| *** Pavo has joined #heat | 00:01 | |
| *** catintheroof has joined #heat | 00:04 | |
| *** dimtruck is now known as zz_dimtruck | 00:16 | |
| *** rajpatel has quit IRC | 00:19 | |
| *** adrian_otto has joined #heat | 00:19 | |
| *** jose-phillips has quit IRC | 00:22 | |
| *** catintheroof has quit IRC | 00:30 | |
| *** jaosorior has quit IRC | 00:46 | |
| openstackgerrit | Steve Baker proposed openstack/heat-agents master: Set labels on containers managed by docker-cmd https://review.openstack.org/428516 | 00:47 |
|---|---|---|
| *** liusheng has joined #heat | 00:56 | |
| *** zz_dimtruck is now known as dimtruck | 01:21 | |
| *** evilt0ne has quit IRC | 01:25 | |
| *** sdake has joined #heat | 01:28 | |
| *** chris_hultin|AWA is now known as chris_hultin | 01:41 | |
| *** evilt0ne has joined #heat | 01:43 | |
| *** evilt0ne has joined #heat | 01:43 | |
| *** chris_hultin is now known as chris_hultin|AWA | 01:43 | |
| *** zhurong has joined #heat | 01:52 | |
| *** Pavo has quit IRC | 02:01 | |
| *** Pavo has joined #heat | 02:05 | |
| *** adrian_otto has quit IRC | 02:06 | |
| *** gongysh has joined #heat | 02:11 | |
| *** dimtruck is now known as zz_dimtruck | 02:13 | |
| *** zz_dimtruck is now known as dimtruck | 02:19 | |
| *** Pavo_ has joined #heat | 02:24 | |
| *** spzala has quit IRC | 02:34 | |
| *** spzala has joined #heat | 02:34 | |
| *** spzala has quit IRC | 02:39 | |
| *** EricGonczer_ has joined #heat | 02:47 | |
| *** EricGonczer_ has quit IRC | 02:52 | |
| *** ricolin has joined #heat | 02:53 | |
| *** dimtruck is now known as zz_dimtruck | 02:54 | |
| *** ramishra has joined #heat | 03:09 | |
| *** Pavo_ has quit IRC | 03:13 | |
| *** KanagarajM_ has joined #heat | 03:29 | |
| *** zhurong_ has joined #heat | 03:36 | |
| *** rajpatel has joined #heat | 03:36 | |
| *** zhurong has quit IRC | 03:38 | |
| *** sdake has quit IRC | 03:41 | |
| *** Pavo has quit IRC | 04:01 | |
| *** Pavo has joined #heat | 04:06 | |
| *** sdake has joined #heat | 04:08 | |
| openstackgerrit | Merged openstack/heat master: Update reno for stable/ocata https://review.openstack.org/428283 | 04:08 |
| *** zhurong_ has quit IRC | 04:11 | |
| *** sdake_ has joined #heat | 04:11 | |
| *** nmathew has joined #heat | 04:12 | |
| *** sdake has quit IRC | 04:12 | |
| *** stingaci has quit IRC | 04:23 | |
| *** ricolin has quit IRC | 04:59 | |
| *** stingaci has joined #heat | 05:05 | |
| *** stingaci has quit IRC | 05:06 | |
| *** adrian_otto has joined #heat | 05:14 | |
| openstackgerrit | Rabi Mishra proposed openstack/heat master: Tail correct log files with devstack https://review.openstack.org/428574 | 05:19 |
| *** ricolin has joined #heat | 05:20 | |
| *** prateek has joined #heat | 05:25 | |
| *** rajpatel has quit IRC | 05:29 | |
| openstackgerrit | Rabi Mishra proposed openstack/heat master: Add 'default_pool' property to lbaasv2 listener https://review.openstack.org/419385 | 05:33 |
| openstackgerrit | Rabi Mishra proposed openstack/heat master: Add 'pools' attribute to lbaasv2 loadbalancer https://review.openstack.org/419383 | 05:33 |
| openstackgerrit | Rabi Mishra proposed openstack/heat master: Add 'loadbalancer' property to lbassv2 pool https://review.openstack.org/419377 | 05:33 |
| *** Pavo has quit IRC | 06:01 | |
| *** Pavo has joined #heat | 06:05 | |
| *** mvpnitesh has joined #heat | 06:13 | |
| mvpnitesh | Hi guys , Happy morning | 06:14 |
| mvpnitesh | I've a doubt, if we delete a stack , the stack details are still in database. Is it an expected behaviour | 06:14 |
| ramishra | mvpnitesh: yeah, that's expected, we soft delete stacks and they can be purged at regular intervals using #heat-manage purge_deleted | 06:22 |
| mvpnitesh | @ramishra : ok thank you | 06:23 |
| *** mvpnitesh has quit IRC | 06:27 | |
| *** mvpnitesh has joined #heat | 06:28 | |
| *** msno has joined #heat | 06:43 | |
