Wednesday, 2014-07-02

openstackgerrit	Steve Baker proposed a change to openstack/heat: Order intrinsic function docs alphabetically https://review.openstack.org/102717	00:00
openstackgerrit	Steve Baker proposed a change to openstack/heat: HOT intrinisic function list_join like Fn::Join https://review.openstack.org/102718	00:00
*** david-lyle has quit IRC		00:05
*** david-lyle has joined #heat		00:06
*** IlyaE has quit IRC		00:09
*** david-lyle has quit IRC		00:10
*** tango has quit IRC		00:16
openstackgerrit	A change was merged to openstack/heat: Fix pagination of stack-list https://review.openstack.org/101605	00:22
*** gokrokve has quit IRC		00:24
*** dims__ has quit IRC		00:25
*** arbylee has joined #heat		00:34
*** daneyon has quit IRC		00:39
*** daneyon has joined #heat		00:39
*** daneyon has quit IRC		00:39
*** shakamunyi has quit IRC		00:42
*** dims__ has joined #heat		00:43
*** alexheneveld has quit IRC		00:45
*** achampion has joined #heat		00:45
*** ccrouch has joined #heat		00:50
*** kfox1111 has quit IRC		00:55
*** TravT\|2 has quit IRC		00:57
*** nati_ueno has quit IRC		00:59
*** randallburt has quit IRC		01:00
*** shakamunyi has joined #heat		01:00
*** tiantian has joined #heat		01:00
*** m_22 has joined #heat		01:01
*** nati_ueno has joined #heat		01:01
openstackgerrit	liusheng proposed a change to openstack/heat: Use None instead of mutables in method params defaults https://review.openstack.org/103707	01:09
*** nati_ueno has quit IRC		01:09
openstackgerrit	Steve Baker proposed a change to openstack/heat: Catch NotFound for every stack delete db operation https://review.openstack.org/103715	01:10
openstackgerrit	Steve Baker proposed a change to openstack/heat: Sleep before stopping threads for delete https://review.openstack.org/103716	01:10
*** nati_ueno has joined #heat		01:10
tiantian	morning	01:10
stevebaker	zaneb: ^ is another crack at the delete race logging error, take a look at the commit message	01:10
*** shakamunyi has quit IRC		01:20
*** nosnos has joined #heat		01:49
*** gokrokve has joined #heat		01:49
*** nosnos has quit IRC		01:55
*** nosnos has joined #heat		01:55
*** Yanyan has joined #heat		02:04
*** nati_ueno has quit IRC		02:07
*** nati_ueno has joined #heat		02:10
*** andersonvom has joined #heat		02:12
elynn	Hi morning all :)	02:16
*** m_22 has quit IRC		02:17
*** danielbruno has quit IRC		02:17
*** nati_ueno has quit IRC		02:21
*** nati_ueno has joined #heat		02:21
*** zhiyan_ is now known as zhiyan		02:22
*** nati_ueno has quit IRC		02:22
*** zhiyan is now known as zhiyan_		02:30
*** Qiming has joined #heat		02:32
*** arbylee1 has joined #heat		02:35
*** arbylee has quit IRC		02:35
*** arbylee has joined #heat		02:37
*** arbylee1 has quit IRC		02:37
*** zhiyan_ is now known as zhiyan		02:42
*** Qiming has quit IRC		02:47
*** harlowja is now known as harlowja_away		02:50
*** ajc_ has joined #heat		02:50
zaneb	stevebaker: so the issue is the second time in a row that you stop the delete (i.e. the third consecutive delete command)?	02:52
stevebaker	zaneb: its timing sensitive, so its hard to tell if it is the third or fourth	02:53
zaneb	ok, but it can't happen on the second is I guess the point	02:54
stevebaker	zaneb: I could reproduce it with just 2 deletes. Even trying different delays between each call	02:55
zaneb	then I'm still confused :D	02:56
*** killer_prince is now known as lazy_prince		02:59
*** lazy_prince is now known as killer_prince		03:00
zaneb	on the 2nd delete, the probability that the first delete will finish within 0.2s is ~0	03:00
stevebaker	I couldn't reproduce it with 2 deletes, but that doesn't mean it can't be reproduced.	03:01
stevebaker	zaneb: unless its an empty stack, such as with the test in question	03:01
*** nati_ueno has joined #heat		03:02
zaneb	ok, I could believe that :)	03:03
zaneb	so we're acknowledging that this doesn't fix any real problems, it just makes tempest stop complaining	03:04
zaneb	stevebaker: why remove the threadgroup.wait() though? that throws us into a polling loop when we could be wait()ing	03:06
*** dims__ has quit IRC		03:09
stevebaker	zaneb: it seemed to not do anything. graceful=True does a wait anyway, and graceful=False each thread is killed. My interpretation of kill docs was that calls to thread.wait would immediately raise GreenletExit after kill has been called http://eventlet.net/doc/modules/greenthread.html (maybe I'm wrong about the "immediately")	03:10
*** ramishra has joined #heat		03:11
zaneb	yeah, it will always raise GreenletExit after that, but not immediately - it will still block until the thread finishes	03:11
zaneb	you're right that it makes no difference in the end, because we sleep() until the lock is released anyway	03:12
zaneb	but the current thread could get rescheduled multiple times during that, unnecessarily	03:12
*** cmyster has joined #heat		03:13
zaneb	I'll leave a review comment	03:13
stevebaker	zaneb: yeah, I'll put the wait back	03:14
zaneb	the eventlet documentation threw me the first 3 times I read it too	03:16
*** cmyster is now known as rabbi_cmyster		03:16
zaneb	they're pretty poor	03:16
openstackgerrit	Steve Baker proposed a change to openstack/heat: Sleep before stopping threads for delete https://review.openstack.org/103716	03:16
*** rabbi_cmyster is now known as cmyster		03:16
cmyster	morning	03:16
zaneb	stevebaker: thanks. I'm inclined to merge, but still struggling to figure out the root cause	03:19
stevebaker	I'm putting it down to internal SQLAlechemy-mumble-mumble not stopping gracefully when its thread gets killed	03:20
*** zhiyan is now known as zhiyan_		03:22
cmyster	a zombie...	03:22
cmyster	kill it with fire...	03:22
zaneb	that's a definite posibility. though when its thread gets killed, GreenletExit gets raised in it... so it definitely gets the chance to handle it gracefully	03:22
zaneb	stevebaker: we think it's the lock release that is failing?	03:23
*** zhiyan_ is now known as zhiyan		03:24
*** nosnos has quit IRC		03:27
*** zhiyan has left #heat		03:29
*** zhiyan has joined #heat		03:37
openstackgerrit	Zhang Yang proposed a change to openstack/heat: Refactor unit test test_neutron_autoscaling https://review.openstack.org/104061	03:38
*** zhiyan is now known as zhiyan_		03:39
zaneb	stevebaker: I've almost convinced myself it's a minor bug in SQLAlchemy	03:39
*** zhiyan_ is now known as zhiyan		03:40
*** TravT has joined #heat		03:45
*** zhiyan is now known as zhiyan_		03:46
*** sab has joined #heat		03:49
*** zhiyan_ is now known as zhiyan		03:50
*** Qiming has joined #heat		03:54
skraynev	Good morning!	03:58
cmyster	morning	04:01
openstackgerrit	A change was merged to openstack/heat: Fix pass incorrect volume_size while building bdm https://review.openstack.org/103822	04:02
*** andersonvom has quit IRC		04:03
*** TravT has quit IRC		04:03
openstackgerrit	liusheng proposed a change to openstack/heat: Fix H201 violations and re-enable gating https://review.openstack.org/102778	04:04
*** rakesh_hs has joined #heat		04:05
*** andersonvom has joined #heat		04:06
stevebaker	zaneb: agreed, and the error itself is not technically a race, but the condition that leads to it is	04:16
*** rakesh_hs has quit IRC		04:16
lifeless	stevebaker: hey q for you	04:17
stevebaker	lifeless: yup	04:17
lifeless	stevebaker: how can I, in a CFN template (sorry ;)) go from a OS::Neutron::Port -> the default route on the subnet	04:17
*** rakesh_hs has joined #heat		04:17
lifeless	stevebaker: specifically I want the gateway IP on the subnet from the first fixed_ip from the port object	04:17
lifeless	stevebaker: presume the subnet already exists (provider network)	04:18
*** tango has joined #heat		04:18
lifeless	stevebaker: I can pull out the subnet_id easily enough	04:18
lifeless	stevebaker: http://paste.ubuntu.com/7734743/	04:19
stevebaker	lifeless: off hand, I don't know	04:19
lifeless	stevebaker: I might be wrong, but I'm thinking its not possible	04:20
lifeless	stevebaker: k, thanks	04:20
lifeless	stevebaker: should I file a bug, do you think ?	04:20
stevebaker	lifeless: if you can't dig it out of the show attribute for the port then we could always add an attribute that exposes that	04:21
stevebaker	lifeless: sure, file a bug	04:21
lifeless	the port just has the subnet_id, not the subnet object AFAICT	04:22
lifeless	hmm	04:22
lifeless	https://bugs.launchpad.net/heat/+bug/1286128 claims to be able to do this	04:22
uvirtbot	Launchpad bug 1286128 in heat "Allow neutron network lookup by name" [Medium,Fix committed]	04:22
* lifeless digs deeper		04:22
stevebaker	any attribute which takes a network or subnet can now use uuid or name	04:23
lifeless	sorry for the burnt cycles. I may end up filing a doc bug somewhere	04:23
lifeless	stevebaker: its not passing in thats a problem ;)	04:23
lifeless	stevebaker: its reading-out	04:23
stevebaker	yeah	04:23
lifeless	stevebaker: we pass in network. Network has N subnets.	04:23
lifeless	stevebaker: one of those subnets is chosen. gateway ip is on that subnet	04:24
*** lazy_prince has joined #heat		04:25
*** kebray has joined #heat		04:27
*** rushiagr_ has joined #heat		04:30
*** rushiagr has joined #heat		04:31
*** rushiagr has quit IRC		04:34
*** bvivek has joined #heat		04:36
*** rakesh_hs2 has joined #heat		04:36
*** rakesh_hs has quit IRC		04:36
*** dims__ has joined #heat		04:37
*** ramishra has quit IRC		04:37
*** nosnos has joined #heat		04:42
*** dims__ has quit IRC		04:42
*** IlyaE has joined #heat		04:42
*** Guest1036 has quit IRC		04:43
*** lazy_prince has quit IRC		04:46
*** killer_prince is now known as lazy_prince		04:47
*** ramishra has joined #heat		04:49
*** kebray has quit IRC		04:53
*** kebray has joined #heat		04:54
*** andersonvom has quit IRC		05:00
*** dims__ has joined #heat		05:02
*** dims__ has quit IRC		05:07
openstackgerrit	Sirushti Murugesan proposed a change to openstack/heat: Remove newline from message to avoid truncating API response https://review.openstack.org/104068	05:07
