| *** spzala has joined #heat | 00:01 | |
| *** jroyall has joined #heat | 00:01 | |
| *** jroyall is now known as joeroyall | 00:01 | |
| *** zns has quit IRC | 00:03 | |
| *** ccrouch has joined #heat | 00:05 | |
| *** Qiming has joined #heat | 00:06 | |
| *** sarob_ has quit IRC | 00:07 | |
| *** hdd has quit IRC | 00:17 | |
| *** spzala has quit IRC | 00:20 | |
| *** sabeen has quit IRC | 00:26 | |
| *** radez is now known as radez_g0n3 | 00:28 | |
| *** kopparam has joined #heat | 00:30 | |
| *** kopparam has quit IRC | 00:38 | |
| *** randallburt has quit IRC | 00:38 | |
| *** randallburt has joined #heat | 00:39 | |
| *** zns has joined #heat | 00:42 | |
| *** gokrokve has joined #heat | 00:44 | |
| *** andersonvom has quit IRC | 00:45 | |
| *** kebray has quit IRC | 00:46 | |
| *** hdd has joined #heat | 00:59 | |
| openstackgerrit | Angus Salkeld proposed a change to openstack/heat: Import the stack module directly https://review.openstack.org/107292 | 01:00 |
|---|---|---|
| *** zns has quit IRC | 01:05 | |
| *** erkules_ has joined #heat | 01:12 | |
| SpamapS | stevebaker: around? | 01:12 |
| stevebaker | SpamapS: hai! | 01:12 |
| SpamapS | stevebaker: https://bugs.launchpad.net/heat/+bug/1366798 .. I'm losing my Heat context right now while distracted with some internal stuff. Can you triage that? | 01:13 |
| uvirtbot | Launchpad bug 1366798 in heat "JSON parameter is returned wrongly by the JsonParameter class" [Undecided,New] | 01:13 |
| SpamapS | stevebaker: I am suspicious that it is just a heatclient bug in the 'show' command, and not an API problem, but I figure you can verify it faster than I can dig out my test rig and poke it. | 01:14 |
| stevebaker | SpamapS: sure thing, I can do that | 01:14 |
| openstackgerrit | Angus Salkeld proposed a change to openstack/heat: Support loading resources from stevedore extensions https://review.openstack.org/103044 | 01:14 |
| openstackgerrit | Angus Salkeld proposed a change to openstack/heat: Use setuptools to install contrib plugins https://review.openstack.org/105334 | 01:14 |
| openstackgerrit | Angus Salkeld proposed a change to openstack/heat: Convert functions into a fixed part of the template https://review.openstack.org/103484 | 01:14 |
| *** erkules has quit IRC | 01:14 | |
| SpamapS | stevebaker: thanks! | 01:15 |
| miguelgrinberg | stevebaker: sorry to bug you. I noticed that 404 errors from the heat API are sometimes returned as text/plain, others as text/html. Are there any plans to fix this? Do you think it is a worthwhile effort? | 01:20 |
| miguelgrinberg | stevebaker: I'm thinking these should be JSON, in case it isn't clear, or at least use content negotiation to return what the client wants | 01:21 |
| stevebaker | miguelgrinberg: I'm not aware of a bug for that, it seems worth fixing to be JSON | 01:22 |
| asalkeld | anyone booking flights to paris? | 01:22 |
| miguelgrinberg | stevebaker: I'll write a bug then. Thanks. | 01:22 |
| asalkeld | wondering if we need time before/after for moar heat convergence chatter | 01:23 |
| *** gokrokve_ has joined #heat | 01:23 | |
| stevebaker | asalkeld: a good proportion of our summit sessions will be convergence | 01:23 |
| asalkeld | yeah, i guess | 01:23 |
| stevebaker | asalkeld: we'll likely drop the 40 minute session format and have something less structured | 01:24 |
| *** hdd has quit IRC | 01:24 | |
| asalkeld | stevebaker, I thought it was going to be a mix of structured and unstructured | 01:25 |
| stevebaker | asalkeld: probably, thats up to the PTL to figure out | 01:25 |
| asalkeld | really? | 01:25 |
| asalkeld | not sure you can just change the format | 01:25 |
| asalkeld | i think structured has it's place | 01:26 |
| *** kebray has joined #heat | 01:26 | |
| *** gokrokve has quit IRC | 01:26 | |
| asalkeld | for people that want to attend other projects | 01:26 |
| asalkeld | (not everyone is solely focused on one project) | 01:26 |
| *** gokrokve_ has quit IRC | 01:27 | |
| *** joeroyall has quit IRC | 01:27 | |
| stevebaker | asalkeld: have you followed all the openstack-dev discussion about the desire to get the summit sessions to be as productive as the mid-cycle meetups? | 01:31 |
| asalkeld | sure, and that seems to be ending in part/part structured/unstructed last i looked | 01:31 |
| stevebaker | yep, and the new PTL can figure out with the rest of us how the heat sessions will be scheduled. I'm pretty sure we're saying the same thing | 01:34 |
| *** Drago1 has joined #heat | 01:34 | |
| asalkeld | cool, such a long thread | 01:34 |
| asalkeld | hard to see the final outcome | 01:35 |
| *** Drago1 has quit IRC | 01:37 | |
| *** hdd has joined #heat | 01:44 | |
| *** nosnos has joined #heat | 01:51 | |
| lipinski1 | How do I create a Neutron network/subnet via Heat but have the gateway disabled? | 01:55 |
| lipinski1 | My problem is that I'm getting a undesired default route from the network I create - when I don't want that behavior. | 01:55 |
| lipinski1 | I'm not specifying any gateway_ip in the OS::Neutron::Subnet parameters | 01:55 |
| *** Tross has quit IRC | 01:58 | |
| *** denis_makogon has quit IRC | 02:10 | |
| *** denis_makogon has joined #heat | 02:10 | |
| *** openstackgerrit has quit IRC | 02:33 | |
| *** hdd has quit IRC | 02:40 | |
| *** sabeen has joined #heat | 02:45 | |
| *** ramishra has joined #heat | 02:55 | |
| *** KanagarajM has joined #heat | 02:57 | |
| *** sabeen has quit IRC | 02:57 | |
| *** sabeen has joined #heat | 02:58 | |
| *** Drago1 has joined #heat | 03:00 | |
| *** ramishra has quit IRC | 03:08 | |
| *** Drago1 has quit IRC | 03:08 | |
| *** ramishra has joined #heat | 03:09 | |
| *** lazy_prince is now known as killer_prince | 03:16 | |
| *** ajc_ has joined #heat | 03:22 | |
| *** achampion has joined #heat | 03:22 | |
| *** openstack has joined #heat | 03:42 | |
| *** sendak.freenode.net sets mode: -o gonery | 03:46 | |
| -sendak.freenode.net- *** Notice -- TS for #heat changed from 1410234098 to 1332364781 | 03:46 | |
| *** sendak.freenode.net sets mode: +cpt-s | 03:46 | |
| *** hdd has joined #heat | 03:46 | |
| *** coolsvap has joined #heat | 03:46 | |
| *** tango has joined #heat | 03:46 | |
| *** xmltok has joined #heat | 03:46 | |
| *** andreaf_ has joined #heat | 03:46 | |
| *** elynn has joined #heat | 03:46 | |
| *** stevelle has joined #heat | 03:46 | |
| *** metral has joined #heat | 03:46 | |
| *** denis_makogon has joined #heat | 03:46 | |
| *** erkules_ has joined #heat | 03:46 | |
| *** stevebak` has joined #heat | 03:46 | |
| *** john-n-s- has joined #heat | 03:46 | |
| *** Guest56441 has joined #heat | 03:46 | |
| *** thegodfather has joined #heat | 03:46 | |
| *** vijendar1 has joined #heat | 03:46 | |
| *** achampio1 has joined #heat | 03:46 | |
| *** Murali__ has joined #heat | 03:46 | |
| *** ccrouch1 has joined #heat | 03:46 | |
| *** KanagarajM has joined #heat | 03:46 | |
| *** ajc_ has joined #heat | 03:46 | |
| *** ramishra has joined #heat | 03:46 | |
| *** sabeen has joined #heat | 03:46 | |
| *** nosnos has joined #heat | 03:46 | |
| *** kebray has joined #heat | 03:46 | |
| *** randallburt has joined #heat | 03:46 | |
| *** Qiming has joined #heat | 03:46 | |
| *** edmund has joined #heat | 03:46 | |
| *** asalkeld has joined #heat | 03:46 | |
| *** daneyon has joined #heat | 03:46 | |
| *** killer_prince has joined #heat | 03:46 | |
| *** gpocentek has joined #heat | 03:46 | |
| *** sbadia has joined #heat | 03:46 | |
| *** grncdr has joined #heat | 03:46 | |
| *** lipinski1 has joined #heat | 03:46 | |
| *** BillArnold has joined #heat | 03:46 | |
| *** htruta has joined #heat | 03:46 | |
| *** reed has joined #heat | 03:46 | |
| *** aweiteka has joined #heat | 03:46 | |
| *** dsneddon has joined #heat | 03:46 | |
| *** sjmc7 has joined #heat | 03:46 | |
| *** sergmelikyan has joined #heat | 03:46 | |
| *** pmyers has joined #heat | 03:46 | |
| *** Daviey has joined #heat | 03:46 | |
| *** rdo has joined #heat | 03:46 | |
| *** blomquisg has joined #heat | 03:46 | |
| *** enterprisedc has joined #heat | 03:46 | |
| *** ekarlso- has joined #heat | 03:46 | |
| *** urulama-afk has joined #heat | 03:46 | |
| *** cmyster has joined #heat | 03:46 | |
| *** jrist has joined #heat | 03:46 | |
| *** praveenkumar has joined #heat | 03:46 | |
| *** boris-42 has joined #heat | 03:46 | |
| *** rbrady has joined #heat | 03:46 | |
| *** justin-8 has joined #heat | 03:46 | |
| *** pscheie has joined #heat | 03:46 | |
| *** liusheng has joined #heat | 03:46 | |
| *** cody-somerville has joined #heat | 03:46 | |
| *** greghaynes has joined #heat | 03:46 | |
| *** IgorYozhikov has joined #heat | 03:46 | |
| *** pbelanyi has joined #heat | 03:46 | |
| *** DandyPandy has joined #heat | 03:46 | |
| *** shadower has joined #heat | 03:46 | |
| *** shardy has joined #heat | 03:46 | |
| *** lsmola_ has joined #heat | 03:46 | |
| *** swygue has joined #heat | 03:46 | |
| *** wpf has joined #heat | 03:46 | |
| *** larsks has joined #heat | 03:46 | |
| *** jpeeler has joined #heat | 03:46 | |
| *** bnemec has joined #heat | 03:46 | |
| *** Kimamisa has joined #heat | 03:46 | |
| *** morganfainberg has joined #heat | 03:46 | |
| *** SpamapS has joined #heat | 03:46 | |
| *** radix has joined #heat | 03:46 | |
| *** skraynev has joined #heat | 03:46 | |
| *** therve has joined #heat | 03:46 | |
| *** kevinbenton has joined #heat | 03:46 | |
| *** sdake has joined #heat | 03:46 | |
| *** dekozo has joined #heat | 03:46 | |
| *** harlowja has joined #heat | 03:46 | |
| *** jasond`` has joined #heat | 03:46 | |
| *** jdandrea has joined #heat | 03:46 | |
| *** Ng has joined #heat | 03:46 | |
| *** adam_g has joined #heat | 03:46 | |
| *** wendar has joined #heat | 03:46 | |