| skraynev | ramishra: hi. are you around? | 06:45 |
| skraynev | ramishra: can we merge this patch https://review.openstack.org/#/c/428281/1 ? | 06:45 |
| *** zhurong has joined #heat | 06:45 | |
| ricolin | skraynev: I think we can right? | 06:46 |
| skraynev | ricolin: oh. it's already stable/ocata - yes :) we can :) | 06:46 |
| ricolin | skraynev: exactly:) | 06:47 |
| *** unlaudable has joined #heat | 06:47 | |
| *** mvpnitesh has quit IRC | 06:47 | |
| *** sdake_ has quit IRC | 06:52 | |
| *** adrian_otto has quit IRC | 07:20 | |
| *** adrian_otto has joined #heat | 07:26 | |
| *** rcernin has joined #heat | 07:26 | |
| *** adrian_otto has quit IRC | 07:33 | |
| *** Pavo has quit IRC | 08:01 | |
| *** Pavo has joined #heat | 08:05 | |
| openstackgerrit | Crag Wolfe proposed openstack/heat master: Store resource attributes in the DB https://review.openstack.org/422983 | 08:20 |
| openstackgerrit | Crag Wolfe proposed openstack/heat master: Data migration enabling storage of resource attributes https://review.openstack.org/422982 | 08:20 |
| *** jpich has joined #heat | 08:32 | |
| *** FL1SK has quit IRC | 08:32 | |
| *** mannamne has joined #heat | 08:38 | |
| *** mannamne has quit IRC | 08:52 | |
| *** shardy has joined #heat | 08:54 | |
| *** zzzeek has quit IRC | 09:00 | |
| *** zzzeek has joined #heat | 09:00 | |
| *** liusheng has quit IRC | 09:07 | |
| *** liusheng has joined #heat | 09:07 | |
| *** nilles has joined #heat | 09:07 | |
| *** derekh has joined #heat | 09:13 | |
| *** ricolin has quit IRC | 09:23 | |
| *** spzala has joined #heat | 09:35 | |
| *** chlong has quit IRC | 09:38 | |
| *** spzala has quit IRC | 09:40 | |
| *** blues-man has joined #heat | 09:52 | |
| *** zhurong has quit IRC | 09:55 | |
| *** gfidente has joined #heat | 09:58 | |
| *** gfidente has joined #heat | 09:58 | |
| *** evilt0ne has quit IRC | 09:59 | |
| *** Pavo has quit IRC | 10:01 | |
| *** Pavo has joined #heat | 10:05 | |
| *** gongysh has quit IRC | 10:19 | |
| openstackgerrit | Peter Razumovsky proposed openstack/heat master: Handle network property for Server on cancel-update https://review.openstack.org/415849 | 10:20 |
| *** msno has quit IRC | 10:33 | |
| *** mannamne has joined #heat | 10:34 | |
| *** FL1SK has joined #heat | 10:36 | |
| *** prazumovsky has joined #heat | 10:52 | |
| *** sdake has joined #heat | 11:18 | |
| *** gongysh has joined #heat | 11:24 | |
| *** jprovazn has joined #heat | 11:25 | |
| *** blues-man has quit IRC | 11:41 | |
| *** sdake_ has joined #heat | 11:43 | |
| *** sdake has quit IRC | 11:46 | |
| *** Pavo has quit IRC | 12:01 | |
| *** gongysh has quit IRC | 12:03 | |
| *** Pavo has joined #heat | 12:05 | |
| *** catintheroof has joined #heat | 12:20 | |
| *** catintheroof has quit IRC | 12:20 | |
| *** catintheroof has joined #heat | 12:21 | |
| *** sdake has joined #heat | 12:27 | |
| *** sdake_ has quit IRC | 12:30 | |
| *** rcernin has quit IRC | 12:31 | |
| *** KanagarajM_ has quit IRC | 12:49 | |
| openstackgerrit | Rabi Mishra proposed openstack/heat master: Use region_name when creating keystone client https://review.openstack.org/428718 | 12:58 |
| *** EricGonczer_ has joined #heat | 12:59 | |
| *** blues-man has joined #heat | 12:59 | |
| *** askb has quit IRC | 12:59 | |
| *** ccamacho is now known as ccamacho|lunch | 13:02 | |
| *** EricGonc_ has joined #heat | 13:03 | |
| *** EricGonczer_ has quit IRC | 13:03 | |
| *** zhurong has joined #heat | 13:09 | |
| *** KanagarajM_ has joined #heat | 13:13 | |
| *** dprince has joined #heat | 13:23 | |
| *** KanagarajM_ has quit IRC | 13:26 | |
| *** KanagarajM_ has joined #heat | 13:26 | |
| *** KanagarajM_ has quit IRC | 13:27 | |
| *** KanagarajM_ has joined #heat | 13:27 | |
| *** EricGonc_ has quit IRC | 13:28 | |
| *** shardy is now known as shardy_mtg | 13:34 | |
| *** hoonetorg has quit IRC | 13:35 | |