openstackgerrit	Sirushti Murugesan proposed a change to openstack/heat: Avoid truncating API Response https://review.openstack.org/104068	05:09
*** kebray has quit IRC		05:14
*** tiantian has quit IRC		05:18
*** rushiagr has joined #heat		05:24
*** rwsu has quit IRC		05:24
*** cody-somerville has quit IRC		05:27
*** cody-somerville has joined #heat		05:27
*** shakamunyi has joined #heat		05:31
*** shakamunyi has quit IRC		05:32
*** shakamunyi has joined #heat		05:32
*** rushiagr has quit IRC		05:35
*** rushiagr has joined #heat		05:36
*** chandan_kumar has joined #heat		05:38
*** sab has quit IRC		05:39
*** sab has joined #heat		05:42
*** shakamunyi has quit IRC		05:42
*** shakamunyi has joined #heat		05:50
*** shakamunyi has quit IRC		05:50
*** bvivek has quit IRC		05:53
*** bvivek has joined #heat		05:56
openstackgerrit	OpenStack Proposal Bot proposed a change to openstack/heat: Imported Translations from Transifex https://review.openstack.org/89750	06:02
*** dims__ has joined #heat		06:03
*** nati_ueno has quit IRC		06:03
*** IlyaE has quit IRC		06:03
*** dims__ has quit IRC		06:08
*** mohits has joined #heat		06:09
*** unmeshg has joined #heat		06:09
*** nkhare has joined #heat		06:11
*** jstrachan has joined #heat		06:14
*** tiantian has joined #heat		06:14
*** nati_ueno has joined #heat		06:16
*** bvivek2 has joined #heat		06:25
*** bvivek has quit IRC		06:25
*** gokrokve has quit IRC		06:29
*** rushiagr_ has joined #heat		06:35
*** rushiagr has quit IRC		06:35
*** tango has quit IRC		06:38
*** ishant has joined #heat		06:45
*** julienvey has joined #heat		06:50
*** nati_ueno has quit IRC		06:51
*** rushiagr_ has quit IRC		06:52
*** TonyBurn_ has joined #heat		06:56
*** tomek_adamczewsk has joined #heat		06:58
*** avquadri has joined #heat		06:59
*** gokrokve has joined #heat		07:00
therve	Good morning!	07:01
*** gokrokve_ has joined #heat		07:02
*** gokrokv__ has joined #heat		07:04
*** rushiagr_ has joined #heat		07:04
*** gokrokve has quit IRC		07:04
*** jamiehannaford has joined #heat		07:04
*** gokrokve_ has quit IRC		07:07
*** julienvey has quit IRC		07:07
openstackgerrit	Ethan Lynn proposed a change to openstack/heat: Implement Flavor Extra-specs Key/Value Pairs https://review.openstack.org/102094	07:07
*** gokrokv__ has quit IRC		07:08
*** tspatzier has joined #heat		07:09
cmyster	morning therve	07:11
*** mohits has quit IRC		07:13
*** bvivek2 has quit IRC		07:14
*** mohits has joined #heat		07:17
*** maxskew has quit IRC		07:19
openstackgerrit	Ethan Lynn proposed a change to openstack/heat: Implement PrivateIpAddress property for EC2::Instance https://review.openstack.org/98580	07:20
*** tomek_adamczewsk has quit IRC		07:21
*** pasquier-s has joined #heat		07:22
skraynev	therve: hi, could you please check bug with 1336225. This id is mentioned in https://review.openstack.org/#/c/103842/. I can not find this bug too, but possibly it's related with rights.	07:22
Qiming	good afternoon	07:23
therve	skraynev, It was tagged as security	07:23
skraynev	therve: I suppose, that it's invisible according to some security policies.	07:23
therve	Although that's not the right process for security issues :/	07:24
skraynev	therve: Do you have some docs how to do id in right way?	07:24
therve	skraynev, https://wiki.openstack.org/wiki/VulnerabilityManagement	07:25
*** maxskew has joined #heat		07:25
therve	skraynev, I think the bug is public now, you should try again	07:25
skraynev	therve: thank you for the link. also you are right, it works for me now :)	07:26
*** bvivek has joined #heat		07:28
*** pas-ha has joined #heat		07:33
pas-ha	morning all	07:33
lifeless	stevebaker: filed https://bugs.launchpad.net/heat/+bug/1336656	07:37
uvirtbot	Launchpad bug 1336656 in heat "existing subnet not looked up on port?" [Undecided,New]	07:37
cmyster	morning pas-ha, Qiming	07:37
Qiming	cmyster, hi	07:38
*** jcoufal has joined #heat		07:38
pas-ha	cmyster, hi	07:38
therve	lifeless, Isn't there a typo? gatewway_ipa ?	07:39
therve	Sorry gateway_ipa	07:40
therve	Maybe it's a new beer brewed by neutron	07:40
lifeless	therve: that was a test I did, shouldn't have let it leak in	07:42
lifeless	therve: heat doesn't throw on bad attribute names	07:42
therve	lifeless, Not with Fn::Select no	07:42
*** mkerrin has joined #heat		07:45
lifeless	therve: makes it hard to see whats up :)	07:45
lifeless	therve: anyhow, just rechecking in case I fumble fingered	07:45
therve	lifeless, I don't think what you're trying to do works	07:47
cmyster	question: I have a stack with an issue in it, and via horizon I click on it. I see in horizon's log that "The server has either erred or is incapable of performing the requested operation." but the HTTP code is 500. What is the bug here? the code or the message ?	07:47
therve	Getting fixed_ips retrieve a dict with subnet_id, not a subnet object	07:47
lifeless	therve: I know it doesn't work, the question is should it :)	07:47
lifeless	therve: see https://review.openstack.org/#/c/82853/16/heat/engine/resources/neutron/port.py	07:47
lifeless	therve: and https://bugs.launchpad.net/heat/+bug/1286128	07:47
uvirtbot	Launchpad bug 1286128 in heat "Allow neutron network lookup by name" [Medium,Fix committed]	07:48
lifeless	therve: "	07:48
lifeless	Additional properties 'network' and 'subnet' are added to	07:48
lifeless	the resources (existing properties 'network_id' and"	07:48
therve	lifeless, That's a property, not an attribute	07:48
*** tomek_adamczewsk has joined #heat		07:50
Qiming	hi, therve	07:51
therve	(It's also not an object, but a name)	07:51
*** ifarkas has joined #heat		07:51
lifeless	therve: anyhow, so - I get that it doesn't work today. Seems desirable for it work though, no?	07:52
lifeless	or there to be some way to specify that I need an object representing that subnet somehow (and I can't know the subnet a-priori, since its chosen by Neutron)	07:53
therve	lifeless, Probably? Neutron is hard, I have no idea.	07:54
lifeless	therve: lol	07:54
lifeless	therve: landscape was hard :) Neutron is a walk in the park.	07:54
therve	lifeless, It was designed correctly though :)	07:55
therve	Neutron implicit relationships between resources is just awful	07:55
Qiming	therve: I'm wondering if we can make OS::stack_id (maybe parent_stack_id as well, if any) and OS::stack_name implicit entries in Nova server metadata.	07:56
lifeless	hmm I saw the thread on it but honestly didn't really get the issue	07:56
therve	Qiming, Why?	07:56
*** lsmola__ has joined #heat		07:57
Qiming	therve: it could be a convenience for in-server operations, but I'm not sure it will expose some security risks	07:57
therve	Qiming, I'd be tempted to just document it. Explicit is always better.	07:57
Qiming	therve: it will eliminate the 'matching_metadata' property we need to specify for AutoscalingGroup	07:57
therve	lifeless, Anyway do you want to rephrase that bug or should I?	07:58
therve	Qiming, I don't really want to eliminate that	07:58
tspatzier	Qiming, therve: there is a patch in review that does this, and it has a config option that brings with it the documentation.	07:59
Qiming	therve, any reasons why we require template authors to specify server.metadata and matching_metadata manually? I know it is not the only way to filter out metrics from ceilometer side.	07:59
*** tspatzier has quit IRC		08:00
therve	Qiming, I prefer it to be explicit.	08:01
therve	There are many ways you can write those rules, it's not clear why we should pick one	08:01
therve	It's really 2 lines in the template most of the time	08:02
Qiming	tspatzier, oh? will check it	08:02
*** gokrokve has joined #heat		08:04
*** dims__ has joined #heat		08:06
*** tspatzier has joined #heat		08:07
shardy	morning all	08:08
therve	tspatzier, Can you give the link to the patch	08:09
therve	Hi shardy	08:09
*** gokrokve has quit IRC		08:09
lifeless	therve: how would you like it rephrased? And sure.	08:10
tspatzier	therve: sure - but I just realised that the patch actually does that for scheduler hints, not for metadata. But could be very similar. One second ...	08:10
tspatzier	therve, Qiming: https://review.openstack.org/#/c/96889	08:11
*** dims__ has quit IRC		08:11
*** loki184 has joined #heat		08:11
therve	lifeless, Something like "Can't access subnet attributes from a port resource"	08:11
cmyster	morning shardy,	08:11
lifeless	therve: retitled	08:11
lifeless	therve: if you want more clarity, my brain is halting, so I'll let you do that ;)	08:12
*** mkollaro has joined #heat		08:14
*** julienvey has joined #heat		08:18
*** rakesh_hs2 has quit IRC		08:19
*** julienvey has quit IRC		08:22
*** rakesh_hs has joined #heat		08:22
*** bvivek has quit IRC		08:22
therve	tspatzier, Why is it a global configuration switch? It doesn't make much sense to me	08:26
Qiming	tspatzier, just took a quick glance over the patch, I agree to therve on the global tags	08:27
therve	I guess it's provider oriented?	08:27
skraynev	therve: do not you mind if missed place for deprecated client call be fixed in depended patch?	08:29
therve	missed place?	08:29
skraynev	therve: I mean https://review.openstack.org/#/c/97973/15	08:29
skraynev	read last comment	08:29
ramishra	:q!	08:30
*** gokrokve has joined #heat		08:30
ramishra	oops.. sorry	08:30
skraynev	skraynev: I missed it early :(	08:30
*** ramishra has quit IRC		08:30
therve	I guess that's fine if it's just a warning	08:30
*** ramishra has joined #heat		08:30
cmyster	shardy: did you find anything interesting with my keypair template ?	08:31
tspatzier	therve: right, this is a provider thing. The use case is that the provider has some nova schedule plugin that evaluates the hints to make placement decision based on the fact that servers belong to one stack	08:31
tspatzier	... and that seemed to make sense to me.	08:31
therve	Mokay	08:31
skraynev	therve: glad hear it.	08:32
shardy	cmyster: yes, basically the name should be optional	08:34