| *** cyli has joined #heat | 03:46 | |
| *** slagle has joined #heat | 03:46 | |
| *** zz_gondoi has joined #heat | 03:46 | |
| *** Chinorro has joined #heat | 03:46 | |
| *** otoolee has joined #heat | 03:46 | |
| *** sileht has joined #heat | 03:46 | |
| *** sirushti has joined #heat | 03:46 | |
| *** nijaba has joined #heat | 03:46 | |
| *** dteselkin has joined #heat | 03:46 | |
| *** FL1SK has joined #heat | 03:46 | |
| *** sgran has joined #heat | 03:46 | |
| *** uberj has joined #heat | 03:46 | |
| *** zaneb has joined #heat | 03:46 | |
| *** tank100 has joined #heat | 03:46 | |
| *** higgins has joined #heat | 03:46 | |
| *** brint has joined #heat | 03:46 | |
| *** EmilienM has joined #heat | 03:46 | |
| *** abramley has joined #heat | 03:46 | |
| *** pleia2 has joined #heat | 03:46 | |
| *** zhiyan has joined #heat | 03:46 | |
| *** Michalik_ has joined #heat | 03:46 | |
| *** wirehead_ has joined #heat | 03:46 | |
| *** russellb has joined #heat | 03:46 | |
| *** sgordon has joined #heat | 03:46 | |
| *** radez_g0n3 has joined #heat | 03:46 | |
| *** lvh has joined #heat | 03:46 | |
| *** mattoliverau has joined #heat | 03:46 | |
| *** serverascode has joined #heat | 03:46 | |
| *** chmouel has joined #heat | 03:46 | |
| *** funzo has joined #heat | 03:46 | |
| *** SergeyLukjanov has joined #heat | 03:46 | |
| *** zz_naotok has joined #heat | 03:46 | |
| *** Guest12333 has joined #heat | 03:46 | |
| *** kragniz has joined #heat | 03:46 | |
| *** PhilK has joined #heat | 03:46 | |
| *** mikal has joined #heat | 03:46 | |
| *** TVR___ has joined #heat | 03:46 | |
| *** uvirtbot has joined #heat | 03:46 | |
| *** DinaBelova has joined #heat | 03:46 | |
| *** andrearosa has joined #heat | 03:46 | |
| *** ruhe has joined #heat | 03:46 | |
| *** mtreinish has joined #heat | 03:46 | |
| *** miguelgrinberg has joined #heat | 03:46 | |
| *** aignatov has joined #heat | 03:46 | |
| *** lifeless has joined #heat | 03:46 | |
| *** anteaya has joined #heat | 03:46 | |
| *** tsufiev has joined #heat | 03:46 | |
| *** zigo has joined #heat | 03:46 | |
| *** gilliard_ has joined #heat | 03:46 | |
| *** ryansb has joined #heat | 03:46 | |
| *** jgrassler has joined #heat | 03:46 | |
| *** cinerama has joined #heat | 03:46 | |
| *** Isotopp has joined #heat | 03:46 | |
| *** mburns has joined #heat | 03:46 | |
| *** shufflebot has joined #heat | 03:46 | |
| *** devx has joined #heat | 03:46 | |
| *** d0ugal has joined #heat | 03:46 | |
| *** ChanServ has joined #heat | 03:46 | |
| *** sendak.freenode.net sets mode: +o ChanServ | 03:46 | |
| *** sendak.freenode.net changes topic to "support @ https://ask.openstack.org | developer wiki @ https://wiki.openstack.org/wiki/Heat | development @ https://launchpad.net/heat | logged @ http://eavesdrop.openstack.org/irclogs/%23heat/" | 03:46 | |
| *** Tross has joined #heat | 03:46 | |
| *** randallburt1 has joined #heat | 03:49 | |
| *** killer_prince has quit IRC | 03:50 | |
| *** lazy_prince has joined #heat | 03:50 | |
| *** lazy_prince has quit IRC | 03:50 | |
| *** lazy_prince has joined #heat | 03:51 | |
| *** lazy_prince has quit IRC | 03:53 | |
| *** randallburt has quit IRC | 03:53 | |
| *** killer_prince has joined #heat | 03:53 | |
| *** killer_prince has quit IRC | 03:53 | |
| *** randallburt1 has quit IRC | 03:54 | |
| *** ramishra has quit IRC | 03:54 | |
| *** lazy_prince has joined #heat | 03:54 | |
| *** ramishra has joined #heat | 03:54 | |
| *** Tross has quit IRC | 03:56 | |
| *** jroyall has joined #heat | 03:59 | |
| *** jroyall is now known as joeroyall | 03:59 | |
| *** lazy_prince is now known as killer_prince | 04:00 | |
| *** mburns has quit IRC | 04:04 | |
| *** joeroyall has quit IRC | 04:04 | |
| *** Qiming has quit IRC | 04:06 | |
| *** asalkeld has quit IRC | 04:06 | |
| *** mburns has joined #heat | 04:11 | |
| *** vijendar1 has quit IRC | 04:12 | |
| *** nkhare has joined #heat | 04:13 | |
| *** andersonvom has joined #heat | 04:14 | |
| *** Murali__ has quit IRC | 04:14 | |
| *** akuznetsov has joined #heat | 04:14 | |
| *** sanjayu has joined #heat | 04:15 | |
| *** Murali has joined #heat | 04:15 | |
| *** edmund has quit IRC | 04:18 | |
| *** asalkeld has joined #heat | 04:19 | |
| *** thegodfather is now known as fabbione | 04:20 | |
| *** achampio1 has quit IRC | 04:20 | |
| *** Murali has quit IRC | 04:23 | |
| *** Murali__ has joined #heat | 04:24 | |
| *** Qiming has joined #heat | 04:24 | |
| *** stevebak` is now known as stevebaker | 04:32 | |
| *** achampion has joined #heat | 04:35 | |
| *** john-n-s- has quit IRC | 04:41 | |
| *** john-n-s- has joined #heat | 04:42 | |
| stevebaker | jasond``: hey, could you put together a SwiftSignal example for heat-templates? | 04:45 |
| *** gokrokve has joined #heat | 04:46 | |
| *** john-n-s- has quit IRC | 04:50 | |
| *** gokrokve has quit IRC | 04:51 | |
| *** FL1SK has quit IRC | 04:51 | |
| *** reed has quit IRC | 04:52 | |
| *** FL1SK has joined #heat | 04:56 | |
| *** john-n-s- has joined #heat | 04:56 | |
| *** john-n-s- has quit IRC | 05:01 | |
| *** john-n-s- has joined #heat | 05:01 | |
| *** killer_prince is now known as lazy_prince | 05:01 | |
| *** FL1SK has quit IRC | 05:01 | |
| *** kopparam has joined #heat | 05:02 | |
| *** FL1SK has joined #heat | 05:04 | |
| *** urulama-afk is now known as urulama | 05:09 | |
| *** Tross has joined #heat | 05:16 | |
| *** sjmc7 has quit IRC | 05:22 | |
| *** sjmc7 has joined #heat | 05:22 | |
| *** sabeen has quit IRC | 05:26 | |
| *** hdd has quit IRC | 05:36 | |
| *** cmyster_ has joined #heat | 05:37 | |
| cmyster_ | morning | 05:39 |
| *** lazy_prince is now known as killer_prince | 05:40 | |
| *** gokrokve has joined #heat | 05:46 | |
| *** harlowja is now known as harlowja_away | 05:47 | |
| *** gokrokve has quit IRC | 05:50 | |
| *** Murali__ has quit IRC | 05:51 | |
| *** Murali_ has joined #heat | 05:53 | |
| *** killer_prince is now known as lazy_prince | 05:56 | |
| *** Murali_ has quit IRC | 05:58 | |
| *** achampion has quit IRC | 05:59 | |
| *** lazy_prince is now known as killer_prince | 05:59 | |
| *** killer_prince has quit IRC | 06:00 | |
| *** lazy_prince has joined #heat | 06:00 | |
| *** achampion has joined #heat | 06:02 | |
| *** jcoufal has joined #heat | 06:12 | |
| *** bmahalakshmi has joined #heat | 06:12 | |
| *** bmahalakshmi has quit IRC | 06:13 | |
| *** cmyster_ has quit IRC | 06:13 | |
| *** cmyster_ has joined #heat | 06:14 | |
| *** bmahalakshmi has joined #heat | 06:14 | |
| *** cmyster_ has quit IRC | 06:19 | |
| *** cmyster_ has joined #heat | 06:21 | |
| *** ccrouch1 has quit IRC | 06:22 | |
| *** andersonvom has quit IRC | 06:22 | |
| *** kopparam has quit IRC | 06:23 | |
| *** kopparam has joined #heat | 06:27 | |
| *** Guest56441 is now known as Adri2000 | 06:27 | |
| *** Adri2000 has quit IRC | 06:27 | |
| *** Adri2000 has joined #heat | 06:27 | |
| *** lipinski1 has quit IRC | 06:29 | |
| *** MohitS has joined #heat | 06:30 | |
| *** ccrouch has joined #heat | 06:34 | |
| *** rakesh_hs has joined #heat | 06:35 | |
| *** jroyall has joined #heat | 06:35 | |
| *** jroyall is now known as joeroyall | 06:35 | |
| *** jprovazn has joined #heat | 06:38 | |
| *** ccrouch has quit IRC | 06:39 | |
| *** kebray has quit IRC | 06:44 | |
| *** gokrokve has joined #heat | 06:46 | |
| *** gokrokve has quit IRC | 06:51 | |
| *** tango has quit IRC | 06:58 | |
| *** ccrouch has joined #heat | 06:59 | |
| *** cmyster_ has quit IRC | 06:59 | |
| *** joeroyall has quit IRC | 07:01 | |
| *** kopparam has quit IRC | 07:02 | |
| *** ifarkas has joined #heat | 07:08 | |
| *** ajc__ has joined #heat | 07:09 | |
| *** ajc_ has quit IRC | 07:10 | |
| *** inc0 has joined #heat | 07:11 | |
| inc0 | good morning | 07:11 |
| *** bmahalakshmi has quit IRC | 07:12 | |
| *** ajc__ has quit IRC | 07:14 | |
| *** pas-ha has joined #heat | 07:14 | |
| pas-ha | morning all | 07:19 |
| *** alexheneveld has joined #heat | 07:21 | |
| *** jstrachan has joined #heat | 07:31 | |
| asalkeld | hi pas-ha | 07:33 |
| *** jistr has joined #heat | 07:33 | |
| *** ishant has joined #heat | 07:36 | |
| *** noTHD has joined #heat | 07:37 | |
| *** jcoufal has quit IRC | 07:39 | |
| *** julienvey has joined #heat | 07:39 | |
| *** kopparam has joined #heat | 07:45 | |
| *** gokrokve has joined #heat | 07:46 | |
| *** pas-ha has quit IRC | 07:47 | |
| *** asalkeld has quit IRC | 07:49 | |
| *** gokrokve has quit IRC | 07:51 | |
| *** pasquier-s has joined #heat | 07:53 | |
| *** derekh has joined #heat | 07:53 | |
| *** sorantis has joined #heat | 07:55 | |
| *** bvandenh has joined #heat | 07:56 | |
| *** pas-ha has joined #heat | 08:00 | |
| *** unmeshg has joined #heat | 08:01 | |
| *** jcoufal has joined #heat | 08:03 | |
| *** ajc_ has joined #heat | 08:03 | |
| *** asalkeld has joined #heat | 08:04 | |
| *** ajc_ has quit IRC | 08:05 | |
| *** ajc_ has joined #heat | 08:07 | |
| *** ajc_ has quit IRC | 08:07 | |
| shardy | morning all | 08:08 |
| *** jstrachan has quit IRC | 08:10 | |
| *** tspatzier has joined #heat | 08:12 | |
| *** jstrachan has joined #heat | 08:13 | |
| inc0 | I'm gathering input for my bp :) https://blueprints.launchpad.net/heat/+spec/haaas-heat-resource any words guys? | 08:17 |
| *** erkules_ is now known as erkules | 08:18 | |
| shardy | inc0: I'd like to see an explanation of why you can't just use an AutoScaling group with a specific number of in-service instances (which could be just one) | 08:22 |
| Isotopp | inc0: this being a cluster-type of thingie, I'd like to see more discussion of state and problems, e.g. why would a single vm fail as opposed to all vms on a host? a cluster thingy should probably have an idea of what's going on. also, on vm restart elsewhere, on a different host, it probably wants to know that this is a restart internally, via cloud-init or something, and take action to reestablish network things, clean up volume ... | 08:22 |