| *** spzala has joined #heat | 13:35 | |
| *** spzala has quit IRC | 13:36 | |
| *** spzala has joined #heat | 13:36 | |
| *** annegentle has joined #heat | 13:42 | |
| *** prateek_ has joined #heat | 13:43 | |
| *** mewald_ has joined #heat | 13:47 | |
| *** hoonetorg has joined #heat | 13:48 | |
| *** annegentle has quit IRC | 13:50 | |
| mewald_ | I am trying to prepare my images for use with OS::Heat::SoftwareDeployment using diskimage-builder. I believe all the software is there but found that at something is not working. Looking at /etc/os-collect-config.conf I found that the auth_url is wrong: https://gist.github.com/9abf5de846d69d56403ea8ee06347034 The URL you are seeing is the internalURL of Keystone which cannot be reached from the instance by design. How can I solve this | 13:51 |
| *** KanagarajM_ has quit IRC | 13:54 | |
| *** KanagarajM_ has joined #heat | 13:54 | |
| *** KanagarajM_ has quit IRC | 13:55 | |
| *** jpich has quit IRC | 13:55 | |
| *** KanagarajM_ has joined #heat | 13:55 | |
| *** annegentle has joined #heat | 14:00 | |
| *** Pavo has quit IRC | 14:01 | |
| *** Pavo has joined #heat | 14:05 | |
| *** KanagarajM_ has quit IRC | 14:07 | |
| *** KanagarajM_ has joined #heat | 14:08 | |
| *** shardy_mtg is now known as shardy | 14:11 | |
| *** annegentle has quit IRC | 14:17 | |
| *** ccamacho|lunch is now known as ccamacho | 14:20 | |
| *** spzala has quit IRC | 14:23 | |
| *** jtomasek has quit IRC | 14:25 | |
| *** jtomasek has joined #heat | 14:26 | |
| *** KanagarajM_ has quit IRC | 14:30 | |
| *** prazumovsky has quit IRC | 14:33 | |
| *** jcoufal has joined #heat | 14:35 | |
| *** paw has quit IRC | 14:39 | |
| *** Pavo_ has joined #heat | 14:42 | |
| *** zz_dimtruck is now known as dimtruck | 14:42 | |
| *** paw has joined #heat | 14:42 | |
| *** Pavo_ has quit IRC | 14:54 | |
| *** spzala has joined #heat | 14:55 | |
| *** spzala_ has joined #heat | 14:57 | |
| *** spzala_ has quit IRC | 14:57 | |
| *** spzala_ has joined #heat | 14:57 | |
| *** ricolin has joined #heat | 14:59 | |
| *** spzala has quit IRC | 15:00 | |
| *** chris_hultin|AWA is now known as chris_hultin | 15:01 | |
| *** crose has joined #heat | 15:02 | |
| *** jasond has joined #heat | 15:07 | |
| *** randallburt has joined #heat | 15:10 | |
| *** chris_hultin is now known as chris_hultin|AWA | 15:10 | |
| *** dimtruck is now known as zz_dimtruck | 15:10 | |
| *** zz_dimtruck is now known as dimtruck | 15:11 | |
| *** mewald_ has quit IRC | 15:13 | |
| *** randallburt has quit IRC | 15:15 | |
| *** adrian_otto has joined #heat | 15:20 | |
| *** stingaci has joined #heat | 15:20 | |
| *** ricolin has quit IRC | 15:21 | |
| *** prateek has quit IRC | 15:23 | |
| *** prateek_ has quit IRC | 15:23 | |
| *** stingaci has quit IRC | 15:24 | |
| *** chlong has joined #heat | 15:24 | |
| *** adrian_otto has quit IRC | 15:25 | |
| *** stingaci has joined #heat | 15:25 | |
| *** randallburt has joined #heat | 15:27 | |
| *** chris_hultin|AWA is now known as chris_hultin | 15:29 | |
| openstackgerrit | Zane Bitter proposed openstack/heat-templates master: Add template for autohealing servers https://review.openstack.org/428776 | 15:30 |
| *** rajpatel has joined #heat | 15:31 | |
| *** crose has quit IRC | 15:31 | |
| *** charliejllewelly has joined #heat | 15:31 | |
| *** unlaudable has quit IRC | 15:33 | |
| *** jcoufal_ has joined #heat | 15:34 | |
| *** jcoufal has quit IRC | 15:36 | |
| *** jaosorior has joined #heat | 15:40 | |
| *** nilles has quit IRC | 15:42 | |
| *** dimtruck is now known as zz_dimtruck | 15:48 | |
| *** unlaudable has joined #heat | 15:48 | |
| *** rajpatel1 has joined #heat | 15:52 | |
| *** rajpatel has quit IRC | 15:53 | |
| *** unlaudable has quit IRC | 15:54 | |
| *** Drago1 has joined #heat | 15:55 | |
| *** annegentle has joined #heat | 15:59 | |