*** gokrokve has quit IRC		08:34
shardy	then you can just get_resource: the keypair resource from the server resource	08:34
openstackgerrit	liusheng proposed a change to openstack/heat: Use None instead of mutables in method params defaults https://review.openstack.org/103707	08:34
shardy	actually, you can do that now, but you need to ensure the parameter name for the key is unique	08:34
*** mkollaro has quit IRC		08:35
shardy	we should just default to the heat generated unique resource name	08:35
shardy	cmyster: So basically, you don't need the depends_on	08:35
liusheng	elynn: Hi, thanks for your review about https://review.openstack.org/10370 a new ps has been uploaded :)	08:36
cmyster	shardy: I see.	08:36
shardy	cmyster: http://paste.fedoraproject.org/114921/40429020	08:36
shardy	I'll post a patch later which makes the name optional and defaults to the resource name (which contains a random suffix)	08:37
cmyster	shardy: on that (almost) subject, I created a stack when some commented out lines, keypair related, and when horizon is asking for that stack we return 500 to it. I think that it should be 400 in this case:http://pastebin.test.redhat.com/219039	08:39
cmyster	I was trying to be a smart arse...	08:39
shardy	cmyster: lol ;)	08:40
shardy	Yeah 500 is never the right response, please can you raise a bug with a reproduce procedure?	08:40
cmyster	and yes I amended that particular template exactly as you	08:40
cmyster	oh	08:40
cmyster	yes I'll try to check that file history...	08:41
*** derekh_ has joined #heat		08:41
*** bvivek has joined #heat		08:41
*** FrancoisBillant has joined #heat		08:43
elynn	liusheng, I will review soon :)	08:43
*** jamiehannaford has quit IRC		08:44
*** jamiehannaford has joined #heat		08:44
*** jcoufal has quit IRC		08:48
*** bvandenh has joined #heat		08:48
openstackgerrit	Kanagaraj Manickam proposed a change to openstack/heat: Don't expose password in heat-keystone-setup-domain logs https://review.openstack.org/103842	08:51
*** jamiehannaford has quit IRC		08:52
*** jamiehannaford has joined #heat		08:56
*** julienvey_ has joined #heat		09:00
*** gokrokve has joined #heat		09:02
*** gokrokve has quit IRC		09:03
*** gokrokve has joined #heat		09:04
*** dims__ has joined #heat		09:07
openstackgerrit	Rabi Mishra proposed a change to openstack/python-heatclient: Alter misleading cli command 'resource-template' https://review.openstack.org/104118	09:08
*** gokrokve has quit IRC		09:09
*** dims__ has quit IRC		09:12
*** saju_m has joined #heat		09:12
openstackgerrit	Rabi Mishra proposed a change to openstack/python-heatclient: Alter misleading cli command 'resource-template' https://review.openstack.org/104118	09:12
*** alexheneveld has joined #heat		09:17
*** titish_maryam has joined #heat		09:18
*** mkollaro has joined #heat		09:19
*** titish_maryam has left #heat		09:19
*** bgorski has joined #heat		09:20
*** andrearosa has quit IRC		09:22
*** andreaf has joined #heat		09:24
*** tomek_adamczewsk has quit IRC		09:24
*** tomek_adamczewsk has joined #heat		09:27
*** ramishra has quit IRC		09:27
*** PragadeeswaranS has joined #heat		09:28
*** ramishra has joined #heat		09:35
*** loki184 has quit IRC		09:36
*** liusheng is now known as liusheng-leave		09:39
*** andrearosa has joined #heat		09:40
*** rakesh_hs has quit IRC		09:41
cmyster	shardy: I hit a snag with the template as you pasted it ^	09:48
cmyster	key_name is heat_test_key yet when I try to create a stack it is telling me that the key heat_key is already created ?	09:48
*** sorantis has joined #heat		09:55
*** rushiagr_ has quit IRC		09:56
shardy	cmyster: well presumably you already have a key with that name?	09:56
zhangyang	shardy:As I know, nove spec will be freezed at july 3, how about heat?	09:56
cmyster	no, its something different,	09:57
Qiming	zhangyang, I don't think that will apply to heat	09:57
shardy	zhangyang: we are not imposing a strict spec freeze like nova	09:57
cmyster	I am looking at it, I created a stack that might have made this one catch sometihng,	09:58
cmyster	I'll let you know :)	09:58
shardy	but if folks have major features they want in Juno, now is the time to propose them ;)	09:58
zhangyang	<shardy>: ok, thanks!	09:58
*** Dev_Jin has joined #heat		10:01
*** gokrokve has joined #heat		10:02
*** Dev_Jin has quit IRC		10:02
*** Dev_Jin has joined #heat		10:03
*** Qiming has quit IRC		10:04
Dev_Jin	Hi.. I am trying to configure AutoScalingGroup and ScalingPolicy using heat.. I am able to launch VM as per the policy and scale-in scale-out also working.. I have a use case where I want to monitor instances created as a part of AutoScalingGroup.. but what I am seeing is all the instance (which are part of group and which are not part of the group) have identical information in terms of project_id, user_id etc.. how do I differentiate which instance are pa	10:06
Dev_Jin	of group and which are ouside group?	10:06
Dev_Jin	I read somewhere about metadata injection.. how does it work?	10:06
*** gokrokve has quit IRC		10:07
*** dims__ has joined #heat		10:08
*** jcoufal has joined #heat		10:13
*** dims__ has quit IRC		10:13
*** Yanyan has quit IRC		10:15
zhangyang	<Dev_Jin>: you can see the instance metadata, if the instancs belongs to a scalinggroup, the group info could be found in metadata, this also is how celiometer monitor autoscalinggroup	10:18
openstackgerrit	Pavlo Shchelokovskyy proposed a change to openstack/heat: Implement OS::Sahara::NodeGroupTemplate resource https://review.openstack.org/100288	10:18
openstackgerrit	Pavlo Shchelokovskyy proposed a change to openstack/heat: Implement sahara client plugin https://review.openstack.org/100588	10:18
openstackgerrit	Pavlo Shchelokovskyy proposed a change to openstack/heat: Implement OS::Sahara::Cluster resource https://review.openstack.org/72336	10:18
openstackgerrit	Pavlo Shchelokovskyy proposed a change to openstack/heat: Implement OS::Sahara::ClusterTemplate resource https://review.openstack.org/101509	10:18
*** ramishra has quit IRC		10:21
*** ramishra_ has joined #heat		10:25
Dev_Jin	thanks for reply zhangyang.. instance metadata is empty. I am executing nova show <instance id>	10:29
Dev_Jin	am I missing something?	10:29
*** dims__ has joined #heat		10:32
*** ramishra_ has quit IRC		10:39
*** rakesh_hs has joined #heat		10:42
*** Qiming has joined #heat		10:47
zhangyang	Dev_Jin:did your scalinggroup have a tag start with 'metering.'? if that, heat won't add the groupname tag.	10:47
Dev_Jin	I see Tags property in AWS::AutoScaling::AutoScalingGroup .. but I am using OS::Heat::AutoScalingGroup.. and I don’t see any tags there..	10:51
*** tiantian has quit IRC		10:54
Dev_Jin	zhangyang: I see Tags property in AWS::AutoScaling::AutoScalingGroup .. but I am using OS::Heat::AutoScalingGroup.. and I don’t see any tags there..	10:55
zhangyang	if you use OS::Heat::AutoScalingGroup, I think you need to add Tag for your resource definition, I'm not sure.	10:55
*** gokrokve has joined #heat		11:02
*** loki184 has joined #heat		11:05
*** mkollaro has quit IRC		11:05
*** bvivek has quit IRC		11:08
*** gokrokve has quit IRC		11:08
*** mohits has quit IRC		11:14
*** rushiagr_ has joined #heat		11:19
*** ishant has quit IRC		11:21
*** pasquier-s has quit IRC		11:26
*** avquadri has quit IRC		11:39
*** sgordon has joined #heat		11:40
*** loki184 has quit IRC		11:41
*** sab has quit IRC		11:44
*** nkhare has quit IRC		11:45
*** mkollaro has joined #heat		11:46
cmyster	therve: high? really?	11:46
cmyster	(the 500 error bug)	11:47
therve	cmyster, Yeah	11:54
*** pasquier-s has joined #heat		11:54
therve	It's most likely a regression because of the oslo.messaging port	11:54
openstackgerrit	Zhang Yang proposed a change to openstack/heat: Get members from DB when deleting loadbalancer https://review.openstack.org/89359	11:55
openstackgerrit	Zhang Yang proposed a change to openstack/heat: Refactor unit test test_neutron_autoscaling https://review.openstack.org/104061	11:55
cmyster	k	11:57
therve	shardy, Hey, around? There is an issue with oslo.messaging errors I think	11:58
*** ajc_ has quit IRC		11:58
*** nosnos has quit IRC		11:59
cmyster	therve: I am trying to introduce a similar issue when horizon is trying to get a stack info on a faulty stack. but just like this, it usually works	11:59
cmyster	arrrg, reproduce*	11:59
therve	cmyster, Don't introduce issues :D	11:59
cmyster	hahaha	11:59
*** nosnos has joined #heat		11:59
cmyster	ever pretty ones ?	11:59
shardy	therve: H	11:59
shardy	Hi	11:59
therve	shardy, So exceptions which are not defined in heat.common.exceptions raise 500	12:00
therve	shardy, Any idea what to do?	12:00
shardy	therve: there's a mapping for each API, sec	12:00
shardy	https://github.com/openstack/heat/blob/master/heat/api/middleware/fault.py#L57	12:01
shardy	https://github.com/openstack/heat/blob/master/heat/api/aws/exception.py#L268	12:01
shardy	I assume there are some missing new types to add to those?	12:01
*** julienvey_ has quit IRC		12:01
shardy	we fall through to a 500 if the type isn't defined	12:02
*** julienvey_ has joined #heat		12:02
therve	Hum I didn't know this was required	12:02
*** gokrokve has joined #heat		12:02
* cmyster smells a new automated test		12:02
*** julienvey_ is now known as julienvey		12:03
*** nosnos has quit IRC		12:04
cmyster	bbl	12:05
*** julienvey has quit IRC		12:06
*** jdandrea has joined #heat		12:06
*** julienvey has joined #heat		12:07
*** gokrokve has quit IRC		12:07
*** bvandenh has quit IRC		12:08
*** tomek_adamczewsk has quit IRC		12:08
unmeshg	Hi, I came across a bug of supporting 'auto' as mountpoint	12:08
unmeshg	there are issues in using it, which I have summarized in the bug https://bugs.launchpad.net/heat/+bug/1260345	12:09
uvirtbot	Launchpad bug 1260345 in heat "OS::Cinder::VolumeAttachment does not support "auto" value for mountpoint" [Medium,Triaged]	12:09
unmeshg	I have also enlisted possible approaches for resolving it	12:09