| Isotopp | ... state or similar stuff. | 08:22 |
| shardy | inc0: FWIW, having a resource called "HighAvailability" makes no sense to me, as it's not backed by a thing | 08:23 |
| Isotopp | inc0: working with contrail for example we have had cases where hosts were available on the message queue, but not via contrail external network. nova scheduler was trying to deploy instances on these hosts, all of them unuseable because of contrail being weird. cluster membership is complicated, this was essentially a weird variant of a split brain | 08:24 |
| shardy | HA is a deployment strategy, which is expressible via template design/structure, not really a resource plugin IMO | 08:24 |
| *** andreaf_ is now known as andreaf | 08:25 | |
| inc0 | Isotopp, about single-vm and all vm, of course all vms will be down in case of host failure, but we might want to take care of only one | 08:25 |
| shardy | inc0: also the "HAEngine" part is very similar to aspects of the convergence design, so we should probably just ensure convergence meets your needs here, instead of reinventing a new feature with a lot of overlap in use-cases | 08:25 |
| inc0 | thsi is *aas, and only few vms may have to be stateful | 08:25 |
| cmyster | indeed | 08:25 |
| cmyster | mt... | 08:25 |
| shardy | inc0: if you're keen to continue with this, it would be great if you can convert the etherpad into a spec review, so we can all comment there via gerrit | 08:26 |
| inc0 | shardy, making that resource resembles current HARestarter and main idea underneath is to make possible to run HAaaS in heat in non-heat deployments | 08:26 |
| inc0 | shardy, I will continue one way or another:) I just want to ask you (heat) guys if you think its good place to do it | 08:27 |
| shardy | inc0: The current HARestarter resource is a historical mistake, we'd like to deprecate it and remove it completely, not enhance it ;) | 08:27 |
| inc0 | shardy, why not make it well... usable?:) | 08:28 |
| shardy | inc0: Because you already satisfy the same use cases via other resources which actually work :) | 08:28 |
| shardy | and/or via plans for convergence | 08:29 |
| inc0 | Isotopp, about clustering, yeah, we'll need fencing on this thing. I was thinking of IPMI-powered STONITH | 08:29 |
| shardy | inc0: I'd like to understand specifics of things you need to do which can't be satisfied via AutoScalingGroup | 08:29 |
| shardy | and/or planned work for convergence | 08:30 |
| inc0 | shardy, you can satisfy making stateful vms using current tools? | 08:30 |
| inc0 | its pets vs cattle thing | 08:30 |
| shardy | inc0: all HARestarter does is fire up a new instance when one dies and stops sending a heartbeat, you can do the same via ASG | 08:31 |
| shardy | the main difference is we won't delete the old broken instance until it comes back online | 08:31 |
| inc0 | shardy, current HARestarter, I want something which will boot instance in case of host heartbeat problem | 08:31 |
| shardy | And the work for convergence is to monitor state and automatically align the world with the stack if they diverge | 08:32 |
| inc0 | and run it on other host | 08:32 |
| inc0 | I'd use current HARestarter as something like placeholder for this thing, allowing of course backward compatibility, but also allowing new features, namely host-down scenerios | 08:33 |
| shardy | inc0: So an AutoScalingGroup with anti-affinity between instances based on some hint like host aggregates? | 08:33 |
| inc0 | shardy, still if we will lose host with vms on it we will boot new vms, not reboot same ones on different host | 08:34 |
| shardy | inc0: sure, I'm just trying to explore whether this will be (or already is) possible via existing stuff | 08:34 |
| shardy | inc0: booting vms on another host is new vms | 08:34 |
| inc0 | shardy, not nexessairly | 08:35 |
| *** ishant has quit IRC | 08:35 | |
| inc0 | necessairly* | 08:35 |
| shardy | say you have 100 vms spread evenly over 10 hosts, and one host dies, heat sees only 90, so it asks nova to build 10 more | 08:35 |
| shardy | If the host comes back up, the 10 oldest will get killed | 08:35 |
| inc0 | shardy, thats AWS like cattle-vms scenerio, which is best one | 08:35 |
| inc0 | but we might for example have 1 vm without replication | 08:36 |
| inc0 | with db for example | 08:36 |
| cmyster | thats just bad planning | 08:36 |
| inc0 | cmyster, yes, but in reality its a thing | 08:36 |
| inc0 | we want to reach out for bad planners as well | 08:36 |
| inc0 | in reality there are few really stateless apps | 08:37 |
| *** unmeshg has quit IRC | 08:37 | |
| inc0 | even openstack itself has issues in that matter | 08:37 |
| shardy | inc0: so, what you want isn't to build new vms on a different host, but reboot those on a host..which you've just lost? | 08:37 |
| cmyster | what you are talking about here is to make a storage cloud thing where bits are replicated between a few phisical hard drive | 08:37 |
| * shardy is confused | 08:37 | |
| *** KanagarajM2 has joined #heat | 08:38 | |
| inc0 | shardy, vms in reallity are just files on compute nodes | 08:38 |
| *** KanagarajM has quit IRC | 08:38 | |
| inc0 | shardy, so if we place these files on shared storage, for example ceph cluster | 08:38 |
| shardy | inc0: Yeah, but I'm trying to distill if this aims to solve the "host went down" scenario or "app on vm crashed" | 08:38 |
| shardy | both are indirectly solveable via existing resources | 08:39 |
| inc0 | its just for host-down case | 08:39 |
| shardy | So you have to use the cattle model then, as your files representing the vm's are gone | 08:39 |
| shardy | fire up some more on a non-gone host and continue :) | 08:40 |
| cmyster | inc0: storage solution in existence like XIV's for instance split actual data ~3 way to ~3 different phisical hard drives. if one of the hard drives die then you have the other 2. VMs as a whole are a different thing. still manageable but you need to define the storage cloud in a very certain way | 08:40 |
| inc0 | shardy, not really, I don't think there is obvious way to reboot machine placed on shared storage on different host with migration of volume bindings, networks, addresses and so on | 08:40 |
| inc0 | cmyster, yes, but there are solutions for it. Either SAN or DRBD | 08:40 |
| inc0 | shardy, thing is, cattle, while being best solution always, require work from everyone, including app developers | 08:41 |
| inc0 | and that is not always a case | 08:41 |
| inc0 | and we want to reach out to people who has for example legacy apps, stateful and impossible to be turned into cattle | 08:42 |
| inc0 | this is one of reasons why openstack still lose to other solutions in terms of enterprise private cloud | 08:42 |
| shardy | inc0: I still think that even if you don't find the AutoScaling solution to your liking, some combination of boot-from-volume and convergence may satisfy this use case | 08:44 |
| inc0 | cmyster, thing is, you will not achieve HA without specific infrastructure configuration, but this little module along with few good howtos may allow people to start having 99%+ SLA on pet-vms on openstack based clouds | 08:45 |
| inc0 | shardy, this will be part of it, but how will you implement fencing, heartbeat of hosts and so on? | 08:45 |
| inc0 | I mean it is possible, ofcourse, I just want to make it easy:) | 08:45 |
| shardy | feel free to post the spec, but if there's a way to direct your engergies to making convergence work well for this use-case, instead of implementing a really domain-specific solution, that would be good | 08:45 |
| cmyster | +1 | 08:46 |
| cmyster | morning shardy btw ;) | 08:46 |
| shardy | o/ | 08:46 |
| *** gokrokve has joined #heat | 08:46 | |
| inc0 | quite frankly I have to read about convergence;) | 08:46 |
| *** urulama has left #heat | 08:47 | |
| *** gokrokve has quit IRC | 08:47 | |
| *** gokrokve has joined #heat | 08:48 | |
| shardy | inc0: the heartbeat is already possible via sample-count alarms, not sure about fencing atm | 08:48 |
| inc0 | shardy, sample count might be too slow, nova has zookeeper, but I want to make few more tests | 08:49 |
| shardy | seems like if you automated deleting an instance from a gone-host, then nova should take care of ensuring it's deleted immediately when the host is recommissioned | 08:49 |
| inc0 | we'd like to have less-than-a-second feedback | 08:49 |
| inc0 | there are few good patterns for fencing | 08:50 |
| shardy | inc0: right, will be interesting to hear how you plan to do that in a scalable way :) | 08:50 |
| cmyster | shared virtual drives across all physical hosts for instance. | 08:51 |
| shardy | inc0: please do post the spec after you've looked at the convergence ones, will be interesting to continue the discussion there | 08:51 |
| inc0 | thats one of problems, yea;) but with host aggregates and so on I think its doable:) | 08:51 |
| inc0 | cmyster, not necessairly all physical hosts...just few of them | 08:51 |
| cmyster | indeed | 08:51 |
| inc0 | like nova cells | 08:52 |
| inc0 | just make sure that new vm will be booted on host connected to this storage | 08:52 |
| cmyster | I don't see how you can manage it without virtual drive balancing | 08:52 |
| *** gokrokve has quit IRC | 08:52 | |
| inc0 | cmyster, what do you mean by "virtual drive balancing"? | 08:53 |
| cmyster | omw out to launch I think I have some guids on it, I'll find them later... | 08:54 |
| cmyster | 1h or so | 08:54 |
| inc0 | cmyster, thanks:) | 08:55 |
| *** ramishra has quit IRC | 08:56 | |
| *** sorantis has quit IRC | 08:57 | |