| *** unlaudable has joined #heat | 15:59 | |
| *** Pavo has quit IRC | 16:00 | |
| *** jcoufal_ has quit IRC | 16:03 | |
| *** jcoufal has joined #heat | 16:03 | |
| *** Pavo has joined #heat | 16:05 | |
| *** unlaudable has quit IRC | 16:07 | |
| *** jcoufal_ has joined #heat | 16:07 | |
| *** sdake_ has joined #heat | 16:08 | |
| *** sdake has quit IRC | 16:08 | |
| *** sdake has joined #heat | 16:09 | |
| *** zz_dimtruck is now known as dimtruck | 16:09 | |
| *** jcoufal has quit IRC | 16:09 | |
| *** ccamacho has quit IRC | 16:11 | |
| *** jcoufal has joined #heat | 16:11 | |
| *** sdake has quit IRC | 16:12 | |
| *** sdake_ has quit IRC | 16:12 | |
| *** ccamacho has joined #heat | 16:13 | |
| *** jcoufal_ has quit IRC | 16:13 | |
| *** unlaudable has joined #heat | 16:20 | |
| *** unlaudable has quit IRC | 16:24 | |
| *** nmathew has quit IRC | 16:25 | |
| *** unlaudable has joined #heat | 16:26 | |
| *** adrian_otto has joined #heat | 16:28 | |
| *** stingaci has quit IRC | 16:32 | |
| *** zhurong has quit IRC | 16:34 | |
| *** chlong has quit IRC | 16:36 | |
| *** unlaudable has quit IRC | 16:41 | |
| *** adrian_otto has quit IRC | 16:51 | |
| *** chlong has joined #heat | 16:52 | |
| *** rajpatel1 has quit IRC | 16:52 | |
| *** d0ugal is now known as d0ugal|skiing | 16:54 | |
| *** stingaci has joined #heat | 16:58 | |
| *** adrian_otto has joined #heat | 17:02 | |
| *** blues-man has quit IRC | 17:06 | |
| *** rajpatel has joined #heat | 17:09 | |
| *** rwsu has quit IRC | 17:15 | |
| *** chlong has quit IRC | 17:16 | |
| *** annegentle has quit IRC | 17:25 | |
| *** EricGonc_ has joined #heat | 17:26 | |
| charliejllewelly | hey all...is it possible to use output from one resource in another in the same file? | 17:27 |
| *** annegentle has joined #heat | 17:28 | |
| Drago1 | charliejllewelly: Yes, that is what get_attr is for | 17:29 |
| charliejllewelly | Ah okay thanks | 17:29 |
| charliejllewelly | I tried that but I guess I'm not referencing it correctly | 17:30 |
| charliejllewelly | are there any examples? | 17:30 |
| *** derekh has quit IRC | 17:30 | |
| *** rajpatel has quit IRC | 17:30 | |
| *** rajpatel has joined #heat | 17:32 | |
| *** steve_mul has joined #heat | 17:33 | |
| zaneb | charliejllewelly: dozens http://git.openstack.org/cgit/openstack/heat-templates/tree/hot | 17:34 |
| charliejllewelly | amazing thanks | 17:35 |
| *** stingaci has quit IRC | 17:35 | |
| *** shardy has quit IRC | 17:38 | |
| *** ccamacho has quit IRC | 17:43 | |
| *** adrian_otto1 has joined #heat | 17:48 | |
| *** dprince has quit IRC | 17:50 | |
| *** adrian_otto has quit IRC | 17:50 | |
| *** EmilienM has quit IRC | 17:56 | |
| *** pkoraca has quit IRC | 17:56 | |
| *** mkoderer___ has quit IRC | 17:56 | |
| *** nwonknu has quit IRC | 17:56 | |
| *** alex_xu has quit IRC | 17:56 | |
| *** EmilienM has joined #heat | 17:56 | |
| *** EmilienM has quit IRC | 17:56 | |
| *** EmilienM has joined #heat | 17:56 | |
| *** andreaf has quit IRC | 17:56 | |
| *** kevinbenton has quit IRC | 17:56 | |
| *** shadower has quit IRC | 17:56 | |
| *** sirushti has quit IRC | 17:56 | |
| *** alex_xu has joined #heat | 17:56 | |
| *** shadower has joined #heat | 17:56 | |
| *** pkoraca has joined #heat | 17:56 | |
| *** mkoderer__ has joined #heat | 17:57 | |
| *** kevinbenton has joined #heat | 17:57 | |
| *** sirushti has joined #heat | 17:57 | |
| *** andreaf has joined #heat | 17:58 | |
| *** jose-phillips has joined #heat | 17:59 | |
| *** Pavo has quit IRC | 18:00 | |
| *** nwonknu has joined #heat | 18:02 | |
| *** blomquisg_gone is now known as blomquisg | 18:03 | |
| *** Pavo has joined #heat | 18:05 | |
| *** alex_xu has quit IRC | 18:06 | |
| *** alex_xu has joined #heat | 18:08 | |
| *** sabeen has joined #heat | 18:16 | |
| *** dprince has joined #heat | 18:20 | |
| *** akuznetsov has joined #heat | 18:22 | |