shardy	unmeshg: last time I checked the issue was that it doesn't work in cinder rather than a heat specific issue	12:10
unmeshg	shardy: I tried it on master branch (on my devstack setup )	12:10
unmeshg	it works fine from nova CLI	12:11
*** tomek_adamczewsk has joined #heat		12:11
unmeshg	also works fine if the mountpoint is defined 'null' (instead of 'auto' ) in the heat template	12:11
unmeshg	please suggest the best approach	12:13
shardy	unmeshg: Oh right, I'm remembering incorrectly, I think my issue was when you actually specify mountpoint it's not respected	12:13
shardy	unmeshg: Looks like zaneb already has - if we make the property optional with no default then we'll pass None to novalclient if it's not specified	12:14
shardy	which will do the right thing without needing to specify anything won't it?	12:14
*** TonyBurn_ has quit IRC		12:15
*** alexpilotti has joined #heat		12:15
unmeshg	yes, with the docs specifying that approach, it would be best!	12:16
*** sorantis has quit IRC		12:20
unmeshg	shardy: thanks	12:20
*** unmeshg has quit IRC		12:21
*** unmeshg has joined #heat		12:22
*** maxskew has quit IRC		12:22
openstackgerrit	A change was merged to openstack/heat: Don't expose password in heat-keystone-setup-domain logs https://review.openstack.org/103842	12:23
*** maxskew has joined #heat		12:24
openstackgerrit	OpenStack Proposal Bot proposed a change to openstack/heat: Updated from global requirements https://review.openstack.org/96507	12:24
*** jdob has joined #heat		12:25
*** sorantis has joined #heat		12:26
*** achampion has quit IRC		12:27
openstackgerrit	OpenStack Proposal Bot proposed a change to openstack/heat: Updated from global requirements https://review.openstack.org/96507	12:33
therve	shardy, So the issue is not there	12:37
therve	It's a combination of allowed_remote_exceptions and exception default arguments	12:37
*** erecio has joined #heat		12:43
*** rushiagr_ is now known as rushiagr		12:48
*** unmeshg has left #heat		12:48
*** dims__ has quit IRC		12:52
*** dims__ has joined #heat		12:53
*** radez_g0n3 is now known as radez		12:55
jdandrea	Q: Mappings (HeatTemplateFormatVersion '2012-12-12') ... are they superseded by anything equivalent in heat_template_version '2013-05-23' ... ? (If so, I beg your pardon - I'm missing it!)	12:56
shardy	jdandrea: HeatTemplateFormatVersion is yaml rendered CFN	12:58
jdandrea	Exactly.	12:58
shardy	heat_template_version is the native HOT format	12:58
jdandrea	Someone here asked me if the HOT format had an equivalent to mappings. scratches head :)	12:58
shardy	jdandrea: No it doesn't	12:58
jdandrea	ok	12:58
shardy	jdandrea: it has the concept of an environment instead	12:58
jdandrea	Ah, true!	12:59
shardy	where you can specify parameters which in most cases could replace the mappings usage in CFN templates	12:59
jdandrea	So not the exact same thing, but ... yes. That.	12:59
jdandrea	Terrific.	12:59
*** gokrokve has joined #heat		13:02
*** chandan_kumar is now known as chandankumar		13:03
*** blues-man has joined #heat		13:04
*** gokrokve has quit IRC		13:07
*** pafuent has joined #heat		13:11
openstackgerrit	Pavlo Shchelokovskyy proposed a change to openstack/heat: UnitTests: fix mock assertions usage https://review.openstack.org/104187	13:14
openstackgerrit	Thomas Herve proposed a change to openstack/heat: Fix serialization of InvalidSchemaError https://review.openstack.org/104189	13:19
*** DaveJ__ has joined #heat		13:22
*** andersonvom has joined #heat		13:30
blues-man	hello	13:31
*** aweiteka has joined #heat		13:33
*** swygue has joined #heat		13:34
saju_m	Hi	13:35
Qiming	shardy?	13:36
saju_m	How to invoke the classes of puppet module "heat" ? What is the order ? I tried this http://dpaste.com/1V4YN00 , but not working .	13:37
saju_m	i invoked the class "heat" in the init.pp, https://github.com/stackforge/puppet-heat/blob/master/manifests/init.pp	13:38
therve	EmilienM, do you know? ^^	13:38
saju_m	but that did't install any heat services	13:38
shardy	Qiming: hi	13:39
Qiming	hi, shardy, not sure if this needs a core team review? https://review.openstack.org/#/c/96507/	13:39
*** arbylee has quit IRC		13:40
EmilienM	therve: saju_m looking now	13:40
*** arbylee has joined #heat		13:40
saju_m	<EmilienM> thanks	13:40
EmilienM	saju_m: you should switch on #puppet-openstack	13:40
therve	Qiming, It's a regular review, so yeah	13:40
Qiming	shardy: 30 patch sets now, I don't know the rules for reviewing/approving these changes	13:40
EmilienM	saju_m: you should not make order	13:40
EmilienM	it's within the module	13:41
tspatzier	shardy: do you have a minute? I would have some questions to a real authentication expert :)	13:41
Qiming	shardy, but it is still there to be merged	13:41
saju_m	<EmilienM> ok, let me ask this in #puppet-openstack	13:41
*** robsparker has joined #heat		13:42
* Qiming is all ears to authentication topic, because he is puzzled as well ...		13:42
*** rbuilta has joined #heat		13:42
*** jpeeler has quit IRC		13:45
*** jpeeler has joined #heat		13:45
*** jpeeler has quit IRC		13:45
*** jpeeler has joined #heat		13:45
*** lazy_prince is now known as killer_prince		13:45
*** rakesh_hs has quit IRC		13:50
*** DandyPandy has joined #heat		13:51
*** rushiagr has quit IRC		13:56
*** tango has joined #heat		13:59
*** IlyaE has joined #heat		14:02
*** gokrokve has joined #heat		14:02
*** ckmvishnu has joined #heat		14:04
*** JAGDISH__ has joined #heat		14:05
*** gokrokve has quit IRC		14:07
bgorski	ls	14:08
zaneb	fail :D	14:09
*** achampion has joined #heat		14:09
bgorski	:)	14:10
*** rwsu has joined #heat		14:10
*** erecio has quit IRC		14:10
*** BillArnold has quit IRC		14:11
*** andersonvom has quit IRC		14:14
*** BillArnold_ has joined #heat		14:14
openstackgerrit	Pavlo Shchelokovskyy proposed a change to openstack/heat: Implement CFN CancelUpdateStack action https://review.openstack.org/102852	14:26
openstackgerrit	Pavlo Shchelokovskyy proposed a change to openstack/heat: Implement cancelling of stack update https://review.openstack.org/102851	14:26
*** gokrokve has joined #heat		14:26
*** gokrokve_ has joined #heat		14:27
*** gokrokve has quit IRC		14:31
*** zz_gondoi is now known as gondoi		14:31
shardy	tspatzier: Hi	14:33
shardy	sorry, was afk for a few minutes	14:33
tspatzier	hey shardy, no problem. I'm in a call at the moment. Will ping you when done.	14:34
*** sabeen2 has joined #heat		14:34
shardy	tspatzier: Ok, np	14:34
*** jcoufal has quit IRC		14:34
*** jcoufal has joined #heat		14:36
*** sabeen2 has quit IRC		14:37
*** sabeen1 has quit IRC		14:38
*** sabeen2 has joined #heat		14:38
*** sorantis has quit IRC		14:38
*** mkollaro has quit IRC		14:40
*** saju_m has quit IRC		14:42
*** pasquier-s has quit IRC		14:43
*** gokrokve_ has quit IRC		14:43
*** gokrokve has joined #heat		14:47
*** daneyon has joined #heat		14:47
*** daneyon has quit IRC		14:47
*** daneyon has joined #heat		14:48
*** sorantis has joined #heat		14:50
*** andreaf has quit IRC		14:51
*** arunrajan has joined #heat		14:54
*** arunrajan has quit IRC		14:54
*** arunrajan has joined #heat		14:54
*** dims__ has quit IRC		14:55
*** pas-ha has quit IRC		14:58
*** nkhare has joined #heat		15:00
*** saju_m has joined #heat		15:01
*** rushiagr has joined #heat		15:02
*** Dev_Jin has quit IRC		15:09
*** gokrokve has quit IRC		15:10
*** ckmvishnu has quit IRC		15:10
*** gokrokve has joined #heat		15:11
*** Dev_Jin has joined #heat		15:15
*** PragadeeswaranS has quit IRC		15:16
openstackgerrit	Steven Hardy proposed a change to openstack/heat: Add OS::Ironic::Chassis resource https://review.openstack.org/103133	15:16
openstackgerrit	Steven Hardy proposed a change to openstack/heat: Add Ironic resource base class https://review.openstack.org/104222	15:16
openstackgerrit	Steven Hardy proposed a change to openstack/heat: Add OS::Ironic::Node resource https://review.openstack.org/104223	15:16
openstackgerrit	Steven Hardy proposed a change to openstack/heat: Add OS::Ironic::Port resource https://review.openstack.org/104224	15:16
*** ericgoncz has joined #heat		15:17
*** kebray has joined #heat		15:18
jasond	shardy: hey	15:18
*** packet has joined #heat		15:19
openstackgerrit	A change was merged to openstack/heat: Use None instead of mutables in method params defaults https://review.openstack.org/103707	15:19
shardy	jasond: Hi!	15:20
jasond	shardy: i'm trying to understand how update works with wait conditions	15:20
jasond	it seems like AWS doesn't support it	15:20
jasond	were you trying to support updates in the wait condition resource you're working on?	15:21
*** randallburt has joined #heat		15:21
*** randallburt has quit IRC		15:21
*** randallburt has joined #heat		15:21
shardy	jasond: I was planning to, but the version posted currently doesn't	15:21
shardy	jasond: I was thinking add a "update" boolean property for the handle, which makes it behave like UpdateWaitConditionHandle on update	15:22
*** dims__ has joined #heat		15:23
jasond	ah, i guess i need to see how UpdateWaitConditionHandle works	15:23
shardy	jasond: It does a slightly weird thing overriding the base-class update method	15:24
shardy	see the original review SpamapS posted for details of why, but basically handle_update isn't called if you don't change the properties	15:24
*** rushiagr has quit IRC		15:25
*** Dev_Jin has quit IRC		15:25
shardy	jasond: did you have time to think about the base-class and interfaces discussed in your review?	15:25
*** gokrokve has quit IRC		15:25
shardy	It would be nice to make the properties for the native API based resource and the swift one compatible	15:26
shardy	so they can be easily interchangeable and/or mapped via the global environment	15:26
jasond	i did. there are some parts of that base class that aren't relevant to the swift resource	15:26
jasond	swift-backed wait condition	15:26
*** rushiagr has joined #heat		15:26
shardy	jasond: yup, I assumed we might need further refactoring, it was mostly the interfaces I was concerned about	15:27