| *** fayablazer has joined #heat | 08:57 | |
| *** ramishra has joined #heat | 09:03 | |
| *** nikunj2513 has joined #heat | 09:05 | |
| *** cdent has joined #heat | 09:06 | |
| *** sarob has joined #heat | 09:07 | |
| *** nikunj2513 is now known as nikunj2512 | 09:08 | |
| *** kopparam has quit IRC | 09:10 | |
| *** ramishra has quit IRC | 09:15 | |
| *** ramishra has joined #heat | 09:15 | |
| *** kopparam has joined #heat | 09:16 | |
| *** tomek_adamczewsk has joined #heat | 09:18 | |
| *** jstrachan has quit IRC | 09:20 | |
| *** jstrachan_ has joined #heat | 09:20 | |
| *** openstackgerrit has joined #heat | 09:20 | |
| *** EricGonczer_ has joined #heat | 09:22 | |
| *** jstrachan has joined #heat | 09:24 | |
| *** che-arne has joined #heat | 09:26 | |
| *** jstrachan_ has quit IRC | 09:27 | |
| *** ramishra has quit IRC | 09:33 | |
| *** ramishra has joined #heat | 09:33 | |
| *** KanagarajM2 has quit IRC | 09:39 | |
| *** mkollaro has joined #heat | 09:39 | |
| openstackgerrit | huangtianhua proposed a change to openstack/heat: Implement AWS::EC2::EIPAssociation updatable https://review.openstack.org/118562 | 09:43 |
| openstackgerrit | huangtianhua proposed a change to openstack/heat: Implement AWS::EC2::EIPAssociation updatable https://review.openstack.org/118562 | 09:43 |
| *** KanagarajM has joined #heat | 09:44 | |
| ekarlso- | Any heat devs around ? | 09:45 |
| ekarlso- | shardy: ? | 09:45 |
| shardy | ekarlso-: hi | 09:45 |
| ekarlso- | shardy: I got a fix for a bug with the jsonparameter | 09:45 |
| ekarlso- | where to unit tests go ? :) | 09:45 |
| shardy | ekarlso-: heat/heat/tests | 09:46 |
| ekarlso- | test_parameters ? | 09:46 |
| shardy | hard to say without seeing what code your fix touches ;) | 09:46 |
| *** tspatzier has quit IRC | 09:46 | |
| *** gokrokve has joined #heat | 09:46 | |
| shardy | but probably, see if the code you changed is tested by that test :) | 09:47 |
| shardy | then add a new test which proves the fix works and existing interfaces aren't broken | 09:47 |
| *** gokrokve has quit IRC | 09:51 | |
| openstackgerrit | Michal Jastrzebski (inc0) proposed a change to openstack/heat-specs: HAaaS Specification https://review.openstack.org/120051 | 09:51 |
| *** ramishra has quit IRC | 09:52 | |
| inc0 | shardy, here you go:) I've seen rst for convergence and I think its different matter- convergence will take care of stack being deployed correctly while its not really pet-caretaker | 09:53 |
| *** asalkeld_ has joined #heat | 09:56 | |
| *** asalkeld has quit IRC | 09:56 | |
| *** lazy_prince has quit IRC | 09:56 | |
| *** EricGonczer_ has quit IRC | 10:02 | |
| *** dsneddon has quit IRC | 10:02 | |
| ekarlso- | shardy: http://paste.openstack.org/show/108810/ | 10:07 |
| *** KanagarajM has quit IRC | 10:08 | |
| *** KanagarajM has joined #heat | 10:11 | |
| openstackgerrit | Michal Jastrzebski (inc0) proposed a change to openstack/heat-specs: HAaaS Specification https://review.openstack.org/120051 | 10:13 |
| *** mkerrin has joined #heat | 10:14 | |
| *** ishant has joined #heat | 10:15 | |
| *** julienvey has quit IRC | 10:25 | |
| *** alexpilotti has joined #heat | 10:25 | |
| *** julienvey has joined #heat | 10:25 | |
| *** alexpilotti has quit IRC | 10:30 | |
| Qiming | inc0, pls revise the spec to remove hard returns and render your ascii diagram in a '::' environment | 10:32 |
| *** nkhare has quit IRC | 10:34 | |
| Qiming | inc0, still around? | 10:35 |
| *** asalkeld_ has quit IRC | 10:36 | |
| *** pasquier-s has quit IRC | 10:41 | |
| *** gilliard_ is now known as gilliard | 10:42 | |
| *** asalkeld has joined #heat | 10:43 | |
| *** lazy_prince has joined #heat | 10:44 | |
| *** gokrokve has joined #heat | 10:46 | |
| *** sorantis has joined #heat | 10:49 | |
| inc0 | Qiming, now yes | 10:50 |
| Qiming | hi, inc0, I am reading your specs | 10:50 |
| inc0 | so replace spaces with : | 10:50 |
| inc0 | thank you:) | 10:51 |
| inc0 | ah, .rst code block | 10:51 |
| Qiming | I have been looking into this for quite some time now | 10:51 |
| *** gokrokve has quit IRC | 10:51 | |
| Qiming | I'm happy that someone else are starting to pay attention to VM/App HA now. | 10:52 |
| Qiming | however, I have some different thoughts on your proposal | 10:52 |
| Qiming | especially the failure detection, notification part | 10:52 |
| inc0 | yes, thats somewhat placeholder there, I'm about to start testing of different existing software for failure detection | 10:53 |
| inc0 | my general idea is to find a solution which will provide us reliable and fast host-down info | 10:54 |
| Qiming | I started from Heat in attempt to do VM HA solely from inside Heat, but later on, I think I have to look into nova and ceilometer as well. | 10:54 |
| inc0 | there is some discussion going about ceilometer alarming | 10:55 |
| inc0 | but I'm afraid latency of this alarm might be too big for us, however thats something I'll check more thoroughly later | 10:55 |
| Qiming | I don't know how easy/difficult it is to have nova send out heartbeats, maybe nova guys don't like to send anything to Heat specifically | 10:55 |
| inc0 | afaik nova uses zookeeper underneath | 10:56 |
| Qiming | zookeeper is only one of the backend they used to maintain their service-group concept | 10:56 |
| inc0 | so we might try to tap into this zookeeper nova uses and use the source | 10:56 |
| Qiming | as a workaround, it is possible to have service_group implementation send out notifications as usual when a physical node is gone. | 10:57 |
| inc0 | Qiming, you mean nova service_group? | 10:58 |
| Qiming | yes | 10:58 |
| inc0 | we may talk to them and check how hard that might be | 10:58 |
| cmyster | hard? its implemented already | 10:59 |
| Qiming | I have a patch at hand that can do this for you | 10:59 |
| Qiming | but I haven't got cycles to propose it to nova | 10:59 |
| inc0 | well, I'll confirm that | 10:59 |
| Qiming | in addition to the host failure, you need to figure out which VMs were running on that specific host, before you send out the notification | 11:00 |
| inc0 | Qiming, also, restart them from given file | 11:00 |
| inc0 | and take care that host is connected to same shared storage (if thats the case) | 11:00 |
| inc0 | and then migrate network configuration | 11:00 |
| Qiming | you had better do this during host failure detection, or else, once you are out of nova domain, you need an admin privilege to figure out which VMs are impacted by a host failure. | 11:01 |
| inc0 | well, this is implementational detail imho, we will have lots of problems like that once we start making the actual thing | 11:02 |
| Qiming | currently, only ceilometer seems the sink where oslo notifications are actively checked | 11:02 |
| Qiming | to avoid duplicate your own event monitoring infrastructure in Heat, you can leverage ceilometer, where you can collect both host level failures and VM level failures | 11:03 |
| inc0 | if that will give us proper alarming with low latency, sure | 11:04 |
| Qiming | just like how auto-scaling is implemented today, ceilometer can translate these kinds of failure notifications into a resource signal back to Heat | 11:04 |
| inc0 | I'm concerned about adding another middleware | 11:04 |
| Qiming | well, it is up to Heat cores to decide whether introduce another daemon is a good idea, given that the convergence work is already planning to do some 'observing' operations | 11:05 |
| *** ishant has quit IRC | 11:05 | |
| inc0 | we could merge with convergence | 11:07 |
| Qiming | so, IMHO, if you want this VM HA to be done in Heat alone, you may want to sync with the convergence work | 11:07 |
| inc0 | this part | 11:07 |
| Qiming | yes, exactly. | 11:07 |
| shardy | \o/ ;) | 11:07 |
| Qiming | shardy, :) | 11:08 |
| *** gilliard is now known as gillilunch | 11:08 | |
| zaneb | inc0: your email is on my list of things to reply to today, but basically I agree with shardy | 11:08 |
| inc0 | hehe | 11:08 |
| *** nkhare has joined #heat | 11:08 | |
| zaneb | to the extent that we can do anything useful in Heat, your proposal duplicates the work of convergence | 11:09 |
| inc0 | not exacly, again, as I understand novergence its still for cattle vms | 11:10 |
| zaneb | and to the extent that it differs from convergence, it belongs probably in Nova rather than Heat | 11:10 |
| zaneb | convergence is for everything, VMs or not | 11:10 |
| *** sdake has quit IRC | 11:11 | |
| inc0 | but it will not restart *the same* vm from given place on shared storage with same configuration | 11:11 |
| Qiming | zaneb, even if nova can provide *some* useful hints about the availability of a VM as a blackbox, eventually, our users care more about their applications/services running inside the VMs | 11:11 |
| zaneb | inc0: right, and nor can we in Heat, so you'll have to build that in Nova | 11:11 |
| zaneb | as a bonus, anything you build in Nova can be used by anyone, whether or not they're using Heat | 11:12 |
| inc0 | zaneb, nova on the other hand doesn't really care which vm is stateful and which is not | 11:12 |
| *** sdake has joined #heat | 11:12 | |
| zaneb | neither does Heat | 11:12 |
| inc0 | I'll argue that autohealing, which its actually is, may fit into orchiestration | 11:12 |
| zaneb | at some point the user has to tell it | 11:12 |
| Qiming | some Heat native 'heartbeat' is needed to be sent from VMs, just like what cfn-pushstats can do | 11:12 |
| *** sorantis has quit IRC | 11:12 | |
| inc0 | zaneb, nothing is now, question is which will | 11:13 |
| inc0 | and yes, at the end user will define which vm will be covered with ha | 11:13 |
| shardy | Qiming: Not to heat, agent to send metric data direct to ceilometer IMO | 11:14 |
| Qiming | yes, shardy | 11:14 |