| *** mewald_ has joined #heat | 18:25 | |
| mewald_ | Is there any documentation on the various clients_* sections in heat.conf? I would like to understand what each means. | 18:25 |
| *** akuznetsov has quit IRC | 18:26 | |
| *** annegentle has quit IRC | 18:30 | |
| *** stingaci has joined #heat | 18:33 | |
| *** jcoufal_ has joined #heat | 18:35 | |
| *** stingaci has quit IRC | 18:36 | |
| *** jcoufal has quit IRC | 18:38 | |
| *** mannamne has quit IRC | 18:39 | |
| *** chlong has joined #heat | 18:39 | |
| *** prateek_ has joined #heat | 18:42 | |
| *** prateek has joined #heat | 18:42 | |
| *** chris_hultin is now known as chris_hultin|AWA | 18:44 | |
| *** gfidente is now known as gfidente|afk | 18:56 | |
| *** sdake has joined #heat | 18:56 | |
| *** rajpatel has quit IRC | 19:04 | |
| *** annegentle has joined #heat | 19:06 | |
| mewald_ | Is there any documentation on the various clients_* sections in heat.conf? I would like to understand what each means. | 19:09 |
| *** chlong has quit IRC | 19:11 | |
| *** chlong has joined #heat | 19:21 | |
| *** mewald_ has quit IRC | 19:25 | |
| *** rajpatel has joined #heat | 19:25 | |
| *** chlong has quit IRC | 19:27 | |
| openstackgerrit | Zane Bitter proposed openstack/heat master: Log stack status changes in unit tests https://review.openstack.org/429001 | 19:28 |
| openstackgerrit | Zane Bitter proposed openstack/heat master: Enable debug logging in unit tests https://review.openstack.org/429002 | 19:28 |
| *** prateek_ has quit IRC | 19:33 | |
| *** prateek has quit IRC | 19:33 | |
| *** stingaci has joined #heat | 19:35 | |
| *** jcoufal has joined #heat | 19:37 | |
| *** jcoufal_ has quit IRC | 19:37 | |
| *** chlong has joined #heat | 19:41 | |
| *** charliejllewelly has quit IRC | 19:43 | |
| *** stingaci has quit IRC | 19:48 | |
| *** jcoufal has quit IRC | 19:53 | |
| *** jcoufal has joined #heat | 19:53 | |
| *** adrian_otto1 has quit IRC | 19:55 | |
| *** adrian_otto has joined #heat | 19:57 | |
| *** Pavo has quit IRC | 20:01 | |
| *** rajpatel has quit IRC | 20:01 | |
| *** rajpatel has joined #heat | 20:02 | |
| *** Pavo has joined #heat | 20:05 | |
| *** jcoufal has quit IRC | 20:06 | |
| *** jcoufal has joined #heat | 20:06 | |
| *** EricGonc_ has quit IRC | 20:12 | |
| *** gfidente|afk has quit IRC | 20:14 | |
| openstackgerrit | Zane Bitter proposed openstack/heat master: Use stored value in OS::Heat::Value resource https://review.openstack.org/429024 | 20:15 |
| zaneb | cwolferh: ^ mea culpa | 20:15 |
| *** mannamne has joined #heat | 20:16 | |
| cwolferh | zaneb, i need to stare at that for awhile, i think that makes sense | 20:20 |
| *** jprovazn has quit IRC | 20:21 | |
| *** jcoufal_ has joined #heat | 20:23 | |
| *** jcoufal has quit IRC | 20:23 | |
| cwolferh | zaneb, sort of related, i think https://review.openstack.org/#/c/422983/ is working despite a test...signal failures. it still needs special logic to store template resource output/attributes though, which is where the big win really would be. | 20:25 |
| zaneb | cwolferh: interesting... I think that would tie in well to the stuff I'm working on atm. I'll try to post something WIP soon, it's getting pretty close | 20:30 |
| cwolferh | oh, and you may have noticed it's convergence only :) not sure non-convergence is worth it | 20:31 |
| *** steve_mul has quit IRC | 20:32 | |
| zaneb | agree | 20:33 |
| zaneb | cwolferh: I had this crazy thought a second ago... | 20:33 |
| *** stingaci has joined #heat | 20:34 | |
| zaneb | what if every output was just a hidden OS::Heat::Value resource? | 20:34 |
| *** chris_hultin|AWA is now known as chris_hultin | 20:34 | |
| *** adrian_otto has quit IRC | 20:40 | |
| zaneb | I think we'd still want to store attrs though, for the case where we're doing an update and don't change the resource, so we don't have to re-fetch 'em | 20:40 |
