Monday, 2014-03-03

« Sunday, 2014-03-02 Index Tuesday, 2014-03-04 »
*** gokrokve has joined #heat00:00
*** e0ne has quit IRC00:04
openstackgerritSteven Dake proposed a change to openstack/heat: Migrate to oslo.messaging from oslo-incubator rpc  https://review.openstack.org/7279800:20
openstackgerritSteven Dake proposed a change to openstack/heat: Always report the action in state_reason as engine encodes it  https://review.openstack.org/7740600:20
*** cfriesen_ has joined #heat00:40
*** cfriesen_ has quit IRC00:44
*** zns has quit IRC00:51
*** arbylee has joined #heat00:56
*** e0ne has joined #heat00:59
*** e0ne has quit IRC01:04
stevebakerAny reviewers for this? I've bumped it down to medium so it won't be getting any extensions https://review.openstack.org/#/c/72919/01:18
*** nosnos has joined #heat01:28
openstackgerritLee Li proposed a change to openstack/python-heatclient: Remove dependent module py3kcompat  https://review.openstack.org/7355801:29
openstackgerritSteven Dake proposed a change to openstack/heat: Migrate to oslo.messaging from oslo-incubator rpc  https://review.openstack.org/7279801:32
*** Tross has joined #heat01:32
SpamapSstevebaker: \o/ thanks!01:39
*** gokrokve has quit IRC01:39
*** zns has joined #heat01:40
openstackgerritMitsuru Kanabuchi proposed a change to openstack/heat: Implement OS::Neutron::ExtraRoute as /contrib  https://review.openstack.org/7489901:41
*** gokrokve has joined #heat01:41
*** KiwiMartin has joined #heat01:45
*** KiwiMartin has quit IRC01:46
*** rpothier has joined #heat01:55
*** e0ne has joined #heat01:58
openstackgerritLee Li proposed a change to openstack/python-heatclient: Using common methods from oslo cliutils  https://review.openstack.org/6712001:58
openstackgerritSteven Dake proposed a change to openstack/heat: Update config generator for oslo.messaging  https://review.openstack.org/7747201:59
openstackgerritSteven Dake proposed a change to openstack/heat: Migrate to oslo.messaging from oslo-incubator rpc  https://review.openstack.org/7279801:59
*** e0ne has quit IRC02:03
openstackgerritSteven Dake proposed a change to openstack/heat: Migrate to oslo.messaging from oslo-incubator rpc  https://review.openstack.org/7279802:07
openstackgerritSteven Dake proposed a change to openstack/heat: Update config generator for oslo.messaging  https://review.openstack.org/7747202:07
*** guohliu has joined #heat02:14
*** erkules has quit IRC02:19
*** ramishra_ has joined #heat02:27
*** erkules has joined #heat02:32
*** ramishra_ has quit IRC02:38
*** ramishra has joined #heat02:39
*** e0ne has joined #heat02:58
*** Tross has quit IRC02:59
*** e0ne has quit IRC03:03
*** jprovazn has joined #heat03:11
*** gokrokve has quit IRC03:13
*** gokrokve has joined #heat03:13
*** liang has joined #heat03:14
*** gokrokve has quit IRC03:17
*** arbylee has quit IRC03:22
openstackgerritJenkins proposed a change to openstack/heat: Updated from global requirements  https://review.openstack.org/7668903:28
*** ramishra_ has joined #heat03:29
*** sergmelikyan has quit IRC03:31
openstackgerritSteven Dake proposed a change to openstack/heat: Migrate to oslo.messaging from oslo-incubator rpc  https://review.openstack.org/7279803:31
*** sergmelikyan has joined #heat03:31
*** ramishra has quit IRC03:32
*** bubba_ has joined #heat03:32
openstackgerritSteven Dake proposed a change to openstack/heat: Migrate to oslo.messaging from oslo-incubator rpc  https://review.openstack.org/7279803:33
openstackgerritSteven Dake proposed a change to openstack/heat: Update config generator for oslo.messaging  https://review.openstack.org/7747203:33
openstackgerritJenkins proposed a change to openstack/python-heatclient: Updated from global requirements  https://review.openstack.org/7669803:34
*** Tross has joined #heat03:34
*** nszceta has quit IRC03:51
*** ramishra_ has quit IRC03:56
*** e0ne has joined #heat03:58
*** e0ne has quit IRC04:03
*** kfox1111 has joined #heat04:14
openstackgerrithuangtianhua proposed a change to openstack/heat: Ensure property 'name' can be updated  https://review.openstack.org/7462404:16
*** akuznetsov has quit IRC04:18
*** akuznetsov has joined #heat04:19
*** kfox1111 has quit IRC04:25
*** bvandenh has joined #heat04:27
openstackgerritA change was merged to openstack/heat: Implement glance image constraint  https://review.openstack.org/7291904:27
*** nosnos has quit IRC04:28
*** nosnos has joined #heat04:34
*** cmyster has joined #heat04:56
*** cmyster has joined #heat04:56
*** Tross has quit IRC04:58
*** e0ne has joined #heat04:58
*** bvandenh has quit IRC04:58
*** e0ne has quit IRC05:03
*** zns has quit IRC05:03
*** jprovazn has quit IRC05:08
*** skraynev_afk is now known as skraynev05:09
*** Tross has joined #heat05:09
skraynevmorning05:09
cmystermorning05:09
* cmyster wonders...05:11
cmysterI am trying to make this tempest work. I guess I should keep using cirrus and not something fancy05:11
*** ramishra has joined #heat05:26
*** rcleere has quit IRC05:26
openstackgerritA change was merged to openstack/heat: Adding "device_owner" property for Port resource  https://review.openstack.org/6965405:27
*** zaneb has quit IRC05:30
*** daneyon has joined #heat05:33
*** nkhare has joined #heat05:36
*** gokrokve has joined #heat05:55
*** e0ne has joined #heat05:58
*** zaneb has joined #heat06:01
*** nosnos has quit IRC06:01
*** e0ne has quit IRC06:03
*** nosnos has joined #heat06:06
openstackgerritJenkins proposed a change to openstack/heat: Imported Translations from Transifex  https://review.openstack.org/7256606:10
*** ramishra has quit IRC06:19
*** daneyon has quit IRC06:25
openstackgerrithuangtianhua proposed a change to openstack/heat: Ensure property 'name' can be updated  https://review.openstack.org/7462406:27
*** lsmola has joined #heat06:28
*** saju_m has joined #heat06:31
*** e0ne has joined #heat06:35
openstackgerritSteven Dake proposed a change to openstack/heat: Migrate to oslo.messaging from oslo-incubator rpc  https://review.openstack.org/7279806:37
*** nkhare has quit IRC06:38
openstackgerritSteven Dake proposed a change to openstack/heat: Update config generator for oslo.messaging  https://review.openstack.org/7747206:38
*** nosnos has quit IRC06:40
*** topol has quit IRC06:45
*** chandankumar_ has quit IRC06:45
*** amritanshu_RnD has joined #heat06:51
*** chandan_kumar has joined #heat06:51
*** amritanshu_RnD is now known as Guest2321306:52
*** nkhare has joined #heat06:54
*** chandan_kumar has quit IRC06:55
*** nosnos has joined #heat07:04
*** tomek_adamczewsk has joined #heat07:06
*** e0ne has quit IRC07:09
*** TonyBurn_ has joined #heat07:10
*** TonyBurn__ has joined #heat07:19
*** TonyBurn_ has quit IRC07:23
*** TonyBurn__ is now known as TonyBurn_07:26
*** ramishra has joined #heat07:30
*** Guest23213 has quit IRC07:31
*** amritanshu__RnD has joined #heat07:31
*** skraynev has quit IRC07:36
*** nkhare has quit IRC07:37
*** skraynev has joined #heat07:37
*** skraynev has quit IRC07:41
*** skraynev has joined #heat07:42
*** ramishra has quit IRC07:45
*** gokrokve has quit IRC07:45
*** nkhare has joined #heat07:50
*** tomek_adamczewsk has quit IRC07:52
*** tspatzier has joined #heat07:53
*** pasquier-s has joined #heat08:04
therveGood morning08:05