| inc0 | at current state openstack doesn't have tools at all to take care about vms on failed host | 11:14 |
| shardy | (which again is already possible if you don't mind creating some credentials and putting them inside the instance) | 11:15 |
| inc0 | convergence will restart parts of stack if host goes down | 11:15 |
| Qiming | inc0, nova has a 'evacuate' api/command, which allows you to rescue (remote restart) VMs on some other hosts | 11:17 |
| inc0 | it might be good idea to teach convergence to boot from shared storage, but as you said, I thing user should define which vms exacly should be booted | 11:17 |
| Qiming | however, the API/command is meant to be used by an admin only | 11:17 |
| Qiming | and you have to manually specify which host is the target | 11:17 |
| inc0 | there are bps for scheduled evacuate | 11:19 |
| inc0 | but there is nothing to trigger that | 11:19 |
| inc0 | automatically | 11:20 |
| inc0 | and there is nothing that will boot from existing volume | 11:20 |
| Qiming | inc0, that's true | 11:20 |
| inc0 | there is something for shared storage, but again, real part is to do it automatically | 11:21 |
| *** DaveJ__ has joined #heat | 11:22 | |
| openstackgerrit | Michal Jastrzebski (inc0) proposed a change to openstack/heat-specs: HAaaS Specification https://review.openstack.org/120051 | 11:24 |
| *** mkulke has joined #heat | 11:25 | |
| *** nkhare has quit IRC | 11:26 | |
| mkulke | hello there, i have a problem with a heat stack stuck in DELETE_IN_PROGRESS state. what can i do about this? | 11:26 |
| *** jstrachan_ has joined #heat | 11:28 | |
| *** jstrachan has quit IRC | 11:30 | |
| *** jstrachan has joined #heat | 11:32 | |
| inc0 | shardy, how will convergence check if host went down? | 11:33 |
| shardy | inc0: If nova loses an entire compute node, I'd expect it to emit some sort of notification, or heartbeat per-vm via ceilometer | 11:34 |
| *** tomek_adamczewsk has quit IRC | 11:34 | |
| inc0 | and convergence will restart parts of stack on different compute nodes right? | 11:35 |
| *** jstrachan_ has quit IRC | 11:35 | |
| Qiming | IIRC, nova today is not sending out a notification when a host is down, though the service_group knows it well ... | 11:36 |
| *** nikunj2512 has quit IRC | 11:36 | |
| *** nkhare has joined #heat | 11:38 | |
| *** sorantis has joined #heat | 11:39 | |
| Qiming | there IS a compute.instance.exists event, which can be treated as a 'heartbeat', IIUC | 11:39 |
| inc0 | Qiming, this is emitted every 20mins | 11:40 |
| *** che-arne has quit IRC | 11:40 | |
| inc0 | by default | 11:40 |
| inc0 | its waaay too long if we want to minimize downtime | 11:40 |
| Qiming | inc0, right, you may want to customize pipleline.yaml then | 11:40 |
| inc0 | also, I'm not sure if powered off instances emits this event | 11:41 |
| inc0 | and, we don't really care about that;) | 11:41 |
| inc0 | we care about hosts | 11:41 |
| inc0 | if compute node goes down, we don't need lack of compute.instance.exists to know that we should act | 11:42 |
| Qiming | inc0, that's why I was talking about a patch to nova, so that the internally maintained service_group can emit notifications when a compute node disappears | 11:42 |
| *** gokrokve has joined #heat | 11:46 | |
| inc0 | Qiming, yup, but again, we need to take care of fencing as well | 11:47 |
| inc0 | if compute node come back up to not restart vms | 11:47 |
| Qiming | inc0, exactly | 11:47 |
| inc0 | although I'm pretty sure it will not | 11:47 |
| inc0 | I'm not sure how about iptables neutron configs and so on | 11:48 |
| Qiming | the problem of fencing is that it can be done by an admin | 11:48 |
| Qiming | some of my coworkers have proposed fencing patches to nova, cinder and neutron | 11:48 |
| inc0 | yes, this system should work in admin context | 11:49 |
| inc0 | but users won't have access to system itself | 11:49 |
| inc0 | besides telling "I want this VM to be restarted on failure" | 11:49 |
| *** lazy_prince has quit IRC | 11:49 | |
| Qiming | right, it may not be a good idea to expose admin-only operations to Heat | 11:50 |
| Qiming | oh, to Heat users | 11:50 |
| inc0 | sure, users won't have anything to do with actual deamon | 11:50 |
| *** gokrokve has quit IRC | 11:51 | |
| *** nkhare has quit IRC | 11:51 | |
| *** rbowen has joined #heat | 11:51 | |
| *** nosnos has quit IRC | 11:52 | |
| *** nosnos has joined #heat | 11:52 | |
| *** tomek_adamczewsk has joined #heat | 11:53 | |
| *** andreaf_ has joined #heat | 11:54 | |
| *** andreaf has quit IRC | 11:56 | |
| *** elynn has quit IRC | 11:56 | |
| *** nosnos has quit IRC | 11:57 | |
| *** elynn has joined #heat | 11:57 | |
| *** denis_makogon has quit IRC | 11:59 | |
| *** denis_makogon has joined #heat | 11:59 | |
| *** gillilunch is now known as gilliard | 11:59 | |
| *** alexpilotti has joined #heat | 11:59 | |
| *** metral has quit IRC | 12:01 | |
| *** stevelle has quit IRC | 12:01 | |
| *** xmltok has quit IRC | 12:01 | |
| *** asalkeld has quit IRC | 12:01 | |
| *** sanjayu has quit IRC | 12:03 | |
| *** metral has joined #heat | 12:05 | |
| *** stevelle has joined #heat | 12:05 | |
| *** xmltok has joined #heat | 12:05 | |
| *** jstrachan_ has joined #heat | 12:07 | |
| *** jstrachan has quit IRC | 12:09 | |
| ekarlso- | anyone here wanna help me get tests going ? | 12:12 |
| ekarlso- | http://paste.openstack.org/show/108862/ atm | 12:12 |
| *** sgordon_ has joined #heat | 12:13 | |
| *** sgordon_ has quit IRC | 12:13 | |
| *** sgordon_ has joined #heat | 12:13 | |
| ekarlso- | noone willing to help with that ? | 12:15 |
| *** jstrachan_ has quit IRC | 12:17 | |
| *** jstrachan has joined #heat | 12:18 | |
| *** jdob has joined #heat | 12:18 | |
| *** lipinski1 has joined #heat | 12:20 | |
| *** KanagarajM has quit IRC | 12:21 | |
| *** julienve_ has joined #heat | 12:22 | |
| *** jstrachan_ has joined #heat | 12:22 | |
| *** uberj has quit IRC | 12:24 | |
| *** uberj_ has joined #heat | 12:24 | |
| ekarlso- | not easy to get help with something here : ) | 12:24 |
| *** swygue has quit IRC | 12:24 | |
| *** julienvey has quit IRC | 12:26 | |
| *** jstrachan_ has quit IRC | 12:27 | |
| *** jstrachan has quit IRC | 12:28 | |
| *** rpothier has joined #heat | 12:28 | |
| *** Drago1 has joined #heat | 12:29 | |
| *** MohitS has quit IRC | 12:32 | |
| *** pasquier-s has joined #heat | 12:32 | |
| *** lipinski1 has quit IRC | 12:33 | |
| Qiming | ekarlso-, your paste is not showing much useful info | 12:34 |
| ryansb | ekarlso-: it looks like your test run was killed (that's what Exception KeyboardInterrupt means) | 12:34 |
| ekarlso- | ryansb: well, I can do | 12:34 |
| ekarlso- | dunno why but now it works:1 | 12:35 |
| *** jstrachan has joined #heat | 12:35 | |
| ryansb | \o/ | 12:36 |
| ekarlso- | p = self.new_parameter('{"a": 1}', {'Type', 'Json'}, validate_value=False) < isn't that valid to create a parameter ? | 12:36 |
| *** julienve_ has quit IRC | 12:38 | |
| *** julienvey has joined #heat | 12:38 | |
| *** swygue has joined #heat | 12:38 | |
| ryansb | ekarlso-: no, you don't give the value at the time | 12:38 |
| ekarlso- | ryansb: got a clue on that ? | 12:38 |
| ryansb | so you'd do something like | 12:39 |
| *** hdd has joined #heat | 12:39 | |
| ryansb | p = self.new_parameter('a', {'Type': 'Json'}, validate_value=False) | 12:39 |
| ekarlso- | what's the 'a' or 'p' for ? | 12:40 |
| *** kopparam has quit IRC | 12:40 | |
| ryansb | umm, the "p =" is assigning the parameter to a variable | 12:40 |
| ryansb | the "a" is the name you're giving the parameter | 12:40 |
| ekarlso- | yea, I get that part | 12:41 |
| ekarlso- | but p = self.new_parameter('p', {'Type', 'Json'}, validate_value=False) | 12:41 |
| *** yassine has joined #heat | 12:41 | |
| ekarlso- | that tells me that the InvalidSchemaError: Invalid parameter (p), expected a mapping | 12:41 |
| *** jstrachan has quit IRC | 12:43 | |
| ryansb | you'll need a value as well | 12:45 |
| *** Drago1 has quit IRC | 12:45 | |
| ryansb | p = self.new_parameter('p', {'Type', 'Json'}, {"some": "map"}, validate_value=False) | 12:45 |
| *** gokrokve has joined #heat | 12:46 | |
| *** Drago1 has joined #heat | 12:47 | |
| inc0 | shardy, what do you think about idea of teaching convergence to schedule evacuate-vm on host failure? | 12:48 |
| *** jstrachan has joined #heat | 12:49 | |
| ekarlso- | ryansb: ahhhh | 12:49 |
| *** crose has joined #heat | 12:49 | |
| *** gokrokve has quit IRC | 12:50 | |
| *** achampion has quit IRC | 12:50 | |
| *** Drago1 has quit IRC | 12:50 | |
| *** che-arne|2 has joined #heat | 12:52 | |
| ekarlso- | ryansb: funny that heat tests all params but never the output of them like what they look like as strings :) | 12:52 |
| *** tango has joined #heat | 12:52 | |
| *** Drago has joined #heat | 12:53 | |
| zaneb | inc0: what do you think about the idea of teaching Nova to do that? | 12:54 |
| inc0 | zaneb, its my understatning that nova doesn't like expanding its features as is. Also, autohealing is a bit above nova in my opinion | 12:57 |
| inc0 | heat was my first guess in this matter | 12:57 |
| inc0 | this will be *aas, and nova is a bit lower level | 12:58 |
| zaneb | inc0: isn't evacuation an admin operation? | 12:59 |
| zaneb | Heat just doesn't concern itself with which machine a VM is running on | 12:59 |
| zaneb | that's supposed to be abstracted by Nova | 12:59 |
| inc0 | but then convergence is closely related to this, and convergence does concern itself with machine going down | 13:01 |
| inc0 | also, HARestartes is in heat after all (i didn't really know thats its considered obsolete:)) | 13:02 |
| inc0 | HARestarter* | 13:02 |
