a good weekend | 22:22 |
| *** datsun180b has quit IRC | 22:22 | |
| *** jrodom has joined #openstack-trove | 22:31 | |
| juice | hub_cap: yes we are | 22:49 |
| juice | wait there was one question outstanding | 22:50 |
| juice | you want to still keep the restriction on flavor_id to uuid or int or full ref while used in the context of instance create, correct? | 22:50 |
| hub_cap | well... | 22:53 |
| hub_cap | i mean nova will barf if we dont pass it something | 22:53 |
| hub_cap | that it knows about | 22:53 |
| hub_cap | so maybe we dont put a restriction at all | 22:53 |
| *** jrodom has quit IRC | 22:55 | |
| *** amcrn has quit IRC | 23:38 | |
| *** KennethWilke has quit IRC | 23:53 | |
| *** yogeshmehra has joined #openstack-trove | 23:55 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!