skraynevhello therve08:06
*** e0ne has joined #heat08:06
cmystermorning08:09
*** e0ne has quit IRC08:13
*** gokrokve has joined #heat08:16
*** yogesh has joined #heat08:18
openstackgerritSteven Dake proposed a change to openstack/heat: Migrate to oslo.messaging from oslo-incubator rpc  https://review.openstack.org/7279808:27
*** giulivo has joined #heat08:27
*** amritanshu__RnD has quit IRC08:28
*** therve_ has joined #heat08:33
*** amritanshu__RnD has joined #heat08:36
*** gokrokve has quit IRC08:38
*** saju_m has quit IRC08:40
*** saju_m has joined #heat08:43
*** ifarkas has joined #heat08:44
*** therve_ has quit IRC08:45
openstackgerritSteven Dake proposed a change to openstack/heat: Migrate to oslo.messaging from oslo-incubator rpc  https://review.openstack.org/7279808:45
*** jistr has joined #heat08:47
*** alexheneveld has joined #heat08:54
*** e0ne has joined #heat08:55
*** saju_m has quit IRC09:02
*** chandan_kumar has joined #heat09:03
*** tomek_adamczewsk has joined #heat09:04
*** CGJamie has joined #heat09:15
openstackgerritThomas Herve proposed a change to openstack/heat: Native ScalingPolicy resource  https://review.openstack.org/7752909:15
*** gokrokve has joined #heat09:16
*** alexheneveld has quit IRC09:20
*** mkollaro has joined #heat09:20
*** gokrokve has quit IRC09:21
*** che-arne has joined #heat09:22
*** che-arne has quit IRC09:24
*** coolsvap has joined #heat09:30
*** dmakogon__ is now known as denis_makogon09:31
*** saju_m has joined #heat09:34
*** CGJamie has quit IRC09:46
*** e0ne_ has joined #heat09:47
*** julienvey1 is now known as julienvey09:49
*** jamieh has joined #heat09:50
*** e0ne has quit IRC09:50
*** ramishra has joined #heat09:53
openstackgerritA change was merged to openstack/heat: Update Oslo wiki link in README  https://review.openstack.org/7715309:55
*** jamieh_ has joined #heat10:03
*** jamieh has quit IRC10:05
*** tomek_adamczewsk has quit IRC10:10
*** tomek_adamczewsk has joined #heat10:11
*** tomek_adamczewsk has quit IRC10:11
openstackgerritZhang Yang proposed a change to openstack/heat: Replace assertEqual(None, *) with assertIsNone  https://review.openstack.org/7755010:13
*** mkollaro has quit IRC10:14
*** gokrokve has joined #heat10:17
*** jamieh_ has quit IRC10:19
*** yogesh has quit IRC10:20
*** yogesh has joined #heat10:20
*** DaveJ__ has joined #heat10:20
*** amritanshu__RnD has quit IRC10:21
*** mkollaro has joined #heat10:21
*** amritanshu_RnD has joined #heat10:21
*** gokrokve has quit IRC10:22
*** jamieh has joined #heat10:22
*** amritanshu_RnD is now known as Guest6052710:22
DaveJ__Hi guys, still having a few issues with creating alarms / meters.  I used the sample https://raw.github.com/openstack/heat-templates/master/cfn/F17/AutoScalingCeilometer.yaml but my alarm shows insufficent data.  Looking at the template the group adds the metadata 'Key: metering.server_group, Value: WebServerGroup'  But the meter matches on10:22
DaveJ__'{'metadata.user_metadata.groupname': {Ref: 'WebServerGroup'}}'.10:22
DaveJ__Should 'groupname' be 'server_group' ?10:22
*** yogesh has quit IRC10:24
*** saju_m has quit IRC10:25
*** jamieh has quit IRC10:25
therveDaveJ__, You should be able to remove this, the code adds the tag by default10:27
DaveJ__therve:  Both the tag and the matching_metadata ?10:27
therveNo just the tag10:28
therveDaveJ__, Do you have data at all?10:28
DaveJ__yeah ceilometer  sample-list -m cpu_util -q resource_id=<my resource> shows samples.10:29
DaveJ__but ceilometer alarm-list reports there is insufficent data for the alarm.10:29
therveAnd if you do q=metadata.user_metadata.groupname == XXX ?10:29
*** jamieh has joined #heat10:31
DaveJ__hmm - that's not returning anything, but I had  changed the values of the tag + matching_metadata.  Let me redeploy and check that10:31
DaveJ__matching_metadata should be {'metadata.user_metadata.groupname': {Ref: 'WebServerGroup'}} ?10:31
openstackgerritAndrey Kurilin proposed a change to openstack/heat: Sync latest oslo.db code  https://review.openstack.org/7653910:36
therveDaveJ__, If you replace the Ref by the real value, yeah10:37
cmysterguys, what's the credentials for the cirros-0.3.1-x86_64-uec image? I'm trying to write a test to start a few stacks and connect to them (IIRC there is an sshd on those)10:39
therveThere are none by default10:39
cmysterok,10:39
therveYou can specify an SSH key when creating the server10:39
shardycmyster: Heat requires an image with cloud-init, so you should be able to log in with an ssh key10:39
cmysterso I need to use F20 probably,10:40
therveI'd be surprised if it doesn't work with the cirros image10:40
cmysterwhat, just ssh -i to it ?10:41
*** liang has quit IRC10:41
cmysterhmmm, sec lemme check for its IP10:41
cmysteroh, seems like in issue with Horizon, it won't show the stack.10:41
*** saju_m has joined #heat10:42
*** e0ne has joined #heat10:45
*** david-lyle has quit IRC10:45
*** coolsvap has quit IRC10:47
*** e0ne_ has quit IRC10:49
openstackgerritZhang Yang proposed a change to openstack/heat: Set statuscode=404 in NotFoundException in tests  https://review.openstack.org/7756511:01
openstackgerritPavlo Shchelokovskyy proposed a change to openstack/heat: Order imports in alphabetical order (6/9)  https://review.openstack.org/7151611:04
*** Guest60527 has quit IRC11:05
*** alexheneveld has joined #heat11:08
*** saju_m has quit IRC11:11
shardytherve: Hey, question re https://review.openstack.org/#/c/77529/ when you have a moment11:11
*** gokrokve has joined #heat11:17
openstackgerritA change was merged to openstack/heat: Add validation of nested property schema  https://review.openstack.org/7551211:17
sgranhello11:18
*** alexpilotti has joined #heat11:18
sgranI'm thinking of working on the AWS::EC2::SecurityGroup Resource to add support for security group rules as separate resources11:18
sgranwould something like this be welcome?11:18
sgranif done correctly, this allows two things I'd like to see - it would allow you to have rules that reference the parent security group (all db servers in the secgroup DbServers can talk to each other on DBPort, say)11:19
sgranand it would allow updating the list of rules for a security group without deleting and recreating it11:19
*** nosnos has quit IRC11:21
*** gokrokve has quit IRC11:21
shardysgran: You mean implement AWS::EC2::SecurityGroupIngress/AWS::EC2::SecurityGroupEgress?11:22
sgranyes11:22
*** alexpilotti has quit IRC11:23
sgranI can see one difficulty - AWS::EC2::SecurityGroupEgress will have to fail if the network client is nova, but I'm sure that can be made to produce a reasonable error11:23
shardysgran: sure, that sounds like a worthwhile addition, although for folks who don't care about AWS portability, we're reccomending using the native neutron resources where possible11:23
sgransure11:23
shardyI wonder if it might be possible to implement as a provider resource template in /etc/heat/templates?11:24
sgranthis will probably be useful for people still on nova network, though, right?11:24
sgrannot sure - I would have thought you'd need to modify the existing security group code as well?11:24
sgranand I've never done one, although that might be interesting to learn how to do :)11:25