| Qiming | agree to zaneb, considering that many detection and recovery operations belongs to the admin domain. Once this is built into nova, Heat can leverage whatever HA properties that can be customized for VMs. | 13:02 |
| zaneb | HARestarter is an abomination | 13:02 |
| inc0 | allright, I didn't knew that;) | 13:03 |
| zaneb | and BTW, it doesn't work the way you think either | 13:03 |
| inc0 | what do you mean? | 13:03 |
| zaneb | to the extent that it can be said to "work" at all | 13:03 |
| zaneb | inc0: e.g. you said "equivalent to current HARestart mechanism, so simply restarting VM in case of alarm." | 13:05 |
| zaneb | HARestarter doesn't "simply restart" | 13:05 |
| zaneb | it doesn't restart at all | 13:06 |
| zaneb | it deletes half your template (try to guess which half) and recreates it | 13:06 |
| *** jasondotstar has joined #heat | 13:07 | |
| zaneb | inc0: also in your email you said "since HA will in fact be separate resource, we can actually enable it for existing VMs (not only those made by heat), much like current HARestarter." | 13:07 |
| *** tomek_adamczewsk has quit IRC | 13:07 | |
| zaneb | this is not even remotely true of the current HARestarter | 13:07 |
| zaneb | which depends absolutely on being defined in the same template as the server that it's monitoring | 13:08 |
| inc0 | ok, thanks for clearing that out, I've misunderstood code then | 13:09 |
| *** radez_g0n3 is now known as radez | 13:10 | |
| *** crose has quit IRC | 13:11 | |
| *** kopparam has joined #heat | 13:11 | |
| inc0 | well, I'll rethink whole idea then, thanks zaneb | 13:11 |
| zaneb | in general, whenever we create a resource in Heat that is just an interface to some code in Heat and not representing an actual real resource with a UUID and an OpenStack API, the result is an abomination | 13:13 |
| inc0 | well, heat could use this feature but now I agree its more nova use-case | 13:14 |
| inc0 | use in sense of heat deploying instances with ha-enabled (for example) | 13:15 |
| *** Drago has quit IRC | 13:15 | |
| *** lipinski1 has joined #heat | 13:16 | |
| ekarlso- | guys, can you really when adding parameter types actually test that heat represents them back to the user / client the way they are passed in ? ;) | 13:17 |
| zaneb | ekarlso-: patches welcome ;) | 13:17 |
| *** kopparam has quit IRC | 13:17 | |
| ekarlso- | zaneb: found 2 broken types atm :'( | 13:18 |
| openstackgerrit | Endre Karlson proposed a change to openstack/heat: Make str(json_param) return string vs python https://review.openstack.org/120084 | 13:19 |
| ekarlso- | there's the first one ;) | 13:19 |
| ekarlso- | need a better commit msg :§ | 13:19 |
| ekarlso- | zaneb: hmmm | 13:21 |
| ekarlso- | dunno but Jan's comment vs my change : https://bugs.launchpad.net/heat/+bug/1366798 | 13:21 |
| uvirtbot | Launchpad bug 1366798 in heat "JSON parameter is returned wrongly by the JsonParameter class" [High,Triaged] | 13:21 |
| ekarlso- | which one is correct ? | 13:21 |
| *** hdd has quit IRC | 13:23 | |
| ekarlso- | ryansb: / zaneb u guys got comments on that ? | 13:23 |
| zaneb | looking... | 13:23 |
| *** crose has joined #heat | 13:24 | |
| *** MohitS has joined #heat | 13:25 | |
| zaneb | ekarlso-: I think your patch is basically correct | 13:26 |
| zaneb | although I -1'd it for other reasons ;) | 13:27 |
| openstackgerrit | Endre Karlson proposed a change to openstack/heat: Make str(json_param) return correctly https://review.openstack.org/120084 | 13:27 |
| ekarlso- | ah ok | 13:27 |
| *** Drago1 has joined #heat | 13:29 | |
| *** inc0 has quit IRC | 13:30 | |
| zaneb | 0816b2963c4c334efe65f2a0cf2748ae6bc708c0 was the patch that commented out the json.dumps | 13:31 |
| *** Drago has joined #heat | 13:32 | |
| *** Drago1 has quit IRC | 13:33 | |
| *** Qiming has quit IRC | 13:36 | |
| *** sabeen has joined #heat | 13:36 | |
| *** Qiming has joined #heat | 13:37 | |
| *** Murali has joined #heat | 13:38 | |
| *** jmckind has joined #heat | 13:39 | |
| *** edmund has joined #heat | 13:39 | |
| openstackgerrit | Endre Karlson proposed a change to openstack/heat: Make str(json_param) return string vs python https://review.openstack.org/120084 | 13:41 |
| ekarlso- | updated zaneb | 13:41 |
| *** mkollaro has quit IRC | 13:43 | |
| *** achampion has joined #heat | 13:44 | |
| *** gokrokve has joined #heat | 13:46 | |
| *** crose has quit IRC | 13:46 | |
| *** jstrachan_ has joined #heat | 13:47 | |
| *** spzala has joined #heat | 13:49 | |
| *** jstracha_ has joined #heat | 13:50 | |
| *** jstrachan has quit IRC | 13:50 | |
| *** mkollaro has joined #heat | 13:50 | |
| *** gokrokve has quit IRC | 13:51 | |
| *** idanmo has joined #heat | 13:51 | |
| *** hdd has joined #heat | 13:51 | |
| *** jprovazn has quit IRC | 13:52 | |
| *** jstrachan_ has quit IRC | 13:53 | |
| *** jstrachan has joined #heat | 13:53 | |
| *** jprovazn has joined #heat | 13:55 | |
| *** pasquier-s_ has joined #heat | 13:55 | |
| *** rbrady has quit IRC | 13:56 | |
| *** idanmo has quit IRC | 13:57 | |
| *** jstracha_ has quit IRC | 13:57 | |
| *** pasquier-s has quit IRC | 13:58 | |
| *** crose has joined #heat | 13:59 | |
| *** nkhare has joined #heat | 14:02 | |
| *** idanmo has joined #heat | 14:03 | |
| erkules | ahoi, I want to template a cluster. For that I need to way to get the internal IPs in advance. Which is obvious not possible. Is there a common trick to get the IPs, at least while the template is executed? | 14:04 |
| *** jasond`` is now known as jasond | 14:05 | |
| *** sorantis has quit IRC | 14:06 | |
| *** jstrachan has quit IRC | 14:07 | |
| *** jstrachan has joined #heat | 14:08 | |
| *** tims has joined #heat | 14:08 | |
| *** sorantis has joined #heat | 14:09 | |
| openstackgerrit | Endre Karlson proposed a change to openstack/heat: Make str(json_param) return string vs python https://review.openstack.org/120084 | 14:10 |
| shardy | erkules: Normally you use the output attributes of one server (which include the network info) as input properties to the thing which needs to know the IP | 14:11 |
| shardy | here's an example: | 14:11 |
| shardy | https://github.com/openstack/heat-templates/blob/master/openshift-origin/F19/hot-template/openshift/openshift.yaml#L464 | 14:11 |
| ekarlso- | zaneb: pushed | 14:12 |
| zaneb | ekarlso-: +2, thanks | 14:13 |
| ekarlso- | zaneb: will fix the "list" / csv type also when I get back from mkdinner and gym :) | 14:13 |
| zaneb | cheers | 14:13 |
| erkules | shardy: thx. Is there a way to tell the other nodes about that attributes? | 14:19 |
| *** jstrachan_ has joined #heat | 14:23 | |
| *** zunguri has joined #heat | 14:25 | |
| *** david-lyle has joined #heat | 14:25 | |
| *** Drago has quit IRC | 14:26 | |
| *** jstrachan has quit IRC | 14:27 | |
| *** Drago1 has joined #heat | 14:27 | |
| *** gokrokve has joined #heat | 14:27 | |
| *** jistr has quit IRC | 14:28 | |
| *** Drago1 has quit IRC | 14:32 | |
| *** pasquier-s has joined #heat | 14:35 | |
| *** Drago has joined #heat | 14:35 | |
| *** jstrachan has joined #heat | 14:37 | |
| *** pasquier-s_ has quit IRC | 14:38 | |
| *** jstrachan_ has quit IRC | 14:39 | |
| *** rakesh_hs has quit IRC | 14:40 | |
| *** MohitS has quit IRC | 14:40 | |
| *** jstrachan_ has joined #heat | 14:40 | |
| *** MohitS has joined #heat | 14:41 | |
| *** lipinski1 has quit IRC | 14:42 | |
| *** jstrachan has quit IRC | 14:43 | |
| *** tomek_adamczewsk has joined #heat | 14:47 | |
| *** sergmelikyan has quit IRC | 14:48 | |
| *** sergmelikyan has joined #heat | 14:49 | |
| *** nkhare has quit IRC | 14:49 | |
| *** jistr has joined #heat | 14:52 | |
| *** jistr has quit IRC | 14:56 | |
| *** jistr has joined #heat | 14:57 | |
| *** aweiteka has quit IRC | 15:03 | |
| *** reed has joined #heat | 15:04 | |
| *** jstrachan has joined #heat | 15:06 | |
| *** jstrachan_ has quit IRC | 15:09 | |
| *** jstrachan_ has joined #heat | 15:11 | |
| *** MohitS has quit IRC | 15:12 | |
| *** jstrachan has quit IRC | 15:12 | |
| *** MohitS has joined #heat | 15:13 | |
| *** aweiteka has joined #heat | 15:18 | |
| *** david-lyle has quit IRC | 15:19 | |
| *** idanmo has quit IRC | 15:20 | |
| *** EricGonczer_ has joined #heat | 15:21 | |
| *** jprovazn has quit IRC | 15:21 | |
| *** Qiming has quit IRC | 15:23 | |
| *** pasquier-s_ has joined #heat | 15:24 | |
| *** pasquier-s has quit IRC | 15:27 | |
| *** lipinski1 has joined #heat | 15:27 | |
| *** andersonvom has joined #heat | 15:28 | |
| *** mkollaro has quit IRC | 15:34 | |
| *** mkollaro has joined #heat | 15:35 | |
| *** sorantis has quit IRC | 15:37 | |
| *** gstaicu has joined #heat | 15:39 | |
| gstaicu | hi | 15:39 |
| gstaicu | I am trying to create a stack on an openstack havana deploymnt | 15:40 |
| gstaicu | and I receive a message like this | 15:40 |
| gstaicu | Resource create failed: BadRequest: Unknown argment : | | | port (HTTP 400) | 15:40 |
| gstaicu | it says that there is an unknown argument | 15:41 |
| gstaicu | port | 15:41 |
| *** gokrokve_ has joined #heat | 15:41 | |
| gstaicu | can somebody explain me what can cause this error? | 15:41 |
| zaneb | gstaicu: what sort of resource was it? | 15:42 |
| gstaicu | first I am creating the resource | 15:42 |
| gstaicu | like this my_instance_port0: type: OS::Neutron::Port properties: network_id: ae3f1317-48d2-494c-af37-a6ae3838e280 fixed_ips: - subnet_id: 2e767321-f8ac-4484-bf54-9d0bbfa4922e | 15:42 |
| gstaicu | succesufully | 15:43 |
| gstaicu | I am verifying using neutron commands | 15:43 |
| zaneb | I wonder if your neutron client is too new | 15:43 |
| gstaicu | and then I am creating a new resource: my_instance | 15:43 |
| gstaicu | with properties: | 15:44 |
| gstaicu | networks: | 15:44 |
| larsks | gstaicu: I wonder if you could post your template somewhere using pastebin where it would be easier to look at it. | 15:44 |