shardy	we can always resolve duplication later :)	15:27
jasond	they're aligned now. see my update from yesterday	15:27
*** dims__ has quit IRC		15:27
shardy	jasond: ah! Thanks, I'm a bit behind on reviews this week	15:27
*** dims__ has joined #heat		15:27
jasond	np	15:28
jasond	although the swiftwaitconditionhandle has a timeout property	15:28
jasond	i guess it could use a default	15:28
jasond	i have to run to a meeting, be back in half an hour	15:29
*** saju_m has quit IRC		15:29
tspatzier	shardy: ping	15:32
*** gondoi is now known as zz_gondoi		15:32
shardy	tspatzier: Hi	15:32
*** sorantis has quit IRC		15:33
tspatzier	shardy: I have an authentication related issue with software config and heat in standalone multi cloud mode - not sure if anyone tried that so far ...	15:33
*** sorantis has joined #heat		15:33
shardy	tspatzier: probably not, auth for signalling gets hard in standalone mode	15:33
tspatzier	are you familiar with how access to metadata and signalling works for software config when using cfn?	15:33
shardy	tspatzier: yes, pretty much	15:34
shardy	not so much the specifics of standalone or multi-cloud though tbh	15:34
tspatzier	shardy: good, I thought so :)	15:34
*** KanagarajM has joined #heat		15:34
*** zz_gondoi is now known as gondoi		15:35
tspatzier	shardy: So here is what I tried and what the result was: I set up heat standalone multi cloud and deployed a template with software config to a remove devstack. For that I have to specify the auth-url on the command line, and I had configured it before as an allowed uri in heat.conf	15:35
*** sorantis has quit IRC		15:35
tspatzier	I guess this also generated EC2 tokens for the instances so the in-instance tools can query the metadata. and the remote keystone is used for this token generation.	15:36
tspatzier	Now when the instances talk back to the standalone heat engine, it tries to validate the tokens against its local keystone ... which fails. Unless I hardcode the remote auth uri in some more properties in heat.conf. But that seems to void the multi-cloud idea, right?	15:38
shardy	tspatzier: Yeah, heat in standalone mode uses the auth_password middleware doesn't it?	15:38
shardy	so the ec2tokens auth won't work	15:39
tspatzier	I'm not too familiar with all that but just found out that the token validation seems to read the auth uri properties from the config file.	15:39
openstackgerrit	A change was merged to openstack/heat: UnitTests: fix mock assertions usage https://review.openstack.org/104187	15:40
tspatzier	shardy: will the same scenario be possible once the native heat api and signalling works?	15:40
openstackgerrit	A change was merged to openstack/heat: use register_cli_opts instead of register_opts https://review.openstack.org/103847	15:40
shardy	tspatzier: Yes the problem is the signed URL generated with the keypair needs to be validated via the ec2tokens auth middleware, which presumably isn't configured if you're in standalone mode	15:40
shardy	tspatzier: You may be better using the HEAT_SIGNAL transport instead, because that uses a username and randomly generated password (which get deployed inside the instance)	15:41
openstackgerrit	OpenStack Proposal Bot proposed a change to openstack/heat: Updated from global requirements https://review.openstack.org/96507	15:42
*** jcoufal has quit IRC		15:42
tspatzier	shardy: so in the case of HEAT_SIGNAL the standalone heat engine will not have to go to keystone (using the wrong one in standalone mode) and verify?	15:42
shardy	tspatzier: by default the agents probably get a token from keystone then send that to heat, so they'd need to be configured to instead send the username/password so the auth_password middleware would work	15:42
shardy	tspatzier: the problem is the other middlewares need to be configured with an admin service user which does the token validation	15:44
*** chandankumar has quit IRC		15:44
shardy	and you don't have that in standalone mode	15:44
shardy	tspatzier: python-heatclient can be configured to send the username/password to heat, but I'm not sure about the os-* agents	15:45
*** tomek_adamczewsk has quit IRC		15:45
shardy	tspatzier: probably a question for SpamapS/stevebaker	15:45
tspatzier	shardy: I think since the heat collector plugin could do it and os-collect-config would not care.	15:46
tspatzier	shardy: so when the heat client sends username and password, where would the standalone heat engine go an verify it?	15:46
shardy	tspatzier: I assume you don't have the stack user domain configured, and that the user creating the stack is admin and just creates the users in the same project as the stack?	15:47
tspatzier	shardy: I think so, at least I did not so any extra config in my environment.	15:48
shardy	tspatzier: Ok, the standalone use-case may actually mean we need to maintain the legacy non-domain paths	15:48
shardy	or come up with another way to authenticate signals from instances in standalone mode	15:48
shardy	unfortunately the cost of keeping things working in standalone mode has turned out to be very high :(	15:49
tspatzier	shardy: or find another mechanism for sharing software deployment metadata and deployment state in another way. I think this came up at the summit, but we never dived into it.	15:50
shardy	tspatzier: If you can get the instance sending the username/password, the auth_password middleware you're already using should work	15:50
*** nati_ueno has joined #heat		15:50
shardy	tspatzier: Yeah, just getting this problem solved for the integrated (non standalone) case is really hard, making it work for both cases is pretty difficult	15:51
shardy	tspatzier: I will think about it and try to find time to test some things though	15:51
tspatzier	shardy: ok, then I can try to get the heat transport running and see how far I get. Seems like there are a few things to be fixed os-* and the image element	15:51
shardy	tspatzier: Ok, sounds good, please let me know how it goes :)	15:52
tspatzier	shardy: will do. and if you have some thoughts, I am eager to hear them :)	15:52
tspatzier	shardy: on standalone in general: is this supposed to continue to be supported? sounds like it causes pain to keep everything in synch.	15:53
*** elynn_ has joined #heat		15:54
shardy	tspatzier: I don't know, probably a good topic for a meeting discussion, but I know some folks want to keep it	15:54
SpamapS	shardy: I think it is o-k to put limitations on standalone like "you have to use cfn for signals"	15:54
shardy	SpamapS: the issue in this case is cfn signals won't work, so we have to make the agents send the username/password and use the native transport (with auth_password middleware configured)	15:55
shardy	That should be a viable workaround for tspatzier, I think	15:55
shardy	pretty sure nobody has tried this yet though ;)	15:55
SpamapS	shardy: what if we made standalone users run a standalone keystone?	15:56
*** elynn_ has quit IRC		15:56
tspatzier	shardy: I'm trying to find a way this work and also can be supported going forward. If we come up with a solution that has restriction, i.e. we do not support all kinds of possible option for standalone, that is fine.	15:57
shardy	SpamapS: I guess, just sounds like yet more complexity tbh	15:57
*** rakesh_hs has joined #heat		15:57
SpamapS	shardy: also to be clear, Helion's public cloud (aka hpcloud.net) will have a Heat service at some point. We're just not sure when that is. Then I will basically say "standalone can die"	15:58
SpamapS	shardy: I'm happy to sign HP people up to do any hard work to keep standalone alive.	15:58
SpamapS	shardy: If you hadn't noticed, our engineers are starting to get their feet wet in the code, so they'll be around to help with things now. :)	15:58
shardy	SpamapS: sure, and I do understand some folks want standalone, but it's always been this kinda partially functional thing IMO, with lots of gotchas and stuff which doesn't work	15:59
shardy	but if folks are prepared to maintain all-the-workarounds, then fair enough :)	15:59
*** daneyon_ has joined #heat		15:59
shardy	SpamapS: yup, I've seen, good news :)	15:59
SpamapS	shardy: partially functioning for all heat use cases, but fully functioning for HP Rover's use case, is fine.	15:59
SpamapS	thus far they do not use updates or signalling	16:00
SpamapS	though I think there is interest in signals	16:00
tspatzier	SpamapS: we stumbled over signals when trying to do software config with standalone	16:00
shardy	I know I've resisted it for the integrated case, but the standalone case seems like an argument for the random-string based auth (managed by heat) which some folks wanted	16:01
*** arunrajan has quit IRC		16:01
shardy	Trouble is, if we do that it'll get used in non-standalone	16:01
*** daneyon has quit IRC		16:02
shardy	maybe everything should be bounced via swift in standalone mode?	16:02
shardy	e.g using the transport jasond is working on?	16:02
tspatzier	shardy: that's also something I thought about. have to think about it more though	16:03
shardy	tspatzier: Ok, I guess getting the user/password thing working is the first step, then we can think about alternative transports :)	16:03
tspatzier	shardy: yes, sounds like a plan.	16:04
*** arunrajan has joined #heat		16:04
tspatzier	SpamapS: I saw there are some patches in review to make heat the default transport in os-* . what is the way to get those changes when I build an image with diskimage-builder? Doesn't it pull released versions of os-* ?	16:05
*** killer_prince is now known as lazy_prince		16:05
*** gokrokve has joined #heat		16:06
*** daneyon_ has quit IRC		16:07
*** bgorski has quit IRC		16:10
*** tspatzier has quit IRC		16:10
*** blues-man has quit IRC		16:12
*** nkhare has quit IRC		16:16
*** IlyaE has quit IRC		16:16
*** jcoufal has joined #heat		16:20
*** alexheneveld has quit IRC		16:21
openstackgerrit	William C. Arnold proposed a change to openstack/heat: Implements stack lifecycle scheduler hints https://review.openstack.org/96889	16:22