*** saju_m has joined #heat11:25
openstackgerritPavlo Shchelokovskyy proposed a change to openstack/heat: Order imports in alphabetical order (7/9)  https://review.openstack.org/7151711:25
shardysgran: I'm not sure either, but just FYI I think we're likely to gradually move towards making AWS compatible resources provider templates where possible, and keeping the python plugins as the native resource implementations11:26
shardyThat process has hardly started though so feel free to just implment a plugin if it's easier, just something to think about :)11:26
sgranfair enough :)11:38
openstackgerritPavlo Shchelokovskyy proposed a change to openstack/heat: Order imports in alphabetical order (8/9)  https://review.openstack.org/7151811:39
openstackgerritA change was merged to openstack/heat: Document schema properties for Neutron router resources  https://review.openstack.org/7666511:40
sgranthe current security group iterates the list of ingress/egress rules and makes rule directly now11:43
sgranin updating to allow for individual rules, I'd need it to instead create new objects11:43
sgranis there an existing example of doing something like that?11:43
shardysgran: don't the rule resources just have to be a container for the data, where Ref: <RuleResource> returns a json dict?11:45
shardyI don't see why SecurityGroup has to change at all11:46
sgranIf they're not resources, you can't have a rule in a group referencing the group11:48
sgranor am I not understanding you?11:48
shardysgran: Oh wait, the interface is backwards from what I thought11:49
shardyyuk11:49
shardysgran: Do you have a use-case for when this is needed?11:50
sgranyes.  I want to spin up a group of postgres servers.  Part of the replication catch-up process is to ssh into the master and copy some write ahead logs.  Normal replication needs access to the database port (typically 5432).  I'd like to be able to create a security group for my postgress servers that allows the db port to be connected to from the other db servers, and the api servers.  In addition I'd like a rule that allows 22 only from the db servers11:52
sgranso this would be 3 rules in the security group, and 2 of them would be self-referential11:52
shardysgran: Ok, perhaps you could raise a blueprint, and include a template snippet example (perhaps in a linked wiki page under https://wiki.openstack.org/wiki/Heat/Blueprints)11:54
sgranbut there are other use cases - tomcat session replication would reference the tomcat server group, and so on11:54
openstackgerritA change was merged to openstack/heat: Add admin_pass as a property  https://review.openstack.org/7725311:54
sgranok, I'll start on it11:54
*** tomek_adamczewsk has joined #heat11:58
openstackgerritA change was merged to openstack/heat: Change software_config storage of config and io columns  https://review.openstack.org/7706112:00
*** jprovazn has joined #heat12:03
therveshardy, Yes?12:04
shardytherve: Hi, I had a query about the signal interface to your native ScalingPolicy resource12:05
shardyso you expect that to only work with native signals?12:06
therveNot really12:06
therveIt should fairly agnostic AFAIU12:06
*** jamieh has quit IRC12:07
shardytherve: what I mean, is do we definitely want to override FnGetRefId, as that won't work with ceilometer will it?12:07
therveAh, I see12:07
shardyI really want us to work out the non ec2 way to do that, but atm I think ceilometer just expects to hit the URL12:07
therveYeah I'm not sure12:08
therveshardy, We can keep it both. Or maybe add ec2 auth to the native API.12:08
shardyI'm not sure if doing what you have is best, then work out how to make it work later, or align with the current ceilometer capability and end up with an interface we don't want to maintain..12:08
*** jamieh has joined #heat12:08
shardytherve: Hmm, I'd rather not do that if possible - IMO we should aim for the point where every native resource works without ec2 auth12:09
therveWe don't provide any guarantee of integration with ceilometer for now12:09
shardyas some deployers I know don't enable it12:09
*** alexpilotti has joined #heat12:10
* therve nods12:10
shardytherve: sure, just wondering if in it's current state it will actually be useful, I suppose it will work if you use the internal watchrule based alarms?12:10
shardytherve: have you tested it, and if so how?12:11
therveshardy, Well you get integrate with any external system12:11
sgranshardy: https://blueprints.launchpad.net/heat/+spec/securitygroupingressegress12:11
therveI haven't tested it yet, no12:11
*** nkhare has quit IRC12:12
shardyideally I think we want to pass ceilometer a trust, and get it to notify us via heatclient on the user's behalf12:12
therveI was expecting the whole stack user thing to land so that we can test everything12:12
shardys/trust/trust_id12:12
therveThat'd be nice.12:12
therveI don't know if we can still make that change in the cycle, though12:12
shardybut that will have to wait till Juno I guess12:12
shardyyeah probably not hence my query on what interface makes sense now12:13
* shardy was expecting the stack-user stuff to land by now too..12:13
shardythe whole remaining series is ready to merge, just need moar reviews :)12:13
therveWe'll get there :)12:13
*** alexpilotti_ has joined #heat12:14
*** tomek_adamczewsk has quit IRC12:14
*** alexpilotti has quit IRC12:15
*** alexpilotti_ is now known as alexpilotti12:15
*** tomek_adamczewsk has joined #heat12:15
DaveJ__therve: Thanks - finally got my stack relaunched.  Removed the tags as suggested, and just referenced the AutoScaling group.  Cheers12:17
*** gokrokve has joined #heat12:17
therveDaveJ__, So it works?12:18
DaveJ__therve:  The alarm is getting the stats now  (reporting OK state) - just going check that the state changes now12:19
*** gokrokve has quit IRC12:22
*** rpothier has quit IRC12:23
therveshardy, I'm getting " You are not authorized to perform the requested action, identity:list_roles." when trying to create a RDSInstance12:25
therveIt seems to fail when creating a WaitConditionHandle12:25
shardytherve: latest master heat?12:26
therveI think so12:27
shardyOk, wasn't sure if you meant you were testing my instance-users branch12:27
cmysterback, meetings...12:27
cmysterright,12:27
cmysterso it seems that demo's stack view in horizon causes a very nice and long exception, has anyone seen something like that ?12:28
shardytherve: if you're not running the instance-users stuff, the user creating the stack is presumably missing the admin role?12:30
therveAh yeah12:31
therveSorry12:32
*** e0ne has quit IRC12:36
openstackgerritA change was merged to openstack/heat: Change Resource timestamps to save correct info  https://review.openstack.org/7664412:37
openstackgerritA change was merged to openstack/heat: Fix stack_get_all call on stack watcher  https://review.openstack.org/7549512:37
openstackgerritA change was merged to openstack/heat: Alter stack_count_all_by_tenant to stack_count_all  https://review.openstack.org/7085312:37
openstackgerritA change was merged to openstack/heat: Unscoped List Stacks  https://review.openstack.org/6304112:38
openstackgerritA change was merged to openstack/heat: Add project to unscoped stack list response  https://review.openstack.org/7278912:42