| *** gokrokve has quit IRC | 15:44 | |
| gstaicu | ok | 15:44 |
| gstaicu | here it is: | 15:45 |
| gstaicu | http://pastebin.com/8sRbKM6m | 15:45 |
| larsks | gstaicu: what versions of heat and neutron are you running? | 15:46 |
| gstaicu | havana | 15:47 |
| gstaicu | 2013.2.3 | 15:48 |
| *** david-lyle has joined #heat | 15:48 | |
| *** cdent has quit IRC | 15:48 | |
| *** cdent has joined #heat | 15:49 | |
| larsks | zaneb: could that be too *old*? | 15:49 |
| larsks | gstaicu: I'm running Icehouse, and that templates works perfectly for me. | 15:49 |
| gstaicu | this is the message in the heat-engine.log | 15:49 |
| gstaicu | http://pastebin.com/7xhaT0y9 | 15:49 |
| *** jroyall has joined #heat | 15:49 | |
| *** jroyall is now known as joeroyall | 15:49 | |
| *** cdent has quit IRC | 15:50 | |
| *** jstrachan has joined #heat | 15:51 | |
| *** daneyon has quit IRC | 15:51 | |
| *** cdent has joined #heat | 15:51 | |
| gstaicu | it looks like on havana does not work | 15:51 |
| *** daneyon has joined #heat | 15:52 | |
| larsks | gstaicu: I don't have a havana environment handy, but it does work for me on icehouse. Looking at the heat havana sources, I *think* the port argument for the 'networks' property should work, but zaneb would in theory know for sure. | 15:52 |
| zaneb | I know nothing | 15:53 |
| *** jstrachan_ has quit IRC | 15:53 | |
| gstaicu | :) | 15:53 |
| *** bvandenh has quit IRC | 15:55 | |
| openstackgerrit | Steven Hardy proposed a change to openstack/heat: Tests update FakeKeystoneClient to align user_id and trustor https://review.openstack.org/120131 | 15:56 |
| openstackgerrit | Steven Hardy proposed a change to openstack/heat: Cleanup trust on delete with stored context if needed https://review.openstack.org/120132 | 15:56 |
| *** Drago has quit IRC | 15:56 | |
| *** Drago1 has joined #heat | 15:56 | |
| *** Drago1 has quit IRC | 15:56 | |
| *** jstrachan has quit IRC | 15:57 | |
| *** jstrachan has joined #heat | 15:58 | |
| *** yassine has quit IRC | 15:59 | |
| *** Drago has joined #heat | 16:00 | |
| *** ifarkas has quit IRC | 16:00 | |
| *** spzala_ has joined #heat | 16:02 | |
| *** jstrachan_ has joined #heat | 16:04 | |
| *** spzala has quit IRC | 16:04 | |
| *** spzala_ is now known as spzala | 16:04 | |
| *** jstrachan has quit IRC | 16:04 | |
| *** MohitS has quit IRC | 16:05 | |
| *** cdent has quit IRC | 16:05 | |
| *** Murali has quit IRC | 16:12 | |
| *** kebray has joined #heat | 16:16 | |
| *** kebray has quit IRC | 16:16 | |
| *** spzala has quit IRC | 16:17 | |
| *** daneyon has quit IRC | 16:18 | |
| *** daneyon has joined #heat | 16:18 | |
| *** jstrachan_ has quit IRC | 16:18 | |
| *** kebray has joined #heat | 16:19 | |
| *** kebray has quit IRC | 16:19 | |
| *** kebray has joined #heat | 16:19 | |
| *** kebray has quit IRC | 16:20 | |
| *** gstaicu has quit IRC | 16:20 | |
| *** kebray has joined #heat | 16:21 | |
| *** jstrachan has joined #heat | 16:23 | |
| *** aweiteka has quit IRC | 16:26 | |
| *** derekh has quit IRC | 16:27 | |
| *** jstrachan_ has joined #heat | 16:30 | |
| zaneb | if any cores want to review https://review.openstack.org/#/c/118562/ we can close one of our FFE bps | 16:32 |
| *** jstrachan has quit IRC | 16:32 | |
| *** pasquier-s has joined #heat | 16:32 | |
| *** gokrokve_ has quit IRC | 16:32 | |
| *** MohitS has joined #heat | 16:33 | |
| *** jstrachan has joined #heat | 16:33 | |
| *** david-lyle has quit IRC | 16:35 | |
| *** pasquier-s_ has quit IRC | 16:36 | |
| *** david-lyle has joined #heat | 16:36 | |
| *** jstrachan_ has quit IRC | 16:36 | |
| *** jstrachan_ has joined #heat | 16:38 | |
| *** aweiteka has joined #heat | 16:39 | |
| *** MohitS has quit IRC | 16:39 | |
| *** jmckind has quit IRC | 16:40 | |
| shadower | we've noticed this while doing stuff with tripleo templates: http://paste.openstack.org/show/108961/ | 16:40 |
| *** jstrachan has quit IRC | 16:40 | |
| shadower | and I can't decide if that's a bug or not | 16:40 |
| *** m_22 has joined #heat | 16:40 | |
| *** vijendar has joined #heat | 16:40 | |
| shadower | looks like YAML parses numeric and bool literals as numbers/booleans and Heat barfs when they get passed to a property that expects a string | 16:41 |
| zaneb | shadower: yeah, YAML is sometimes too clever for its own good | 16:41 |
| ryansb | you can just do "42" instead | 16:41 |
| zaneb | key_name: "42" will work, of course | 16:41 |
| shadower | yep I know that | 16:41 |
| shadower | I wonder if we should coerce these to strings anyways | 16:41 |
| *** m_22 has left #heat | 16:41 | |
| sjmc7 | it's a particular problem with hex strings | 16:41 |
| sjmc7 | where they can be all numeric and get treated as such | 16:41 |
| zaneb | I can't see any harm in coercing an int to a string | 16:42 |
| zaneb | the reverse would be more of a problem | 16:42 |
| shadower | yeah | 16:43 |
| ryansb | can you make keys in nova with int keys? | 16:43 |
| shadower | ryansb: that was just an example to illustrate the failure (it fails on validation -- before the stack even gets ceated) | 16:44 |
| ryansb | well sure, just curious if nova would barf if you tried to make a key named 42. | 16:45 |
| *** jstrachan_ has quit IRC | 16:45 | |
| *** pasquier-s has quit IRC | 16:45 | |
| *** pas-ha has quit IRC | 16:45 | |
| shadower | yeah | 16:45 |
| *** tomek_adamczewsk has quit IRC | 16:46 | |
| *** andreaf has joined #heat | 16:53 | |
| *** Daviey has quit IRC | 16:53 | |
| *** joeroyall has quit IRC | 16:54 | |
| *** kebray has quit IRC | 16:57 | |
| *** fayablazer has quit IRC | 16:58 | |
| *** Daviey has joined #heat | 16:59 | |
| openstackgerrit | unmesh-gurjar proposed a change to openstack/heat: Fixed adopt failure for stack with resource group https://review.openstack.org/119355 | 17:08 |
| *** coolsvap has quit IRC | 17:10 | |
| *** alexheneveld has quit IRC | 17:11 | |
| *** cmyster_ has joined #heat | 17:12 | |
| *** alexheneveld has joined #heat | 17:14 | |
| *** dsneddon has joined #heat | 17:17 | |
| *** gokrokve has joined #heat | 17:21 | |
| *** coolsvap has joined #heat | 17:24 | |
| *** harlowja_away is now known as harlowja | 17:25 | |
| *** spzala has joined #heat | 17:27 | |
| *** julienvey has quit IRC | 17:27 | |
| *** spzala has quit IRC | 17:27 | |
| *** julienvey has joined #heat | 17:28 | |
| *** Drago1 has joined #heat | 17:28 | |
| *** Drago has quit IRC | 17:28 | |
| *** cmyster_ has quit IRC | 17:30 | |
| *** julienvey has quit IRC | 17:32 | |
| *** Drago1 has quit IRC | 17:33 | |
| *** Drago1 has joined #heat | 17:34 | |
| *** jcoufal has quit IRC | 17:34 | |
| *** jistr has quit IRC | 17:35 | |
| *** Drago1 has quit IRC | 17:39 | |
| openstackgerrit | Endre Karlson proposed a change to openstack/heat: Make str(json_param) return string vs python https://review.openstack.org/120084 | 17:40 |
| *** sorantis has joined #heat | 17:41 | |
| *** Drago1 has joined #heat | 17:45 | |
| *** edmund has quit IRC | 17:47 | |
| *** spzala has joined #heat | 17:48 | |
| *** mkollaro has quit IRC | 17:49 | |
| *** david-lyle has quit IRC | 17:50 | |
| *** Drago1 has quit IRC | 17:51 | |
| *** kfox1111 has joined #heat | 17:56 | |
| *** devx has quit IRC | 18:03 | |
| *** devx has joined #heat | 18:03 | |
| openstackgerrit | Endre Karlson proposed a change to openstack/heat: Fix so that CSV as str return value https://review.openstack.org/120171 | 18:16 |
| *** andersonvom has quit IRC | 18:16 | |
| *** andersonvom has joined #heat | 18:17 | |
| *** julienvey has joined #heat | 18:19 | |
| ekarlso- | yet another fix :) | 18:19 |
| ekarlso- | if someone cares to take a look : ) | 18:20 |
| openstackgerrit | Endre Karlson proposed a change to openstack/heat: Fix so that CSV as str return value https://review.openstack.org/120171 | 18:28 |
| *** jdob has quit IRC | 18:29 | |
| *** jdob has joined #heat | 18:29 | |
| *** che-arne|2 has quit IRC | 18:29 | |
| *** uberj_ has quit IRC | 18:30 | |
| *** uberj has joined #heat | 18:30 | |
| *** jistr has joined #heat | 18:33 | |
| *** jistr has quit IRC | 18:34 | |
| *** jistr has joined #heat | 18:34 | |
| *** kebray has joined #heat | 18:36 | |
| *** kebray has quit IRC | 18:36 | |
| *** jdob has quit IRC | 18:36 | |
| *** jdob has joined #heat | 18:37 | |
| *** kebray has joined #heat | 18:37 | |
| *** sgordon_ has quit IRC | 18:59 | |
| *** sgordon_ has joined #heat | 19:00 | |
| *** sdake_ has joined #heat | 19:01 | |
| *** openstackgerrit has quit IRC | 19:02 | |
| *** sabeen has quit IRC | 19:03 | |
| *** sabeen has joined #heat | 19:04 | |
| *** julienvey has quit IRC | 19:08 | |
| *** DaveJ__ has quit IRC | 19:08 | |
| *** spzala has quit IRC | 19:12 | |
| *** kebray has quit IRC | 19:12 | |
| *** jroyall has joined #heat | 19:13 | |
| *** jroyall is now known as joeroyall | 19:13 | |
| *** tspatzier has joined #heat | 19:28 | |
| *** skyler has joined #heat | 19:28 | |
| skyler | Hi, so I just booted up a server with Heat. When I launch without heat, the default user is "ubuntu" when I launch with heat the default user is "ec2-user". Is there a way to disable all ec2 features like this? | 19:30 |
| sjmc7 | skyler - that's determined by heat config | 19:32 |
| sjmc7 | in heat.conf, 'instance_user' | 19:33 |
| skyler | sjmc7: Thanks, that is helpful. When I try to use sudo I get this message: sudo: /etc/sudoers.d/heat-instance-ec2-user is mode 0644, should be 0440 | 19:33 |
| sjmc7 | i don't think there is a way to disable it, though i would be happy to be corrected | 19:33 |
| skyler | Is this also something I can fix in heat.conf? | 19:34 |
| skyler | With the wrong mode, passwordless sudo doesn't work and I don't have a password for my initial user. | 19:35 |
| *** spzala has joined #heat | 19:35 | |