*** Qiming has quit IRC		16:24
*** daneyon has joined #heat		16:27
*** noTHD has quit IRC		16:27
*** Qiming has joined #heat		16:28
KanagarajM	heat is having REST API for list_stacks and with_count->no.of stacks	16:30
*** david-lyle has joined #heat		16:32
KanagarajM	in the implementation, there are two rpc call made between api and engine service, this could be improved by introducing another rpc call list_stacks_with_metadata, where more details about stack list could be returned such as 'count'	16:32
KanagarajM	will this should be a blue print or a defect?	16:33
*** nati_ueno has quit IRC		16:33
*** lazy_prince is now known as killer_prince		16:34
SpamapS	KanagarajM: defect.	16:34
KanagarajM	ok thanks	16:34
*** lsmola__ has quit IRC		16:34
*** ifarkas has quit IRC		16:34
*** JAGDISH__ has quit IRC		16:36
*** JAGDISH__ has joined #heat		16:38
*** Qiming has quit IRC		16:42
*** julienvey has quit IRC		16:42
*** FrancoisBillant has quit IRC		16:43
KanagarajM	SpamapS: i have filed a bug for the same https://bugs.launchpad.net/heat/+bug/1336857 and i have mentioned the return value of the new method list_stacks_with_metadata., could you validate it once and kindly suggest me if we could provide additional details as well. thanks.	16:44
uvirtbot	Launchpad bug 1336857 in heat "improve list_stacks with count API implementation to use single rpc call" [Undecided,New]	16:44
*** arbylee has quit IRC		16:46
*** arbylee1 has joined #heat		16:46
openstackgerrit	Steven Hardy proposed a change to openstack/heat: heat_keystoneclient add get_user_token https://review.openstack.org/102885	16:46
openstackgerrit	Steven Hardy proposed a change to openstack/heat: stack user add _user_token https://review.openstack.org/102887	16:46
openstackgerrit	Steven Hardy proposed a change to openstack/heat: clients make heat_url public https://review.openstack.org/102886	16:46
openstackgerrit	Steven Hardy proposed a change to openstack/heat: Convert WaitConditionHandle to use handle_signal https://review.openstack.org/101351	16:46
openstackgerrit	Steven Hardy proposed a change to openstack/heat: Update waitcondition API to use signal RPC interface https://review.openstack.org/101352	16:46
openstackgerrit	Steven Hardy proposed a change to openstack/heat: Refactor waitcondition resources to allow easier subclassing https://review.openstack.org/101353	16:46
openstackgerrit	Steven Hardy proposed a change to openstack/heat: Add an OS::Heat::WaitCondition resource https://review.openstack.org/101354	16:46
openstackgerrit	Steven Hardy proposed a change to openstack/heat: Add native WaitConditionHandle resource https://review.openstack.org/102888	16:46
*** Qiming has joined #heat		16:48
*** zhiyan is now known as zhiyan_		16:50
*** daneyon has quit IRC		16:50
*** daneyon has joined #heat		16:51
*** zhiyan_ is now known as zhiyan		16:51
*** harlowja_away is now known as harlowja		16:52
*** arunrajan has quit IRC		16:55
*** IlyaE has joined #heat		16:55
*** Qiming has quit IRC		16:57
morganfainberg	SpamapS, ping re: https://review.openstack.org/#/c/99225/ i responded to your inline comment about the tests. trying to figure out if we can do that better, but i am not seeing a way.	16:59
morganfainberg	SpamapS, i'll plan on fixing the commit to note the new options/option rename today	17:00
*** Dev_Jin has joined #heat		17:03
*** stannie has quit IRC		17:05
*** arbylee1 has quit IRC		17:08
*** arbylee has joined #heat		17:08
*** gondoi is now known as zz_gondoi		17:08
*** zhiyan is now known as zhiyan_		17:08
*** bharath has joined #heat		17:09
bharath	Hey I am new to heat, May I know why can't invisible_to_admin user create a stack ?	17:10
*** derekh_ has quit IRC		17:11
*** zz_gondoi is now known as gondoi		17:11
SpamapS	morganfainberg: you can just split the tests into two classes, one that makes the call, and one that doesn't.	17:12
*** loki184 has joined #heat		17:13
morganfainberg	SpamapS, how does that work with inheriting the tests from the parent?	17:13
morganfainberg	SpamapS or are you saying we should not do that	17:13
SpamapS	morganfainberg: split the parent	17:13
morganfainberg	SpamapS, AH!	17:13
SpamapS	morganfainberg: note that this is _not_ a requirement.	17:13
morganfainberg	SpamapS, now that makes sense (forgive my lack of coffee [and shabu shabu] filled brain)	17:14
morganfainberg	SpamapS, nah, i like it.	17:14
SpamapS	morganfainberg: but less test code is less headache IMO :)	17:14
morganfainberg	SpamapS, it's a good approach and makes things a bit less headachy	17:14
SpamapS	nice	17:14
morganfainberg	SpamapS, ++++++++	17:14
morganfainberg	SpamapS, need to do lunch again	17:15
*** tango has quit IRC		17:16
*** gondoi is now known as zz_gondoi		17:16
*** IlyaE has quit IRC		17:17
*** SnowDust has joined #heat		17:20
*** arbylee has quit IRC		17:37
*** arbylee has joined #heat		17:37
*** DaveJ__ has quit IRC		17:38
*** sab has joined #heat		17:40
*** sab has quit IRC		17:44
*** arbylee has quit IRC		17:44
*** arbylee has joined #heat		17:45
*** Dev_Jin has quit IRC		17:45
*** openstackgerrit has quit IRC		17:49
*** openstackgerrit has joined #heat		17:50
*** denis_makogon has quit IRC		17:51
*** dmakogon_ has joined #heat		17:51
*** KanagarajM has quit IRC		17:52
*** bandarji has joined #heat		17:52
*** jstrachan has quit IRC		17:55
*** samuelmz_ is now known as samuelmz		17:55
*** saju_m has joined #heat		17:56
*** loki184 has quit IRC		17:57
*** rushiagr has quit IRC		17:57
*** bharath has quit IRC		17:58
*** daneyon_ has joined #heat		17:59
*** tomek_adamczewsk has joined #heat		17:59
samuelmz	Hi, I've a heat template in which I've a OS::Nova::Server and a OS::Heat::AutoScalingGroup. Both have a script into the user_data section. How can I configure the template to execute the AutoScalingGroup user_data after the Nova::Server one? Any thought on this?	18:00
*** tango has joined #heat		18:01
*** daneyon has quit IRC		18:01
*** jdandrea has quit IRC		18:02
*** nati_ueno has joined #heat		18:05
*** bandarji has quit IRC		18:05
*** bandarji has joined #heat		18:05
*** packet has quit IRC		18:06
*** arunrajan has joined #heat		18:09
*** jdandrea has joined #heat		18:09
*** IlyaE has joined #heat		18:10
*** bandarji has quit IRC		18:10
*** bandarji has joined #heat		18:10
*** bandarji has quit IRC		18:11
*** arbylee has quit IRC		18:11
*** shakamunyi has joined #heat		18:14
*** packet has joined #heat		18:19
*** arunrajan has left #heat		18:20
openstackgerrit	David J Hu proposed a change to openstack/python-heatclient: Add keystone v3 auth support https://review.openstack.org/92728	18:23
*** jdandrea has quit IRC		18:26
morganfainberg	SpamapS, trying to split this code up, there is a lot more change here than i thought when you recommended the split (test_heatclient)	18:27
morganfainberg	SpamapS, let me make sure it passes tests and i'll post it. let me know what you think.	18:28
*** kebray has quit IRC		18:30
*** jamiehannaford has quit IRC		18:31
*** kebray has joined #heat		18:31
*** jamiehannaford has joined #heat		18:32
BillArnold_	I see no agenda for a heat meeting today at 2000 UTC. Is there a meeting today? Want to suggest an item, isolation of openstack AZs in practice.	18:32
SpamapS	morganfainberg: cool	18:37
*** dims__ has quit IRC		18:37
*** arbylee has joined #heat		18:37
morganfainberg	stevebaker, shardy, I think heat should stop gating on sample config being up to date. if say oslo.messaging (or keystoneclient) changed options that you guys relied on all changes would get blocked/wedged until someone pushes an update sample config	18:38
morganfainberg	shardy, stevebaker, most other projects have droped that from pep8 for that very reason	18:38
SpamapS	BillArnold_: there should be one today, but perhaps zaneb just hasn't added it?	18:38
openstackgerrit	Morgan Fainberg proposed a change to openstack/heat: Support using domain_name instead of domain_id https://review.openstack.org/99225	18:38
*** arbylee has quit IRC		18:39
*** arbylee has joined #heat		18:39
morganfainberg	SpamapS, not sure i like that ^ as much as the previous iteration	18:39
zaneb	BillArnold_, SpamapS: my bad, added.	18:40
SpamapS	morganfainberg: no.. that did not turn out as awesome as I'd have expected.	18:43
BillArnold_	zaneb thanks. This derives from the discussions on the mailing list about autoscaling groups that Mike Spreitzer started yesterday.	18:43
zaneb	I figured :)	18:44
*** IlyaE has quit IRC		18:44
morganfainberg	SpamapS, yeah. it felt like a rabbit hole, but wanted to make sure i wasn't crazy before i said ... "yeah lets not do that"	18:44
*** nati_ueno has quit IRC		18:44
morganfainberg	SpamapS, I say we go back one patchset and just fix the commit message. any complaints?	18:45
*** nati_ueno has joined #heat		18:45
*** dims__ has joined #heat		18:47
*** nati_uen_ has joined #heat		18:53
*** nati_ueno has quit IRC		18:56
*** rbuilta has quit IRC		18:57
*** cj has joined #heat		18:57
*** kebray has quit IRC		18:58
*** harlowja is now known as harlowja_away		18:59
openstackgerrit	Morgan Fainberg proposed a change to openstack/heat: Support using domain_name instead of domain_id https://review.openstack.org/99225	19:00
openstackgerrit	Jason Dunsmore proposed a change to openstack/heat: Add utility functions for Swift TempURLs https://review.openstack.org/104302	19:02
*** harlowja_away is now known as harlowja		19:03
cj	hello. I'm trying to create a HOT template with the OS::Glance::Image resource type but Heat is reporting an error "Unknown resource type: OS::Glance::Image". I have the latest of the master DevStack branch for Heat. Any suggestions on how I can "register" this resource type? I was thinking it would be known by default	19:10