openstackgerritZhang Yang proposed a change to openstack/heat: Ensure parameter timeout_mins available in update  https://review.openstack.org/7621112:43
*** mkollaro has quit IRC12:45
*** e0ne has joined #heat12:46
*** ramishra has quit IRC12:47
*** chandan_kumar has quit IRC12:48
*** faramir has joined #heat12:49
*** chandan_kumar has joined #heat12:50
*** saju_m has quit IRC12:52
*** pshchelo has joined #heat12:53
*** chandankumar_ has joined #heat12:54
*** chandan_kumar has quit IRC12:57
*** lsmola has quit IRC12:57
*** ramishra has joined #heat12:59
*** topol has joined #heat12:59
*** sgordon has joined #heat13:06
*** sgordon has quit IRC13:06
*** sgordon has joined #heat13:06
*** guohliu has quit IRC13:07
*** mkollaro has joined #heat13:08
*** rpothier has joined #heat13:08
cmystershardy: ping13:09
openstackgerritcalfonso proposed a change to openstack/heat-templates: Updating yum repo variables and making variables consistent  https://review.openstack.org/7758313:10
*** lsmola has joined #heat13:12
*** aweiteka has joined #heat13:13
*** lsmola has quit IRC13:13
*** lsmola has joined #heat13:14
*** ramishra_ has joined #heat13:15
*** dims_ has quit IRC13:18
*** ramishra has quit IRC13:18
*** gokrokve has joined #heat13:19
*** gokrokve has quit IRC13:23
*** jamieh has quit IRC13:23
*** e0ne_ has joined #heat13:24
openstackgerritSergey Kraynev proposed a change to openstack/heat: Validation for Outputs template section  https://review.openstack.org/7035813:25
*** e0ne has quit IRC13:27
openstackgerritRabi Mishra proposed a change to openstack/heat: Allows vip from a different subnet for lbaas resource  https://review.openstack.org/7708413:28
*** jamieh has joined #heat13:30
*** dims_ has joined #heat13:30
openstackgerritAlexander Ignatov proposed a change to openstack/heat: OS::Savanna::Cluster resource implementation  https://review.openstack.org/7233613:34
*** kevinbenton has quit IRC13:35
*** kevinbenton has joined #heat13:35
*** e0ne has joined #heat13:36
*** wchrisj has joined #heat13:37
*** e0ne has quit IRC13:39
*** e0ne has joined #heat13:39
*** e0ne_ has quit IRC13:40
*** wchrisj has quit IRC13:42
*** sballe_ has joined #heat13:44
*** clarkb has quit IRC13:44
*** radez_g0n3 is now known as radez13:46
*** sballe has quit IRC13:48
*** cmyster has quit IRC13:48
*** achampion has quit IRC13:49
*** ramishra_ has quit IRC13:50
*** funzo_ has quit IRC13:54
*** funzo has joined #heat13:55
*** rustlebee is now known as russellb13:56
*** tteggel has quit IRC13:58
*** tteggel has joined #heat13:59
zigoAs it stands, we have https://review.openstack.org/#/c/69703/ (Add -N option to force using the local environment) which has been abandonned, and https://review.openstack.org/#/c/70070/ (removal of run_tests.sh) that has a bad review. IMO, one of the 2 should be approved ...14:07
zigoPreferably the former, IMO.14:07
shardychmouel: ^^14:08
shardyzigo: that patch looks fine to me (the -N one), so if chmouel can restore it I expect we can get it in14:09
shardyI don't quite get what jpeeler's objection was14:09
sgranshardy: hmm.  First hurdle, and I'm looking for advice14:10
sgrando you have a minute for me to pick your brain?14:10
shardysgran: sure14:10
sgranso, if secgroupingress/egress are separate rules, I need to create them as individual entities (not, I think as a subtemplate - I'll explain in a sec)14:10
sgranso I'm looking to see if heat has a mechanism already to do such a thing?14:11
sgranthe problem with a subtemplate, I think, is that it delays creation of the parent until the subtemplate is created, which creates a dependency loop14:11
sgrannext I'll have to figure out how to check and delete default egress rules if there are any egress rules defined, but I think a call back aking to _lb_reload can do that (maybe)14:12
sgranakin*14:12
*** rbuilta has joined #heat14:12
therveYeah _lb_reload is not a really good example to follow :)14:13
sgranok :)14:13
sgranI'll keep looking, then14:14
therveI'm not sure I understand your problem though14:15
sgranok, let me try to restate the problem14:15
shardysgran: Isn't there some underlying API the resource can just call, e.g whatever nova secgroup-add-rule calls?14:15
sgranin terms of ordering of resources, what I think is correct is that the security group itself should be created, as a standalone and empty entity, then each of the rules should be created afterwards, referencing that security group id14:16
shardysgran: In general you want to avoid any non-trivial coupling between resources, just have the ID of one thing passed into the other via Ref, then use API calls to do your required stuff14:16
sgranthis allows you to have self-referential rules in a security group14:16
sgranbut since the declaration allows you to declare rules in a block inside a group declaration, I have to somehow create the rule resources 'inline'14:17
*** gokrokve has joined #heat14:17
sgranbut not as a subtemplate, since as I understand it, that reverses the ordering and makes the group dependent on the rules14:17
shardysgran: Is it not possible to leave AWS::EC2::SecurityGroup alone, and just make the new resources work independently?14:18
sgranwhat do you do with rules created inside the AWS::EC2::SecurityGroup definition?14:18
shardyCreate the group and add the rules?14:19
shardyI think the depenency-hack nature of these resources is still not that clear to me tbh, did you do a wiki page with some examples?14:19
therve(For me the discussion would need to step back again, I still don't know what we're talking about)14:20
sgranthere's a sample in the blueprint14:20
*** grncdr has quit IRC14:20
sgranhttps://blueprints.launchpad.net/heat/+spec/securitygroupingressegress14:20
*** wchrisj has joined #heat14:20
*** yogesh has joined #heat14:21
pshchelohi all, I have a question on resource updates. how can I reference the old value of the parameter in the handle_update?14:21
pshchelo it seems something has changed recently, as now self.properties in handle_update has properties of the updated stack, but before it was still holding the old values14:21
sgrancurrently, that template sample creates a dependency loop and raises an exception14:21
*** mkollaro has quit IRC14:21
*** gokrokve has quit IRC14:22
*** e0ne has quit IRC14:22
*** e0ne has joined #heat14:22
shardyhttps://blueprints.launchpad.net/heat/+spec/securitygroupingressegress14:22
*** grncdr has joined #heat14:22
shardy(for the benefit of therve and anyone else following..)14:23
*** varora has left #heat14:23
therveMaybe it's a generic problem we want to solve14:23
therveself-reference14:23
sgranlet's not let the perfect be the enemy of the good here :)14:23
sgranI think it's a good thing that, in general, self-referential things cause loops14:24
sgranbut where they are needed, I think it means that finer grained resource management is needed14:24
*** cfriesen_ has joined #heat14:25
*** yogesh has quit IRC14:25
shardysgran: If you were to have any one of the three SecurityGroup* resources able to create the group, and if any found it already existed, ignore it and just add rules, would that solve the problem?14:27
sgranpotentially, but it seems an odd way to model the resources, to me14:28
shardywell they are odd resources ;)14:28