| cwolferh | zaneb, hmm. the problem with that is sort of related to the problem with your patch. sometimes the expression we are resolving has non_cacheable attributes | 20:41 |
| zaneb | and at that point I don't know if it's worth storing the outputs separately as well, since they should be ~fairly quickly computable | 20:41 |
| cwolferh | yeah, if we are storing only attributes, especially the outputs of a template resource which are also attributes, that should be good enough | 20:42 |
| zaneb | cwolferh: can you explain that previous point more? I don't know if I believe in non-cacheable attributes | 20:43 |
| cwolferh | i would like to not believe in them either | 20:43 |
| zaneb | let's say we don't believe in them | 20:44 |
| zaneb | life will be simpler | 20:44 |
| cwolferh | lol | 20:44 |
| zaneb | no, seriously | 20:44 |
| cwolferh | well, there are only two workarounds in my patch... one has to do with group size https://review.openstack.org/#/c/422983/8/heat/engine/resources/aws/autoscaling/autoscaling_group.py | 20:45 |
| cwolferh | which can change outside of a traversal | 20:45 |
| cwolferh | so i wipe out the stored attributes for that resource whenever it changes and life is good | 20:45 |
| cwolferh | but then there is https://review.openstack.org/#/c/422983/8/heat/engine/resources/openstack/neutron/lbaas/pool.py which is more irksome | 20:46 |
| *** dimtruck is now known as zz_dimtruck | 20:46 | |
| zaneb | is this something to do with https://review.openstack.org/#/c/94755/ ? | 20:46 |
| cwolferh | yes, i'm piggy backing off of that. poolmembers may be added to a pool after a pool has been updated in a traversal, so for now i'm just not storing attributes for a pool since one of its attributes is poolmembers | 20:48 |
| cwolferh | well, i should say i'm piggy backing off the existing CACHE_LOCAL and CACHE_NONE convention, it seems more or less aligned. i didn't dive into the history of it | 20:50 |
| zaneb | I mention because I find that patch bizarre... afaict it doesn't do *anything* in 99.8% of cases, which made me wonder whether the whole CACHE_LOCAL vs. CACHE_NONE thing was real or as much a figment of the imagination as the supposed performance benefits | 20:51 |
| cwolferh | CACHE_LOCAL gives you performance. CACHE_NONE gives extra security, i guess (haven't cared about that one too much yet), and maybe something more accurate for an underlying thing that changes a lot. i have to admit i don't get the first bullet in that patch commit "Derived from any resource's metadata, resource data or resource_id", like why is that not cacheable? | 20:54 |
| *** rajpatel has quit IRC | 20:54 | |
| zaneb | CACHE_LOCAL doesn't actually give you any performance though | 20:55 |
| cwolferh | oh, yeah, since the that cache intersects with input data, it makes sense you don't want cleartext secrets passed around in there | 20:55 |
| *** rajpatel has joined #heat | 20:56 | |
| zaneb | "Whenever any resource changes state, all resource attribute caches are cleared just in case the state change has side-effects in other resources." <- so that guarantees zero performance benefit except in the rare case where you refer to the same attribute twice or more in the same resource | 20:57 |
| zaneb | "* An API call which returns a secret" <- this makes no sense to me either, why would it matter? it's all in-memory anyway | 20:58 |
| cwolferh | from a convergence perspective, when a node propagates its attributes, that is stored in the sync_point. then when the stack is built in memory in the next resource, those attributes are the cached attributes, ie CACHE_LOCAL ones. i think. | 21:00 |
| zaneb | "* Derived from any resource's metadata, resource data or resource_id" <- this sort-of makes sense, in the sense that all of that stuff can already be cached in variously places, which already have their own invalidation logic | 21:01 |