*** tomek_adamczewsk has quit IRC		19:15
*** tomek_adamczewsk has joined #heat		19:16
*** jamiehannaford has quit IRC		19:21
openstackgerrit	Jason Dunsmore proposed a change to openstack/heat: OS::Swift::Signal resource https://review.openstack.org/96947	19:22
*** IlyaE has joined #heat		19:23
*** dims__ has quit IRC		19:33
*** DandyPandy_ has joined #heat		19:35
*** DandyPandy has quit IRC		19:36
*** sorantis has joined #heat		19:37
*** cj has quit IRC		19:39
*** cj has joined #heat		19:39
*** vijendar has joined #heat		19:43
*** dims has joined #heat		19:44
*** SnowDust has quit IRC		19:46
*** dims_ has joined #heat		19:48
*** dims has quit IRC		19:49
*** shakamunyi has quit IRC		19:52
*** dims_ has quit IRC		19:54
*** dims has joined #heat		19:54
gpocentek	stevebaker: hi! Do you me if I take other https://review.openstack.org/#/c/95979/ to add the doc bit?	19:55
*** pas-ha has joined #heat		19:55
gpocentek	stevebaker: that is "Do you mind*"...	19:56
*** IlyaE has quit IRC		19:57
*** jdob has quit IRC		19:57
*** jdob has joined #heat		19:58
*** SnowDust has joined #heat		19:58
*** dims_ has joined #heat		19:58
zaneb	meeting time	20:00
*** tspatzier has joined #heat		20:00
stevebaker	gpocentek, its yours! thanks	20:00
*** dims has quit IRC		20:00
skraynev	stevebaker: new patches with stevedore kill me... reinstall heat for adding new plugin ;)	20:04
gpocentek	stevebaker: stevebaker thanks	20:04
*** IlyaE has joined #heat		20:05
*** dims_ has quit IRC		20:05
*** jmckind has joined #heat		20:05
*** jmckind has quit IRC		20:05
*** rakesh_hs has quit IRC		20:06
stevebaker	skraynev, but really, how often do you add a new plugin?	20:06
*** dims has joined #heat		20:06
*** shakamunyi has joined #heat		20:07
skraynev	stevebaker: yeah, rarely. I just get so big emotions after testing patches :)	20:09
*** jmckind has joined #heat		20:09
*** harlowja is now known as harlowja_away		20:09
skraynev	stevebaker: So now I prefer use latest instead of downloading one by one.	20:10
*** SnowDust has quit IRC		20:11
*** radez is now known as radez_g0n3		20:13
bodepd	how do you specify arbitrary events and block for events in userdata (like you can in cfn)?	20:16
bodepd	just a pointer to the docs would be greatly appreciated :) I'm sure I just missed it	20:17
*** cj has quit IRC		20:19
*** vijendar has quit IRC		20:20
openstackgerrit	Gauvain Pocentek proposed a change to openstack/heat: Makefile option to build xml docs https://review.openstack.org/95979	20:21
randallburt	skraynev: did you publish those comments for https://review.openstack.org/#/c/97975/16? I'm not seeing them.	20:23
randallburt	skraynev: or were you referring to stevebaker's comments?	20:23
skraynev	randallburt: second	20:23
openstackgerrit	Jason Dunsmore proposed a change to openstack/heat: OS::Swift::Signal resource https://review.openstack.org/96947	20:23
randallburt	skraynev: ah, ok. I'll remove −1 then. I just wanted to make sure I wasn't missing something.	20:23
skraynev	randallburt: no, you have not missed something :)	20:24
*** achampion has quit IRC		20:30
openstackgerrit	Morgan Fainberg proposed a change to openstack/heat: Support using domain_name instead of domain_id https://review.openstack.org/99225	20:30
*** alexpilotti has quit IRC		20:31
*** swygue has quit IRC		20:32
openstackgerrit	Ying Li proposed a change to openstack/heat: Move calculating new capacity to a function https://review.openstack.org/99470	20:33
*** aweiteka has quit IRC		20:35
*** harlowja_away is now known as harlowja		20:39
*** swygue has joined #heat		20:42
*** swygue has quit IRC		20:46
*** swygue has joined #heat		20:46
*** sorantis has quit IRC		20:48
*** zz_gondoi is now known as gondoi		20:50
*** tomek_adamczewsk has quit IRC		20:52
*** IlyaE has quit IRC		20:52
*** IlyaE has joined #heat		20:54
SpamapS	did I miss anything important in the meeting?	20:55
SpamapS	got stuck running errands forever :-P	20:55
*** dims has quit IRC		20:55
pas-ha	SpamapS, everybody review stevebaker series :)	20:55
SpamapS	yeah I need to review a lot	20:56
SpamapS	zaneb: is there a group rate for hotel rooms or anything for the meetup?	20:57
skraynev	SpamapS, and reminding about booking for mid-cycle meetup	20:57
*** tomek_adamczewsk has joined #heat		20:57
zaneb	SpamapS: no. experience from TripleO is that it is a giant PITA for little benefit	20:57
*** alexheneveld has joined #heat		20:58
SpamapS	zaneb: agreed	20:58
* skraynev go to sleep		20:58
SpamapS	zaneb: HP also has good standing rates with most hotel chains so sometimes the negotiated rate is worse. ;)	20:58
*** lurker5 has joined #heat		20:58
* SpamapS watches those well oiled corporate muscles gleam in the sun		20:59
zaneb	SpamapS: yeah, the group rate is worse than the RH rate for those two hotels	20:59
SpamapS	ever marvelled at it's BEAUTY	20:59
*** dims has joined #heat		21:00
zaneb	if you call them up and ask for the Red Hat rate, you will quite possibly get it	21:00
ekarlso	btw, I know I missed the meeting. but is there plans to get partial updates to stacks during Junp ?	21:02
ekarlso	currently one needs to resubmit the whole thing which is tedious when u want to say swap out flavors or images	21:03
ekarlso	for upgrades / capacity increase	21:03
stevebaker	ekarlso: that could be implemented entirely in python-heatclient	21:04
ekarlso	stevebaker: rly ?	21:05
ekarlso	far as I understood heat api is missing PATCH support	21:05
stevebaker	ekarlso: heatclient could fetch the previous template if it is not supplied on the cli	21:05
SpamapS	ekarlso: heatclient could just fetch the current template and submit the changes	21:05
SpamapS	:)	21:05
ekarlso	SpamapS: suuure	21:06
ekarlso	but the heat api requires u to also pass all params no ?	21:06
ekarlso	..	21:06
SpamapS	yes, and fetching the stack shows them	21:06
*** gokrokve has quit IRC		21:06
randallburt	ekarlso: a pet peeve of mine as well, but again heatclient could sort it.	21:06
SpamapS	no doubt PATCH would be more efficient	21:06
SpamapS	and should be implemented	21:07
ekarlso	SpamapS: So, how would you then do if it's a password or secret ? ;P	21:07
SpamapS	but .. bigger fish to fry	21:07
ekarlso	it's tedious... :p	21:07
randallburt	SpamapS: agreed, plus hidden params break the client solution	21:07
randallburt	ekarlso: jinx :)	21:07
ekarlso	we're trying to upgrade for example a database instance, that requires us to pass in the whole params list each time which could have been as simple as passing in a new imasge id	21:08
*** gokrokve has joined #heat		21:08
randallburt	ekarlso: do you know if there's an existing bp for this? IIRC, Tim Schnell may have submitted one a while back.	21:09
SpamapS	ekarlso: nobody thinks it is a bad idea.	21:09
SpamapS	ekarlso: but AFAIK, nobody is driving such work ATM	21:09
ekarlso	randallburt: there's a bug at least I think that pas-ha linked me once	21:09
ekarlso	SpamapS: I just asked if it was something for J	21:09
SpamapS	yeah the bug is at least a year old	21:09
randallburt	ekarlso: k	21:09
SpamapS	ekarlso: it could be... if it had developers assigned. :)	21:10
ekarlso	I think it would make life easier for #tripleo stuff as well :p	21:10
SpamapS	https://bugs.launchpad.net/heat/+bug/1224828	21:10
uvirtbot	Launchpad bug 1224828 in tuskar "heat stack-update wants all parameters and the template all over again" [Medium,Triaged]	21:10
SpamapS	ekarlso: agreed. TripleO's Heat resources are focused on convergence.	21:10
zaneb	ekarlso: there is a plan to fix the part about having to resubmit params. tango is working on it	21:10
SpamapS	because zomg if we don't start now	21:11
*** saju_m has quit IRC		21:11
ekarlso	zaneb: PATCH support or ?	21:11
SpamapS	zaneb: tango should take ownership of that bug then	21:11
zaneb	ekarlso: not PATCH support, no	21:11
ekarlso	zaneb: boo	21:11
zaneb	https://blueprints.launchpad.net/heat/+spec/troubleshooting-low-level-control	21:11
ekarlso	SpamapS: convergence vs ?	21:11
SpamapS	ekarlso: vs. workflow	21:12
zaneb	SpamapS: I'll link that bug to the blueprint	21:12
SpamapS	ekarlso: https://review.openstack.org/#/c/95907/	21:12
zaneb	oh, it is linked already	21:12
tango	ekarlso: please take a look at the bp to see if it does what you need, or add comment to the whiteboard	21:12
SpamapS	zaneb: troubleshooting-low-level-control is a really broad topic.. :P	21:12
* SpamapS facepalms at Launchpad putting linked blueprints below subscribed parties		21:13
*** tspatzier has quit IRC		21:13
tango	SpamapS: Yeah, agree, I should have come up with a more precise name	21:13
ekarlso	still dont see how that makes it more convenient to say update a image or a password even and not requiring the whole param mess	21:13
zaneb	SpamapS: yeah, it was originally something else altogether. I pretty much hijacked that blueprint to make it this	21:13
*** dims_ has joined #heat		21:13
ekarlso	if u have 30+ parameters it's a bit pain having to reconstruct it for update	21:14
SpamapS	ekarlso: environment files FTW	21:14
ekarlso	SpamapS: rly ?	21:14
zaneb	ekarlso: the point of that bug/bp is that you won't have to pass the parameters again	21:14
randallburt	and while we're on the subject, lets also be able to omit the template as well.	21:14
randallburt	SpamapS: boo.	21:14
SpamapS	randallburt: #lean	21:14
*** IlyaE has quit IRC		21:14
zaneb	randallburt: why? you can always get the template from the API	21:14
SpamapS	do what you can today to enable a use case today. :)	21:15
randallburt	zaneb: because lazy	21:15
*** dims has quit IRC		21:15
SpamapS	ok, I think its time I went into the cone of silence	21:15
randallburt	lol	21:15
ekarlso	heat stack-update mystack -P 'password=meh' ...	21:15
SpamapS	ekarlso: heat stack-update mystack -e all-my-password-params.json ;)	21:15
zaneb	like stevebaker said, python-heatclient can handle that	21:15
ekarlso	sure env files are ok, but if you're doing this for other services that use heat then u don't store env files all around.	21:15
SpamapS	ekarlso: We agree. No argument is necessary. _WORK_ is necessary.	21:16