sgranthey're not that strange :)14:29
shardyThe problem is in (nearly) all cases, we expect a resource to create and own the underlying thing, only in this case, we don't because the dependency constraints mean that's not actually possible14:29
sgranit's just a many-to-one relationship, the flip side of the autoscaling group one-to-many relationship14:29
thervesgran, FWIW your problem seems solved in OS::Neutron::SecurityGroup14:30
sgranok, let me look there.14:30
therveAlbeit in a non super great fashion14:30
*** rpothier_ has joined #heat14:30
sgranyeah, almost.  It doesn't allow for rules being created outside of the security group declaration, but maybe that's ok14:32
sgranfor now, at least14:32
*** achampion has joined #heat14:32
*** john-n-seattle has joined #heat14:33
*** rpothier has quit IRC14:33
thervesgran, Does the AWS one allows it for neutron?14:33
therveIt actually feels weird that it does for nova14:34
sgranhmm.  No, wait.  I don't the neutron security group resource will allow for self-referential rules, will it?14:34
therveYou don't... see how?14:35
sgranyes, sorry - thinking faster than typing :)14:36
sgranI don't think that you can create rules referencing the containing security group with that code.  Did I miss something?14:36
therveIIUC, if you say  remote_mode=remote_group_id and remote_group_id=None, it will self-reference14:36
sgranbecause you would still have a resource with a ref: to itself14:36
thervesgran, https://github.com/openstack/heat/blob/master/heat/engine/resources/neutron/security_group.py#L17014:37
therveSure you can't self-reference14:37
sgranI see.  I think I agree that it's a non super great way :)14:39
sgranbut it looks functional, at least14:39
*** sabeen has joined #heat14:39
therveWell, yeah14:40
thervesgran, The best way would be to add a SecurityGroupRule resource14:40
therveI think14:40
sgranyes, I agree :)14:41
sgranthat's what I'm looking at doing for the AWS type14:41
therveAh I guess they have separate AWS rules?14:41
sgranyeah, SecurityGroup for the group, SecurityGroupIngress and SecurityGroupEgress for the rules14:42
therveWell I'm not super excited for introducing new ones, but why not14:42
therveIt should be fairly straightforward14:42
therveAh it's interesting, they mention they introduced them just because of self-referencing14:43
*** mkollaro has joined #heat14:44
*** wchrisj has quit IRC14:45
*** sabeen has quit IRC14:48
*** sabeen has joined #heat14:48
*** jergerber has joined #heat14:49
*** zns has joined #heat14:50
*** vijendar has joined #heat14:52
*** andersonvom has joined #heat14:53
*** wchrisj has joined #heat14:53
*** jpeeler has left #heat14:54
*** jpeeler has joined #heat14:54
*** jpeeler has quit IRC14:54
*** jpeeler has joined #heat14:54
openstackgerritJason Dunsmore proposed a change to openstack/heat: Make server resources use nova_utils.refresh_server()  https://review.openstack.org/7200914:55
*** ccorrigan has quit IRC14:58
*** alexheneveld has quit IRC15:02
*** bada has joined #heat15:04
*** daneyon has joined #heat15:04
*** blomquisg has joined #heat15:05
*** daneyon has quit IRC15:05
*** daneyon has joined #heat15:06
*** faramir has quit IRC15:09
*** daveed has quit IRC15:16
*** rcleere has joined #heat15:16
*** daveed has joined #heat15:16
*** daveed has quit IRC15:16
*** gokrokve has joined #heat15:17
*** gokrokve has quit IRC15:21
*** pablosan has joined #heat15:23
*** gokrokve has joined #heat15:26
therveSo I've found a bunch of issues with TemplateResource15:27
therveIt doesn't work with resource-list, which means for example resources are missing from the documentation15:28
therveAnd resource-show breaks for them15:28
therveI also noticed that there is not template validation15:28
therveWould it make sense to generate a class with properties_schema/attributes_schema, instead of doing it in init?15:29
therveIt's a bit meta-programming-y, but it's the path of least change15:29
therveshardy, ^^^15:31
openstackgerritPavlo Shchelokovskyy proposed a change to openstack/heat: Add handle_update to VolumeAttachment  https://review.openstack.org/7268115:33
*** david-lyle has joined #heat15:37
*** coolsvap has joined #heat15:40
*** clarkb has joined #heat15:41
*** chandan_kumar has joined #heat15:42
*** beekneemech is now known as bnemec15:43
openstackgerritAnderson Mesquita proposed a change to openstack/heat: Restructure contrib/ directories  https://review.openstack.org/7341715:53
openstackgerritAnderson Mesquita proposed a change to openstack/heat: Generate docs for contrib plugins  https://review.openstack.org/6875115:53
*** jmckind has joined #heat16:00
*** tomek_adamczewsk has quit IRC16:00
*** chandan_kumar has quit IRC16:03
*** bada has quit IRC16:03
*** alexheneveld has joined #heat16:04
*** kfox1111 has joined #heat16:08
*** ramishra has joined #heat16:08
*** lazy_prince is now known as killer_prince16:12
*** kfox1111 has quit IRC16:13
*** ifarkas has quit IRC16:13
*** chandan_kumar has joined #heat16:16
openstackgerritJeff Peeler proposed a change to openstack/heat: Add host_routes property to Neutron subnet resource  https://review.openstack.org/7695016:17
*** radez is now known as radez_g0n316:19
*** randallburt has joined #heat16:20
*** randallburt has quit IRC16:21
*** randallburt has joined #heat16:21
*** tims has joined #heat16:22
*** jprovazn has quit IRC16:24
*** ramishra has quit IRC16:25
openstackgerritRandall Burt proposed a change to openstack/heat: Refactor CLB to work with groups  https://review.openstack.org/6558616:26
*** dtalton has joined #heat16:29
*** ramishra has joined #heat16:29
*** dtalton has left #heat16:30
*** TonyBurn_ has quit IRC16:30
radixgood morning16:31
openstackgerritRabi Mishra proposed a change to openstack/heat: Allows vip from a different subnet for lbaas resource  https://review.openstack.org/7708416:33
radixtherve: nice, just saw your ScalingPolicy :)16:37
therveradix, Yeah I hope it does the trick16:38
therveI'm a bit annoyed by ChangeInCapacity and all16:38
therveBut changing it looks tricky at that point16:39
*** gokrokve has quit IRC16:39
radixhmm16:39
radixtherve: I don't think it'll be too bad16:39
therveNo me neither16:39
radixtherve: here's the trick: separate the interface from the internal message passing16:40
radixwe can continue to use "ChangeInCapacity" as the thing that gets passed to methods16:40
radixI was just about to write a comment about that :)16:40
therveYeah it's not that trivial :)16:40
radixhow come?16:41
zigoshardy: Chmouel restored his patch: https://review.openstack.org/6970316:42
therveradix, Well I need to reimplement some code if I try to do that16:42
therveNamely the call to adjust I think16:42
radixtherve: oh, yeah, I think it'll require a bit of a refactor there16:42
radixbut at least you don't need to touch *Group16:42
sdakemorning16:43
radixtherve: but that can be a super small refactor, just replace CHANGE_IN_CAPACITY with self.CHANGE_IN_CAPACITY and add the class variables16:43
shardyjpeeler: ^^ can you clarify your objection to that patch pls?16:43
shardyzigo: Thanks, +2 from me16:43
openstackgerritJenkins proposed a change to openstack/heat: Updated from global requirements  https://review.openstack.org/7668916:46