| *** cmartin has joined #heat | 21:04 | |
| zaneb | cwolferh: yep, that's correct | 21:04 |
| *** sdake has quit IRC | 21:06 | |
| cwolferh | so that's the performance benefit, especially with your recent patch ;) maybe instead of CACHE_NONE we could have CACHE_SECRET for the secretive cases so it gets encrypted and passed along like everything else in the input_data | 21:07 |
| *** catintheroof has quit IRC | 21:11 | |
| zaneb | are we not storing the CACHE_NONE attributes in the SyncPoint? I was pretty sure we were | 21:12 |
| zaneb | the performance benefit came from convergence, not that patch, which empties the cache every time it's about to become useful | 21:13 |
| cwolferh | nevermind, i think we are now that i look at it. | 21:17 |
| *** askb has joined #heat | 21:18 | |
| *** askb has quit IRC | 21:22 | |
| *** rwsu has joined #heat | 21:23 | |
| cwolferh | so, for the one case that of poolmembers being an attribute of pool, 1/ it makes sense not to store them in the db and refer to them. 2/ maybe that patch makes more sense in non-convergence land if that pool is kept around in memory as part of a stack, but 3/ i don't see how it helps in convergence. | 21:25 |
| zaneb | I think we should deprecate the poolmembers attribute on the grounds that it's stupid | 21:29 |
| *** cmartin has quit IRC | 21:31 | |
| *** randallburt has quit IRC | 21:33 | |
| cwolferh | i'm not sure the pool's non-SHOW attributes even work. | 21:34 |
| zaneb | even more reason to delete them :) | 21:37 |
| zaneb | they probably do, NeutronResource has lots of magic in it | 21:37 |
| zaneb | cwolferh: btw did you see my comment on https://review.openstack.org/#/c/422983/8/heat/engine/resources/openstack/neutron/lbaas/pool.py ? | 21:38 |
| *** chlong has quit IRC | 21:44 | |
| cwolferh | zaneb, yes, replied. | 21:44 |
| Drago1 | I really wish there was an intermediate state to create/update in progress, something like "in progress but some attributes that other resources depend on are ready" | 21:44 |
| Drago1 | It would help avoid waiting on everything in a child template to complete when you need a server IP from it ASAP, for instance. | 21:45 |
| zaneb | cwolferh: no you haven't ;) | 21:46 |
| cwolferh | :D | 21:46 |
| cwolferh | now i have | 21:46 |
| zaneb | :D | 21:47 |
| *** chris_hultin is now known as chris_hultin|AWA | 21:47 | |
| zaneb | Drago1: yeah, at one point I thought about making outputs in a nested stack nodes in a dependency graph that spans across parent/child stacks | 21:48 |
| zaneb | Drago1: I can't remember why I gave up on that idea... probably because it seemed a little too ambitious | 21:49 |
| Drago1 | zaneb: I'm sure there are funny edge cases, but if every output was its own resource, they could go create_complete individually and kick off dependent resources | 21:51 |
| Drago1 | Same with resources attributes | 21:52 |
| Drago1 | *resource | 21:52 |
| zaneb | I can see the nested stack thing but I can't think of a legitimate use for different attributes becoming available at different times | 21:53 |
| *** jcoufal has joined #heat | 21:53 | |
| zaneb | Drago1: can you give a for-instance? | 21:53 |
| *** jcoufal_ has quit IRC | 21:54 | |
| Drago1 | zaneb: Not for the normal attribute case, no. Nested stacks are the only thing I can think of where outputs can become available at very different speeds | 21:55 |
| zaneb | yeah, ok, I agree that would be nice | 21:55 |
| *** stingaci has quit IRC | 21:55 | |
| *** spzala_ has quit IRC | 21:55 | |
| cwolferh | couldn't one break up one nested stack into two, where the fast things are in one stack, for instance? | 21:56 |
| Drago1 | cwolferh: Yes, and that's what I've been forced to do. It's uglier, but works. | 21:57 |
| *** adrian_otto has joined #heat | 21:57 | |