| sjmc7 | not seen that before, skyler | 19:35 |
| sjmc7 | where are you getting the image from? | 19:36 |
| skyler | The image is ubuntu cloud img, 12.04.4 amd64. | 19:36 |
| skyler | I don't have this problem when launching manually | 19:36 |
| *** tomek_adamczewsk has joined #heat | 19:38 | |
| *** julienvey has joined #heat | 19:38 | |
| sjmc7 | heat inserts some userdata which includes the default instance name when it calls out to nova | 19:39 |
| larsks | skyler: if you set "user_data_format: RAW" in your os::nova::server resource, Heat will not add it's own user-data information (which, among other things, sets up the ec2-user user) | 19:39 |
| skyler | larsks: Thanks! That is exactly the kind of thing I was hoping exists | 19:40 |
| *** david-lyle has joined #heat | 19:40 | |
| larsks | skyler: An example of that here: https://github.com/larsks/rdo-hangout-heat-intro/blob/master/templates/wp-naive.yaml#L156 | 19:40 |
| *** openstackgerrit has joined #heat | 19:42 | |
| *** kebray has joined #heat | 19:44 | |
| *** crose has quit IRC | 19:46 | |
| *** andersonvom has quit IRC | 19:51 | |
| *** andersonvom has joined #heat | 19:53 | |
| *** tomek_adamczewsk has quit IRC | 19:54 | |
| *** joeroyall has quit IRC | 19:57 | |
| *** jdob has quit IRC | 20:00 | |
| *** kebray has quit IRC | 20:00 | |
| *** jasond` has joined #heat | 20:01 | |
| *** andersonvom has quit IRC | 20:01 | |
| openstackgerrit | Ionut-Madalin Balutoiu proposed a change to openstack/heat-templates: HOT for IIS with Drupal https://review.openstack.org/120202 | 20:03 |
| openstackgerrit | Ionut-Madalin Balutoiu proposed a change to openstack/heat-templates: HOT for IIS with Drupal https://review.openstack.org/120202 | 20:04 |
| *** killer_prince has joined #heat | 20:06 | |
| *** killer_prince is now known as lazy_prince | 20:06 | |
| *** kebray has joined #heat | 20:08 | |
| *** andersonvom has joined #heat | 20:09 | |
| *** jistr has quit IRC | 20:14 | |
| *** dsneddon has quit IRC | 20:16 | |
| *** dsneddon has joined #heat | 20:16 | |
| *** sabeen1 has joined #heat | 20:20 | |
| *** sabeen has quit IRC | 20:21 | |
| *** tspatzier has quit IRC | 20:24 | |
| openstackgerrit | Ionut-Madalin Balutoiu proposed a change to openstack/heat-templates: HOT for IIS with Drupal https://review.openstack.org/120202 | 20:32 |
| *** julienvey has quit IRC | 20:33 | |
| *** radez is now known as radez_g0n3 | 20:37 | |
| kfox1111 | so, Kubernetes seems to compete with Heat and Nova? | 20:43 |
| ekarlso- | kfox1111: well, for docker.. | 20:43 |
| kfox1111 | yeah, but ideally, openstack would provide docker support. | 20:44 |
| kfox1111 | was looking at the openshift roadmap and they are reworking to use Kubernetes. | 20:44 |
| kfox1111 | Was hoping OpenShift would get more integrated with OpenStack, but seems like its going the other way. | 20:44 |
| ekarlso- | kfox1111: doesn't mean u can use openstack stuff to deploy services :) | 20:46 |
| ekarlso- | nor that u can't use ironic / nova to deploy a flurry of hosts with kubernetes on the, | 20:47 |
| kfox1111 | true. but when I looked at what Kubernates can provide, it doesn't seem to provide a lot of the useful features Heat provides, like nested templates. | 20:48 |
| kfox1111 | or Nova schedulaing antiafinity. | 20:48 |
| *** andreaf has quit IRC | 20:48 | |
| kfox1111 | Maybe I'll just have to wait for Solum to mature. | 20:48 |
| ekarlso- | kfox1111: cloudfoundry ? | 20:49 |
| ekarlso- | flynn | 20:49 |
| kfox1111 | maybe. haven't looked at it much yet. maybe time to do so. | 20:51 |
| *** jasondotstar has quit IRC | 20:56 | |
| *** rbowen has quit IRC | 21:03 | |
| zaneb | kfox1111: I don't think that Kubernetes competes with Heat, for the record | 21:05 |
| *** kebray has quit IRC | 21:10 | |
| *** kebray has joined #heat | 21:13 | |
| *** kebray has quit IRC | 21:13 | |
| *** kebray has joined #heat | 21:14 | |
| *** kebray has quit IRC | 21:14 | |
| *** EricGonczer_ has quit IRC | 21:19 | |
| openstackgerrit | Vijendar Komalla proposed a change to openstack/heat: Allow filtering of stacks by username, tenant and owner_id https://review.openstack.org/120235 | 21:20 |
| *** achampion has quit IRC | 21:22 | |
| *** vijendar has quit IRC | 21:23 | |
| *** zunguri has quit IRC | 21:24 | |
| *** zunguri has joined #heat | 21:24 | |
| *** xmltok has left #heat | 21:25 | |
| *** sorantis has quit IRC | 21:28 | |
| *** zunguri has quit IRC | 21:28 | |
| *** kebray has joined #heat | 21:34 | |
| *** jasond` has quit IRC | 21:38 | |
| *** rpothier has quit IRC | 21:47 | |
| *** 77CAAEEHU has joined #heat | 21:48 | |
| *** MohitS has joined #heat | 21:50 | |
| *** 77CAAEEHU is now known as che-arne | 21:50 | |
| *** asalkeld has joined #heat | 21:51 | |
| uberj | will heat tell me if I have a dependency cycle? | 21:54 |
| *** ccrouch has quit IRC | 21:56 | |
| *** Drago has joined #heat | 21:57 | |
| *** tango|2 has joined #heat | 21:59 | |
| *** tango has quit IRC | 22:01 | |
| *** Drago has quit IRC | 22:02 | |
| *** kebray has quit IRC | 22:04 | |
| stevebaker | uberj: yes | 22:10 |
| uberj | cool, thanks | 22:10 |
| *** andersonvom has quit IRC | 22:13 | |
| *** kebray has joined #heat | 22:15 | |
| *** kebray has quit IRC | 22:16 | |
| *** alexpilotti has quit IRC | 22:22 | |
| *** MohitS has quit IRC | 22:25 | |
| *** ccrouch has joined #heat | 22:28 | |
| *** sdake has quit IRC | 22:31 | |
| *** jroyall has joined #heat | 22:35 | |
| *** jroyall is now known as joeroyall | 22:35 | |
| *** kebray has joined #heat | 22:35 | |
| *** achampion has joined #heat | 22:40 | |
| kfox1111 | zaneb: https://github.com/GoogleCloudPlatform/kubernetes/blob/master/DESIGN.md under Is Kubernetes, then, a Docker "orchestration" system? | 22:50 |
| kfox1111 | seems like it has a lot of overlap. not as flexable as heat, and includes some nova like things like scheduling. | 22:51 |
| *** tims has quit IRC | 22:51 | |
| *** kebray has quit IRC | 22:53 | |
| *** zunguri has joined #heat | 22:54 | |
| *** zunguri has quit IRC | 22:58 | |
| *** ccrouch has quit IRC | 23:00 | |
| *** ccrouch has joined #heat | 23:00 | |
| stevebaker | kfox1111: that sounds like a reasonable comparison | 23:03 |
| *** ccrouch1 has joined #heat | 23:04 | |
| *** ccrouch has quit IRC | 23:05 | |
| *** hdd has quit IRC | 23:05 | |
| *** ccrouch1 has quit IRC | 23:08 | |
| kfox1111 | with openshift then basing stuff on it, running both openshift and openstack then someone needs to learn two different orchestration systems. :/ | 23:11 |
| kfox1111 | I'm kind of disapointed. | 23:11 |
| stevebaker | kfox1111: sure, but openshift needs to run on more than just openstack | 23:15 |
| *** david-lyle has quit IRC | 23:15 | |
| stevebaker | kfox1111: soon I'm going to write some software-config hooks so that a config resource represents a container on a server. One option I have is for the hook to just build a kubernetes template, so use kubernetes as a single-server container manager. Or the hook might just manage the containers directly | 23:18 |
| *** ccrouch has joined #heat | 23:19 | |
| kfox1111 | stevebaker: yeah, but if you can run run heat/nova+docker driver and get the same functionality + more then kubernetes, and still run in other clouds, why not? | 23:20 |
| kfox1111 | you should be able to run a minimal openstack in gce or azure or wherever in vm's, right? | 23:21 |
| stevebaker | kfox1111: that would be a bit crazy, but you could ;) | 23:22 |
| kfox1111 | yeah. that sounds good. the docker heat driver bothers me in the authentication bits. | 23:22 |
| *** spzala has quit IRC | 23:22 | |
| kfox1111 | stevebaker: not any more crazy then trying to reimplement nova/heat all over again. ;) | 23:22 |
| stevebaker | yeah, need to shut down that public docker api | 23:22 |
| kfox1111 | I still would like to see things in openstack go a slightly different way. consider this... you have three types of resources. bare metal, vm's, docker containers. | 23:23 |
| kfox1111 | nova has the capability to split a resource up in to several. | 23:23 |
| *** joeroyall has quit IRC | 23:24 | |
| kfox1111 | have a set of "trusted" images. images provided by the cloud provider. | 23:24 |
| kfox1111 | the user can request a bare metal node for example, and run a trusted image on it that has nova with the vm driver in it. when launched the tenant can't login to it, but it runs a new availability zone in it that only the tenant can schedual on. | 23:24 |
| kfox1111 | same with a vm -> docker nova instance. | 23:25 |
| kfox1111 | this allows users to request a bare metal node, or a vm node, and then suballocate it easly to themselves using the same openstack api. | 23:25 |
| kfox1111 | heat can then orchistrate the whole thing. | 23:26 |
| stevebaker | yes, that would be a nice flexible private cloud seup | 23:26 |
| stevebaker | setup | 23:26 |
| kfox1111 | that would be awesome. :) | 23:27 |
| kfox1111 | as a user, I wouldn't have to learn really anything new. and as a cloud provider, I wouldn't have to do much to provide the functionality. | 23:27 |
| kfox1111 | right now, I'm already overloaded a bit learning all of the openstack components, let alone having a completely different stack for docker. :/ | 23:28 |
| *** rbowen has joined #heat | 23:30 | |
| stevebaker | I think nova docker driver, the container service, and using heat to manage containers on your own VMs will all have their place in the future | 23:31 |
| kfox1111 | yeah. should be interesting times ahead. :) | 23:33 |
| *** david-lyle has joined #heat | 23:34 | |
| *** DrBacchus has joined #heat | 23:37 | |
| *** rbowen has quit IRC | 23:40 | |
| *** Qiming has joined #heat | 23:41 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!