*** TonyBurn_ has joined #heat16:46
jpeelershardy: i just thought that such a change would sync all virtual env related, and we were missing the warning about running a local copy of flake816:48
therveradix, Yeah I meant it's little bit more than that :)16:50
shardyjpeeler: Ok, cool - another iteration to fix that up seems fine, but do you think adding some comments to the patch explaining exactly what is missing would be helpful? It's not that clear what needs to be done from your comment IMO :)16:51
*** ifarkas has joined #heat16:54
*** TonyBurn_ has quit IRC16:55
radixtherve: hmm, what else am I missing?16:55
radixoh, right16:55
radixtherve: maybe properties for getting the value to pass16:55
*** zns has quit IRC16:56
radixor accessors anyway16:56
radixself.change_in_capacity() or whatever16:56
radixI think I'm going to need to do something similar for the rolling updates code16:56
*** e0ne has quit IRC16:58
*** spzala has joined #heat16:58
*** gokrokve has joined #heat16:58
*** che-arne has joined #heat16:59
jpeelershardy: any suggestions for better wording on https://review.openstack.org/#/c/76655/? for the most part, i've been trying to be consistent with phrasing found elsewhere17:00
DaveJ__therve: Thanks for you help earlier.  Looks like the alarms / meters are partially working.  I have two mestled stacks in my template ''AWS::CloudFormation::Stack'.  They both reference the same template, which defines a scaling group.  Looks like one group gets meters created, but the second one doesn't.  I'll dig a bit more, but I'm guessing this might be17:01
DaveJ__some sort of issue with nestled stacks ?17:01
openstackgerritTim Schnell proposed a change to openstack/heat: Adds Parameter Label to template validate call  https://review.openstack.org/6532517:03
*** lsmola has quit IRC17:03
*** andrew_plunk has joined #heat17:03
*** jistr has quit IRC17:08
*** coolsvap has quit IRC17:10
*** coolsvap has joined #heat17:11
*** gokrokve has quit IRC17:14
*** gokrokve has joined #heat17:17
*** gokrokve has quit IRC17:20
*** julienvey has left #heat17:21
*** chandan_kumar has quit IRC17:25
shardyjpeeler: Ok, I see the admin-only badness is rife throughout the Neutron resources, so we can consider how to fix that in some future patch17:35
openstackgerritThomas Herve proposed a change to openstack/heat: Native ScalingPolicy resource  https://review.openstack.org/7752917:40
*** Michalik has quit IRC17:42
*** 1JTAAFJP3 has quit IRC17:45
*** killer_prince is now known as lazy_prince17:45
*** harlowja_away is now known as harlowja17:46
*** pvaneck has joined #heat17:46
*** harlowja is now known as harlowja_away17:47
DaveJ__therve:  Collapsed my nestled stack into just a single template.  With two auto-scaling groups, I don;t seem to get meters create for the second group.  Seems like a potential bug. I didn't see anything in launchpad, do you know of anything or should I raise it ?17:48
therveDaveJ__, If you can offer some reproducible behavior, please open it17:49
*** harlowja_away is now known as harlowja17:50
therveThat sounds weird though17:50
*** john-n-s- has quit IRC17:50
*** alexheneveld_ has joined #heat17:53
zanebI think we should start using the term "nestled stack" :)17:53
*** zns has joined #heat17:53
*** alexheneveld has quit IRC17:55
*** alexheneveld_ is now known as alexheneveld17:55
*** rwsu has joined #heat17:55
*** john-n-s- has joined #heat17:55
*** jdob has joined #heat17:57
randallburtzaneb:  +1 sounds so much more cozy17:59
*** ramishra has quit IRC18:00
jasond`zaneb: don't forget to update the glossary :)18:00
zaneb"Nestled Stack: like a Nested Stack, but cozier"18:01
*** derekh has joined #heat18:03
*** jprovazn has joined #heat18:11
*** Michalik has joined #heat18:12
*** nati_ueno has joined #heat18:13
aweitekai tried locking down my heat stack with AWS::EC2::SecurityGroup. Egress rules are failing with "Egress rules are only allowed when Neutron is used and the 'VpcId' property is set."18:21
*** daneyon has quit IRC18:23
aweitekais this a known issue? http://paste.openstack.org/show/71840/18:23
*** nszceta has joined #heat18:23
aweitekajpeeler, fyi ^^ if you attemp to lock down your openshift heat template18:23
shardyaweiteka: Can you use OS::Neutron::SecurityGroup instead?18:24
shardyand/or set the VpcId as stated in the error?18:25
radixtherve: woot :)18:25
aweitekashardy, i missed that one. i'll give that a try. otherwise you're suggesting a bogus vpcid string to satisfy the aws requirement?18:26
*** arbylee has joined #heat18:28
openstackgerritJeff Peeler proposed a change to openstack/heat: Add host_routes property to Neutron subnet resource  https://review.openstack.org/7695018:28
*** alexheneveld has quit IRC18:29
shardyaweiteka: Specifying VpcId will make the resource use neutron instead of nova-network, which doesn't support egress rules18:29
shardyaweiteka: I would assume it needs to be a neutron network ID not a dummy value though18:29
aweitekashardy, ok. i'll try neutron::securitygroup. thx18:30
*** yogesh has joined #heat18:30
*** radez_g0n3 is now known as radez18:34
*** arbylee has quit IRC18:39
*** arbylee has joined #heat18:39
*** tspatzier has quit IRC18:42
*** cody-somerville has joined #heat18:53
*** cody-somerville has quit IRC18:53
*** cody-somerville has joined #heat18:53
*** piousbox has joined #heat18:55
*** e0ne has joined #heat18:59
*** Tross has quit IRC19:00
*** rbuilta has quit IRC19:05
*** rcleere has quit IRC19:12
*** daneyon has joined #heat19:14
*** rcleere has joined #heat19:15
*** rbuilta has joined #heat19:18
stevebakermorning19:26
*** cody-somerville has quit IRC19:27
aweitekashardy, looks like os:nova:securitygroup was added in icehouse. i'm on havana. i'll try the vpcid at some point19:27
sdakeshardy seen this before?19:27
sdake2014-03-03 12:27:33.862 WARNING heat.engine.parser [req-0daabd94-5de1-4b3c-add7-3aab444be372 None admin] Unable to set parameters StackId identifier19:27
*** coolsvap has quit IRC19:29
stevebakeraweiteka: The vpcid should be the neturon network id19:29
openstackgerritJason Dunsmore proposed a change to openstack/heat: Keep trying to get SSH transport for up to 30 sec  https://review.openstack.org/7767719:30
aweitekastevebaker, ok, thanks. i'll pop that in an give it a go.19:30
*** jmckind has quit IRC19:31
*** jmckind has joined #heat19:32
*** e0ne has quit IRC19:35
*** tspatzier has joined #heat19:35
openstackgerritA change was merged to openstack/heat: Remove redundant default value None for dict.get  https://review.openstack.org/7561219:35
openstackgerritA change was merged to openstack/heat: Restructure contrib/ directories  https://review.openstack.org/7341719:36
openstackgerritA change was merged to openstack/heat: Generate docs for contrib plugins  https://review.openstack.org/6875119:36
*** e0ne has joined #heat19:36
*** pablosan has quit IRC19:39
*** cody-somerville has joined #heat19:39
*** sballe has joined #heat19:39
*** rbuilta has quit IRC19:40
*** sballe_ has quit IRC19:40
*** zns has quit IRC19:52
*** Tross has joined #heat19:53
*** Tross has quit IRC19:53
openstackgerritA change was merged to openstack/heat: Document schema properties for Neutron subnet resource  https://review.openstack.org/7665519:54