SpamapS	and the PATCH method in particular will be more useful when convergence happens because you can just patch individual params and the template independent of one another if you don't care about atomicity	21:16
SpamapS	and thus, I go to the cone of silence to focus	21:17
ekarlso	SpamapS: what's teh timeframe for that ? ;P	21:17
ekarlso	SpamapS: also for a newbie at this, convergence vs ?	21:17
*** lurker5 has quit IRC		21:18
randallburt	ekarlso: too late. cone of silence.	21:18
ekarlso	hah	21:18
*** dims_ has quit IRC		21:18
*** IlyaE has joined #heat		21:18
randallburt	ekarlso: and atm, there's no real timeframe for that other than starting the convergence work now and having "something" around it for Juno. That doesn't mean PATCH in Juno. Beyond that, its kinda up to the community and folks willing to work on it.	21:19
*** pafuent has left #heat		21:19
*** tomek_adamczewsk has quit IRC		21:19
ekarlso	randallburt: what's convergence btw?	21:20
randallburt	ekarlso: your best bet atm is to look at the relevant specs and bps and weigh in on those. Your bestest bet is to submit your own specs and patches.	21:20
randallburt	ekarlso: 1 sec, will link the spec.	21:20
ekarlso	randallburt: u mean the low level thing tango / zaneb talked about ?	21:21
*** tomek_adamczewsk has joined #heat		21:21
randallburt	ekarlso: https://review.openstack.org/#/c/95907/	21:21
randallburt	ekarlso: also, we had some discussions at the summit and on the ML that would help clarify, but don't have the links for those ATM.	21:21
ekarlso	tango: it doesn't seem to be the same as doing a PATCH towards updating a stack though +	21:24
*** sgordon` has quit IRC		21:27
tango	ekarlso: can you point me to some info on PATCH?	21:27
*** sgordon has quit IRC		21:27
*** DandyPandy_ has quit IRC		21:28
*** sgordon` has joined #heat		21:28
ekarlso	tango: it's the HTTP Verb used when doing a partial update, that's what u do when u stack-update no ?	21:28
randallburt	ekarlso: stack update is PUT IIRC.	21:29
*** IlyaE has quit IRC		21:29
ekarlso	randallburt: even when you're updating 1 param ?	21:29
openstackgerrit	Anderson Mesquita proposed a change to openstack/heat: Add resource attrs and properties to stack-preview https://review.openstack.org/94479	21:29
tango	ekarlso: I guess you mean using the concept of PATCH in http for stack-update parameters?	21:30
ekarlso	tango: yeah	21:31
*** jdob has quit IRC		21:31
tango	ekarlso: so you don't have to list all the parameters again, change only a few and let the rest be the same	21:31
ekarlso	it's silly to pass all the information if u just want to update 1 passowrd...	21:31
*** dims has joined #heat		21:31
ekarlso	adds complexity to things using it ..	21:31
ekarlso	if I knew where to look I would allmost be tempted to try to add support for it :p	21:32
*** daneyon_ has quit IRC		21:32
*** daneyon has joined #heat		21:32
tango	ekarlso: That's what I have in mind for the bp. I know the bp sounds different but the idea is to make it easy to reuse info currently in the DB	21:33
*** IlyaE has joined #heat		21:33
ekarlso	tango: if u have a service that's using heat to setupresources aka trove and u say want to upgrade the cluster instances aka changing flavor it would be uber easier to just pass in instanceX_flavor = 'm1.large'	21:34
ekarlso	vs having to first get existing, ensure all params are really the correct type (if not it breaks when calling the stacks.update(), populating passwords again and so on ..	21:34
ekarlso	oh and also sometimes some params are implicit so u need to filter those too	21:35
*** tomek_adamczewsk has quit IRC		21:36
tango	ekarlso: I can update the bp to add this use case.	21:37
ekarlso	tango: it would be awesome if it coudl incorporate partial updates like supporting just updating 1 parameter	21:38
ekarlso	or even updating the template w/o parameters using existing parameters	21:38
tango	ekarlso: It's a little further generalization from what Zane suggested, but it seems useful.	21:38
openstackgerrit	Anderson Mesquita proposed a change to openstack/heat: Add resource attrs and properties to stack-preview https://review.openstack.org/94479	21:39
ekarlso	tango: it would make services using heat and probably others as well easier ;p	21:40
tango	ekarlso: right, that case is currently covered. How would you suggest the CLI option for patching just one parameter, to avoid ambiguity?	21:41
*** tomek_adamczewsk has joined #heat		21:41
tango	ekarlso: we can't use ---parameters again since it would be ambiguous	21:43
ekarlso	tango: good q ;)	21:43
ekarlso	tango: btw, a PATCH can be more then 1 parameter, it implies a partial update..	21:43
tango	ekarlso: right, I would expect a partial list	21:44
stevebaker	lifeless: btw, by making this comment I'm not fobbing off your request https://bugs.launchpad.net/heat/+bug/1336656/comments/1	21:44
uvirtbot	Launchpad bug 1336656 in heat "Can't access subnet attributes from a port resource" [Medium,Triaged]	21:44
*** jmckind has quit IRC		21:44
ekarlso	tango: it could be u would update f ex flavor and image of a instance	21:44
ekarlso	or multiple passwords	21:45
ekarlso	or users	21:45
ekarlso	anywho's tango it would super awesome if it was added, I gotta go hit the sack :)	21:46
tango	ekarlso: ok, will do	21:46
ekarlso	tango: i'll owe u a beer / dinner @ paris if u do ;p	21:46
tango	ekarlso: hehe, will code for beer :)	21:47
*** lindsayk1 has joined #heat		21:48
stevebaker	ekarlso: btw there is this https://bugs.launchpad.net/heat/+bug/1224828	21:49
uvirtbot	Launchpad bug 1224828 in tuskar "heat stack-update wants all parameters and the template all over again" [Medium,Triaged]	21:49
stevebaker	oh, thats already been pasted	21:49
ekarlso	stevebaker: that was what I meant earlier.	21:49
ekarlso	:)	21:49
randallburt	tango/ekarlso FWIW, unless this is implemented in the API and not just in the client, you only solve the problem for folks using python-heatclient. Some folks use other language bindings or the ReST api directly.	21:51
stevebaker	randallburt: yes, PATCH would be preferred	21:52
ekarlso	randallburt: I was hoping for API side as well :)	21:52
ekarlso	but nighty folks (:	21:53
ekarlso	too late here :\|	21:53
*** kebray has joined #heat		21:55
tango	randallburt: agree, I am doing both client and API	21:56
randallburt	tango: cool, thanks!	21:56
*** arbylee has quit IRC		21:56
*** BillArnold_ has quit IRC		21:59
*** JAGDISH__ has quit IRC		21:59
*** arbylee has joined #heat		22:01
*** lindsayk1 has quit IRC		22:03
*** ccrouch has quit IRC		22:04
*** maxskew has quit IRC		22:05
*** pas-ha has quit IRC		22:09
*** kebray has quit IRC		22:10
*** kebray has joined #heat		22:11
*** lindsayk1 has joined #heat		22:14
*** pas-ha has joined #heat		22:15
*** shakamunyi has quit IRC		22:17
*** ccrouch has joined #heat		22:17
*** ericgoncz has quit IRC		22:19
*** tomek_adamczewsk has quit IRC		22:19
*** shakamunyi has joined #heat		22:20
*** gondoi is now known as zz_gondoi		22:24
*** DandyPandy has joined #heat		22:25
*** DandyPandy has quit IRC		22:25
*** DandyPandy has joined #heat		22:26
*** pas-ha has quit IRC		22:26
*** sorantis has joined #heat		22:26
*** zhangyang has quit IRC		22:27
*** pas-ha has joined #heat		22:28
*** zhangyang has joined #heat		22:28
*** pas-ha has quit IRC		22:29
openstackgerrit	A change was merged to openstack/heat: Guaranteed password complexity using OS::Heat::RandomString https://review.openstack.org/101634	22:33
*** lindsayk1 has quit IRC		22:38
*** lindsayk1 has joined #heat		22:40
*** lindsayk1 has joined #heat		22:41
*** DandyPandy has quit IRC		22:52
*** david-lyle has quit IRC		22:53
tango	zaneb: hi Zane	22:53
zaneb	hey	22:53
openstackgerrit	A change was merged to openstack/heat-specs: Add specification to implement EC2 instance bdm https://review.openstack.org/102139	22:53
tango	I ran into an error scenario with stack-update which looks like a bug	22:54
*** david-lyle has joined #heat		22:54
tango	want to run it by you to see if I am doing something weird	22:54
tango	So I deploy a template win 1 resource (a server)	22:54
tango	then I run stack-update to add a second resource with a parameter. The update fails, and in the DB, the stack entry shows the template reverting to the previous version with 1 resource	22:56
tango	but the parameters column contains the parameters from the failed stack-update	22:56
openstackgerrit	A change was merged to openstack/heat-specs: Specification for filtering stacks by status for cfn API https://review.openstack.org/101140	22:57
tango	afterward, any CLI on this stack fails with the message: ERROR: The Parameter (mycustomprop_one) was not defined in template.	22:57
tango	It fails on validating the parameters, because the previous template is used, which does not have the new resource with this parameter	22:58
tango	Even stack-delete would not work	22:58
tango	You have this fix pending: https://review.openstack.org/#/c/101288/	22:59
tango	I wonder if it's related	22:59
*** david-lyle has quit IRC		22:59
*** tomek_adamczewsk has joined #heat		22:59
*** blamar_ has joined #heat		23:01
*** blamar has quit IRC		23:01
*** blamar_ is now known as blamar		23:01
*** arbylee has quit IRC		23:01
*** blamar has quit IRC		23:04
*** tomek_adamczewsk has quit IRC		23:05
*** swygue has quit IRC		23:11
*** shakamunyi has quit IRC		23:11
*** arbylee has joined #heat		23:19
*** JAGDISH__ has joined #heat		23:24
*** ccrouch has quit IRC		23:26
*** sabeen2 has quit IRC		23:26
*** swygue has joined #heat		23:28
*** m_22 has joined #heat		23:37
*** jcoufal has quit IRC		23:48
*** nati_uen_ has quit IRC		23:49
*** arbylee has quit IRC		23:51
*** swygue has quit IRC		23:52
lifeless	stevebaker: didn't think you were for a second	23:53
openstackgerrit	OpenStack Proposal Bot proposed a change to openstack/heat: Updated from global requirements https://review.openstack.org/96507	23:54
*** sorantis has quit IRC		23:55

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!