| zaneb | it's the classic question of whether you optimise your design for performance or maintainability | 21:58 |
| zaneb | (design in this context = how you arrange your resources in various template) | 21:59 |
| zaneb | it's clearly better if the same design can get you both performance + maintainability, so as a feature request this makes sense | 21:59 |
| *** Pavo has quit IRC | 22:00 | |
| zaneb | my hidden OS::Heat::Value resource idea could be part of the solution here, but the other part is to somehow knit together all of the dependency graphs of all the stacks in the tree. that part seems hard | 22:01 |
| *** stingaci has joined #heat | 22:04 | |
| *** Pavo has joined #heat | 22:05 | |
| *** jcoufal has quit IRC | 22:06 | |
| Drago1 | zaneb: If get_attr resolved to the hidden OS::Heat::Value, wouldn't it be knit correctly automatically? As long as the nested stack would properly update them whenever resources inside of itself went create_complete. | 22:07 |
| Drago1 | *If a get_attr targeting a nested stack resource | 22:08 |
| zaneb | there's no current way for a parent stack to watch resources in a child stack. they're not part of the same graph | 22:09 |
| Drago1 | zaneb: I'm not saying it would though. Instead, resources from a parent stack would "see" a OS::Heat::Value instead of the nested stack resource | 22:09 |
| zaneb | oh | 22:10 |
| *** jcoufal has joined #heat | 22:10 | |
| Drago1 | The OS::Heat::Value resources would be on the "outside" of the nested stack resource | 22:10 |
| zaneb | then you have the opposite problem, of how the child stack can update them | 22:11 |
| Drago1 | zaneb: Maybe I'm naive, but what would be difficult about letting the nested stack resource know about the synthetic Values? | 22:13 |
| Drago1 | Your idea of them being "hidden" implies some sort of special relationship to me anyway | 22:14 |
| *** dprince has quit IRC | 22:24 | |
| *** sdake has joined #heat | 22:31 | |
| *** sdake has quit IRC | 22:32 | |
| *** sdake has joined #heat | 22:32 | |
| *** jcoufal has quit IRC | 22:39 | |
| *** jcoufal has joined #heat | 22:40 | |
| *** chlong has joined #heat | 22:42 | |
| *** jcoufal has quit IRC | 22:50 | |
| *** stingaci has quit IRC | 22:51 | |
| *** jcoufal has joined #heat | 22:51 | |
| *** jaosorior has quit IRC | 22:52 | |
| *** stingaci has joined #heat | 22:53 | |
| *** mannamne has quit IRC | 22:56 | |
| *** stingaci has quit IRC | 22:57 | |
| *** jasond has quit IRC | 23:01 | |
| *** spzala has joined #heat | 23:02 | |
| *** chris_hultin|AWA is now known as chris_hultin | 23:02 | |
| *** chris_hultin is now known as chris_hultin|AWA | 23:03 | |
| *** Pavo_ has joined #heat | 23:05 | |
| *** adrian_otto has quit IRC | 23:05 | |
| *** jcoufal has quit IRC | 23:06 | |
| *** Pavo is now known as Guest29709 | 23:06 | |
| *** spzala has quit IRC | 23:06 | |
| *** Pavo_ has quit IRC | 23:08 | |
| *** Pavo has joined #heat | 23:09 | |
| *** jcoufal has joined #heat | 23:09 | |
| *** Pavo has quit IRC | 23:10 | |
| *** Guest29709 has quit IRC | 23:11 | |
| *** Pavo has joined #heat | 23:11 | |
| *** sdake has quit IRC | 23:11 | |
| *** Pavo has quit IRC | 23:17 | |
| *** Pavo has joined #heat | 23:17 | |
| *** EricGonczer_ has joined #heat | 23:20 | |
| *** jcoufal has quit IRC | 23:21 | |
| *** EricGonczer_ has quit IRC | 23:21 | |
| *** jcoufal has joined #heat | 23:21 | |
| *** EricGonczer_ has joined #heat | 23:22 | |
| *** EricGonczer_ has quit IRC | 23:23 | |
| *** zz_dimtruck is now known as dimtruck | 23:25 | |
| *** rajpatel has quit IRC | 23:28 | |
| *** adrian_otto has joined #heat | 23:30 | |
| *** annegentle has quit IRC | 23:35 | |
| *** dbecker has quit IRC | 23:36 | |
| *** rajpatel has joined #heat | 23:43 | |
| *** blomquisg is now known as blomquisg_gone | 23:43 | |
| *** paw has quit IRC | 23:51 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!