stevebakerMurano have their incubation discussion in the TC meeting today for those who are interested in how it relates to Heat19:54
stevebakershardy, sdake, zaneb ^19:54
jpeelerwas it decided the vim headers could all go away?19:55
zanebstevebaker: thanks for the heads-up, that might be interesting19:55
stevebakergrab yr popcorn19:55
zanebjpeeler: I think there was consensus within the Heat project that they could19:55
*** Tross has joined #heat19:56
*** e0ne has quit IRC19:58
*** lindsayk has joined #heat19:59
openstackgerritJason Dunsmore proposed a change to openstack/heat: Remove vim header from files.  https://review.openstack.org/7637820:00
openstackgerritPaul Van Eck proposed a change to openstack/heat: Allow proper instance with volume suspension  https://review.openstack.org/7612620:00
stevebakeroh, wait. TC meeting is tomorrow20:01
stevebakerduh20:02
zanebstevebaker: you had me all psyched up for it :/20:03
*** daneyon has quit IRC20:03
stevebakerzaneb: you've got 24 hours to decide if building an imperative DSL to generate heat templates is a good idea20:04
openstackgerritAnderson Mesquita proposed a change to openstack/python-heatclient: Add stack-preview support  https://review.openstack.org/7449820:04
zanebstevebaker: I don't need 24 hours to know that that is not a good idea20:05
stevebakerexactly20:05
zanebis that seriously what they're doing now?20:05
stevebakeryes, currently XML based20:06
zaneboh God20:06
zanebit's getting worse20:06
*** tspatzier has quit IRC20:07
*** lindsayk1 has joined #heat20:07
*** lindsayk has quit IRC20:08
*** Tross has quit IRC20:08
stevebakerzaneb: making the DSL yaml probably wouldn't help https://etherpad.openstack.org/p/murano-new-dsl-example20:08
zanebagree, but XML is adding insult to injury ;)20:09
*** jmckind has quit IRC20:12
*** tims1 has joined #heat20:13
*** e0ne has joined #heat20:15
*** tims has quit IRC20:16
openstackgerritZane Bitter proposed a change to openstack/heat: Parse JSON as JSON when converting to YAML  https://review.openstack.org/7768720:18
randallburtyo dog. I heard you like imperative DSL, so I put some imperative DSL to generate your declarative DSL.20:21
zanebrandallburt: I suspect at least 70% of the people in this channel are not going to get that reference ;)20:24
randallburtzaneb:  its ok. I sometimes just amuse myself ;)20:24
*** tims has joined #heat20:26
zaneb"Change has been successfully merged into the git repository."20:26
zanebandersonvom: ^ and not a moment too soon :)20:26
*** tims1 has quit IRC20:26
*** mspreitz has joined #heat20:27
andersonvomzaneb: I was about to ping you because I saw this morning that YET ANOTHER rebase was needed, but then I saw you had accepted it already o/20:27
andersonvomzaneb: change away now! ;)20:28
zanebyaaaaay! I can finally rebase20:28
mspreitzHas anybody run DevStack in the last 23 hours?20:29
mspreitz(I did one, and it produced an installation in which deleting a stack fails)20:31
andersonvom\o/20:31
*** tspatzier has joined #heat20:32
*** saurabhs has joined #heat20:32
*** arbylee has quit IRC20:33
*** zns has joined #heat20:33
*** tspatzier has quit IRC20:33
*** asalkeld has quit IRC20:36
*** ifarkas has quit IRC20:38
*** mspreitz has left #heat20:39
*** wchrisj has quit IRC20:47
*** asalkeld has joined #heat20:49
*** arbylee has joined #heat20:50
*** mkollaro has quit IRC20:51
*** mkollaro has joined #heat20:52
*** tims has quit IRC20:53
*** tims has joined #heat20:53
*** asalkeld has quit IRC20:54
*** daneyon has joined #heat20:55
*** alexheneveld has joined #heat20:56
*** tims has quit IRC20:58
*** jamieh has quit IRC21:07
*** asalkeld has joined #heat21:07
*** zns has quit IRC21:08
*** denis_makogon has quit IRC21:10
*** dmakogon_ has joined #heat21:11
*** e0ne has quit IRC21:21
openstackgerritJason Dunsmore proposed a change to openstack/heat: Make Keystone client pluggable  https://review.openstack.org/7419021:23
openstackgerritJason Dunsmore proposed a change to openstack/heat: Add Keystone V2 plugin  https://review.openstack.org/7419121:23
radixwas that imperative DSL stuff about murano?21:23
*** zns has joined #heat21:23
randallburtradix:  yes21:23
radixok.21:23
openstackgerritAndrew Plunk proposed a change to openstack/heat: Nova keypair validation breaks client plugins  https://review.openstack.org/7344121:31
*** andersonvom_ has joined #heat21:41
*** andersonvom has quit IRC21:42
*** tango|2 has joined #heat21:42
*** andersonvom_ has quit IRC21:43
*** david-lyle has quit IRC21:43
*** andersonvom has joined #heat21:43
aweitekaanyone have experience with LBaaS session_persistence? i've tried types APP_COOKIE and SOURCE_IP with no luck.21:51
shardystevebaker: hey, around?21:54
stevebakershardy: yo21:54
shardystevebaker: Hi, just wanted to sync up re the FF deadline and instance-users remaining patches..21:54
shardythanks for all the reviews so far, do you have any blocker issues with the remaining patches or are we looking good to get them in?21:55
shardyI saw the _suspend_user comment and replied, can't we just use the superclass handle_suspend?21:55
shardyshadowing all the handle_foo methods seems pretty cumbersome unless we really need to21:56
*** rpothier__ has joined #heat21:56
stevebakershardy: my only concern is that the tripleo CI cloud will fall over if instance-users lands without them putting in config changes first. Is that a think that could happen?21:56
stevebakershardy: like with https://review.openstack.org/#/c/73978/7/heat/common/heat_keystoneclient.py will heat even fail to start if there is no configured stack_user_domain?21:57
shardystevebaker: I'm really not sure, I pinged SpamapS re this and the move to trusts and he was +1 on it, ref comment on bug #128615721:58
shardystevebaker: probably not but all the stuff related to deferred auth will fail21:59
*** rpothier_ has quit IRC21:59
stevebakershardy: also could you clarify, are instance-users still used if trusts isn't?21:59
shardystevebaker: really, I'm not clear on how or where to get the required changes in to tripleo, we've done all the work to make everything in to allow the gate jobs to work, so I don't see how we can gate features on another project22:00
stevebakershardy: as long as they are aware, that is likely all we can do22:01
shardystevebaker: they are two different use-cases - instance-users solves the admin requirement when creating users associated with resources (e.g SignalResponder)22:01
shardylifeless, SpamapS: ^^22:01
shardystevebaker: and trusts solves the requirement to pass a password as well as a token22:02
shardy(as well as improving the general security model)22:02
shardythe two things can be merged independently, but it seemed to make sense to get instance-users in first, because of the switch of users in tempest from admin to demo22:03
stevebakerok, thanks22:03
shardystevebaker: however I've now posted a devstack patch which adds heat_stack_owner role to admin and demo, so it doesn't matter22:03
shardystevebaker: The issue with switching to trusts by default is it's a one line heat patch which breaks $all the tests22:04
shardyI'm still working on solving that22:04
stevebakerlol22:04
*** harlowja is now known as harlowja_away22:05
*** harlowja_away is now known as harlowja22:05
*** radez is now known as radez_g0n322:07
stevebakershardy: btw I'm working on changes to Server and SoftwareDeployment to support keystone auth for polling and signalling22:07
shardystevebaker: sounds good, other than the auth those patches all look great to me22:08
*** gokrokve has joined #heat22:08
stevebakershardy: I'm OK with it going in for now, but is there any reason to set self.password rather than just passing password in on _create_user()?22:09
shardystevebaker: Not really, I did it like that because I originally expected subclasses to set self.password then call the superclass handle_create22:10
shardyrather than having a separate create method22:10
stevebakerah, I might change that in my series22:10
shardyok, cool, either is fine with me :)22:11
stevebakershardy: also, could you provide a list of auth info that needs to be sent to the server for native auth (username, password, auth_url, instance-user-project-name?, region-name?)22:12
shardystevebaker: not sure what you mean, we can just use heat resource-signal and a username/password?22:13
stevebakerwill need auth_url, and won't the tenant name come from the project used for instance-users?22:14
shardystevebaker: Oh yeah sorry, username/password/project/auth_url22:15
stevebakerand the project comes from CONF?22:15
shardywe will probably need to use project_id unless we also pass the domain22:15
shardyNo the project is created per stack and stored in stack.stack_user_project_id22:16
stevebakerah, ok22:16
shardyThe domain ID comes from the conf as it's a one-time thing22:16
shardybut I wanted project isolation between each stack22:17
*** nszceta has quit IRC22:17
stevebakershardy: Only one nit https://review.openstack.org/#/c/72763/22:20
stevebakershardy: you need to shop around for more reviewers now22:20
openstackgerritRandall Burt proposed a change to openstack/heat: Return None when get_attr cannot resolve a value  https://review.openstack.org/7771322:22
shardystevebaker: Thanks, doh missed that commented stuff, 2mins I'll push a new series22:23
*** zns has quit IRC22:23
openstackgerritSteven Hardy proposed a change to openstack/heat: migrate User/AccessKey resources to StackUser base class  https://review.openstack.org/7276322:28
openstackgerritSteven Hardy proposed a change to openstack/heat: StackUser add _delete_keypair function  https://review.openstack.org/7276222:28
openstackgerritSteven Hardy proposed a change to openstack/heat: engine: allow stack_user_project users to retrieve stack  https://review.openstack.org/7130022:28
openstackgerritSteven Hardy proposed a change to openstack/heat: Add test for StackUser._create_keypair  https://review.openstack.org/7276122:28
openstackgerritSteven Hardy proposed a change to openstack/heat: Add config options to specify stack domain admin  https://review.openstack.org/7603522:28
openstackgerritSteven Hardy proposed a change to openstack/heat: StackUser add suspend/resume support  https://review.openstack.org/7193022:28
openstackgerritSteven Hardy proposed a change to openstack/heat: migrate StackUser base class to stack domain users  https://review.openstack.org/7121022:28
openstackgerritSteven Hardy proposed a change to openstack/heat: heat_keystoneclient add delete_stack_domain_user_keypair  https://review.openstack.org/7192922:28
openstackgerritSteven Hardy proposed a change to openstack/heat: Modify stack_user_domain config option to take domain ID  https://review.openstack.org/7397822:28
shardystevebaker: when are you planning to branch for i3?22:28
stevebakershardy: not sure yet, I haven't spoken with ttx22:29
shardystevebaker: Ok cool not imminent then, we should have tomorrow at least to get thru the ever growing review backlog ;)22:30
stevebakershardy: I intend to ask for an extension for any High priority blueprint not yet landed22:31
shardystevebaker: Ok, sounds good - if possible I'd really like instance users to make the i3 cut though, so we can increase the test exposure (and time to fix fallout in the event of problems ;)22:32
*** aweiteka has quit IRC22:32
stevebakershardy: yes, agreed22:32
shardystevebaker: Ok, well thanks, off for some sleep now, I'll hit the reviews in the morning!22:33
stevebaker\o22:34
*** rpothier__ has quit IRC22:35
*** yogesh has quit IRC22:35
*** sgordon has quit IRC22:37
openstackgerritAnderson Mesquita proposed a change to openstack/heat: Add admin_pass as attribute to CloudServer  https://review.openstack.org/7274522:38
*** zns has joined #heat22:40
*** vijendar has quit IRC22:40
*** alexheneveld has quit IRC22:41
*** yogesh has joined #heat22:45
stevebakerandersonvom: can management-api be marked Implemented? all posted changes are merged22:46
* SpamapS reads backscroll22:47
SpamapSshardy: reading22:47
SpamapSshardy: I'm not sure I understand 100%. The change to trusts requires a domain? Or the instance-users requires a domain? And if so.. is there something we need to do in TripleO to enable that? can we just assert that the domain exists and be done with it?22:49
andersonvomstevebaker: done!22:51
*** dims_ has quit IRC22:52
*** mkollaro has quit IRC22:53
zanebfor everyone else sick of the voting interface for talks: http://zaneb.fedorapeople.org/juno/talks.html22:57
zanebyou're welcome22:57
*** nati_uen_ has joined #heat22:58
*** nati_ueno has quit IRC23:00
SpamapSzaneb: \o/23:00
zanebSpamapS: I'm glad it's not just me23:00
zanebI complained to them last time23:00
zanebthey said they would appreciate feedback23:01
zanebI told them exactly how to fix it23:01
zaneband this time it's exactly the same :(23:01
*** jergerber has quit IRC23:02
zanebhttps://www.openstack.org/vote-atlanta/Presentation/continuous-integration-from-design-to-production-of-highly-distributed-applications-using-docker-openstack-and-tosca <- does this guy have _any_ idea what he's talking about?23:03
*** dims_ has joined #heat23:08
*** wchrisj has joined #heat23:09
asalkeldsuper sucky web site23:09
asalkeld(https://www.openstack.org/vote-atlanta/Presentation)23:10
*** sabeen has quit IRC23:16
*** david-lyle has joined #heat23:19
*** daneyon has quit IRC23:20
zanebasalkeld: it sure is23:21
asalkeldoftc any one?23:23
asalkeld#openstack-heat23:23
zanebasalkeld: who/what is there?23:24
*** pablosan has joined #heat23:24
asalkeldwe are moving from freenode23:24
asalkeldto OFTC23:24
zanebreally?23:24
asalkeldcome on get with the times;)23:24
asalkeldall the dos attacks on freenode me thinks23:24
zanebI just removed OFTC from my favourites list23:24
*** pablosan has quit IRC23:25
zanebhaven't actually seen anyone in the #matahari room in years23:25
*** achampion has quit IRC23:25
asalkeldyeah I still had that23:25
*** pablosan has joined #heat23:25
*** blomquisg has quit IRC23:27
*** andrew_plunk has quit IRC23:32
*** arbylee has quit IRC23:33
*** rcleere has quit IRC23:35
*** yogesh has quit IRC23:39
*** blomquisg has joined #heat23:41
*** dims_ has quit IRC23:46
*** rpothier has joined #heat23:51
*** lindsayk1 has quit IRC23:56
*** lindsayk has joined #heat23:57
*** lindsayk1 has joined #heat23:58
*** ramishra has joined #heat23:58
*** lindsayk1 has quit IRC23:59
« Sunday, 2014-03-02 Index Tuesday, 2014-03-04 »

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