Tuesday, 2018-06-12

*** lifeless has quit IRC00:09
*** jmlowe has joined #heat00:41
*** mdnadeem has joined #heat01:13
openstackgerritJiye Yu proposed openstack/heat-dashboard master: Template Generator: Add support for OS::Designate::Zone  https://review.openstack.org/57421001:16
*** mdnadeem has quit IRC01:20
*** dyasny__ is now known as dyasny01:25
*** mdnadeem has joined #heat01:33
openstackgerritJiye Yu proposed openstack/heat-dashboard master: Template Generator: Add support for OS::Designate::Zone  https://review.openstack.org/57421001:36
*** pengdake has joined #heat01:36
*** lifeless has joined #heat01:41
*** hongbin has joined #heat02:02
*** ricolin__ has joined #heat02:06
openstackgerritKazunori Shinohara proposed openstack/heat-templates master: Add sample templates for Blazar  https://review.openstack.org/57361402:17
*** lifeless has quit IRC02:32
*** lifeless has joined #heat02:33
*** rcernin is now known as rcernin|lunch02:35
openstackgerritJiye Yu proposed openstack/heat-dashboard master: Template Generator: Add support for OS::Designate::Zone  https://review.openstack.org/57421002:40
flwangricolin__: around? i have a question about the heat template in Magnum02:50
flwangricolin__: currently, Magnum is using this way https://github.com/openstack/magnum/blob/master/magnum/drivers/common/templates/environments/with_master_lb.yaml to enable/disable something in template02:51
flwangricolin__: now both master LB and etcd lb are using same 'fake' resource type and that means, when disable the master lb, the etcd lb.02:52
flwangif we would like to enable master lb but disable etcd lb, does that mean we have to create another 'fake' resource type for etcd lb ? thanks02:52
*** mdnadeem has quit IRC02:53
*** ysandeep has joined #heat02:58
*** skramaja has joined #heat03:18
*** hongbin has quit IRC03:26
*** ysandeep has quit IRC03:58
*** mdnadeem has joined #heat03:59
*** mdnadeem has quit IRC04:00
*** mdnadeem has joined #heat04:00
ricolin__flwang, are you talking about something like `"Magnum::Optional::Neutron::LBaaS::XXXX`: {some template to enable master lb or etcd lb}04:08
*** mdnadeem has quit IRC04:08
flwangyes04:08
openstackgerritRabi Mishra proposed openstack/heat-tempest-plugin master: Fix software config/deployment api tests  https://review.openstack.org/57454204:09
flwangricolin__: does the way i'm using work? https://review.openstack.org/#/c/574540/04:09
*** jmlowe has quit IRC04:10
*** ramishra has joined #heat04:10
ricolin__flwang, so api_gateway_switcher_pool== enable etcd lb?04:11
*** mdnadeem has joined #heat04:11
flwangricolin__: by current design, if user set master_lb_enabled==True, then both master LB and etcd LB will be enabled04:11
flwangbut we want to see only enable master LB but don't create LB for etcd04:12
ricolin__flwang, looks valid to me04:15
ricolin__since the resource type only use internal in magnum, provide a new resource type for etcd lb is fine IMO04:16
flwangricolin__: so you mean  "Magnum::Optional::Etcd::Neutron::LBaaS::LoadBalancer": "Magnum::Optional::Neutron::LBaaS::LoadBalancer" is valid?04:17
flwangheat can find the correct resource for Magnum::Optional::Neutron::LBaaS::LoadBalancer ?04:17
ricolin__no04:18
ricolin__I mean to create a resource type is correct04:19
ricolin__but if you use "Magnum::Optional::Etcd::Neutron::LBaaS::LoadBalancer": "Magnum::Optional::Neutron::LBaaS::LoadBalancer"04:19
ricolin__Isn't that means if we disable master lb, `Magnum::Optional::Neutron::LBaaS::LoadBalancer` going to be None04:20
ricolin__and so `"Magnum::Optional::Etcd::Neutron::LBaaS::LoadBalancer` will be none04:20
flwangoh, sure, could you please help review my patch ?04:20
ricolin__flwang, sure04:20
flwangthanks ;)04:21
ricolin__flwang, NP!:)04:21
*** ricolin__ has quit IRC04:21
*** pengdake has quit IRC04:21
*** ricolin has joined #heat04:22
openstackgerritKazunori Shinohara proposed openstack/heat-templates master: Add sample templates for Blazar  https://review.openstack.org/57361404:27
*** flwang1 has quit IRC04:52
*** mdnadeem has quit IRC04:57
*** mdnadeem has joined #heat04:58
openstackgerritMehdi Abaakouk (sileht) proposed openstack/heat-tempest-plugin master: gabbi require host to be set to something valid  https://review.openstack.org/57455005:01
ricolinkazsh_, reviewed you Blazar patch(sorry took that long), good stuff, but I think it needs some further works:)05:16
kazsh_ricolin: thx, will check your commnet05:17
*** pengdake has joined #heat05:20
*** AlexeyAbashkin has joined #heat05:39
*** pengdake has quit IRC05:39
*** Alexey_Abashkin has joined #heat05:42
*** Alexey_Abashkin has quit IRC05:42
*** Alexey_Abashkin has joined #heat05:42
*** AlexeyAbashkin has quit IRC05:43
*** Alexey_Abashkin is now known as AlexeyAbashkin05:43
*** AlexeyAbashkin has quit IRC05:45
*** pengdake has joined #heat05:54
*** zzzeek has quit IRC05:54
*** zzzeek has joined #heat05:58
*** jtomasek has joined #heat06:11
*** jtomasek has quit IRC06:11
*** jtomasek_ has joined #heat06:13
*** jtomasek_ has quit IRC06:16
openstackgerritKazunori Shinohara proposed openstack/heat master: Add Blazar support to Heat  https://review.openstack.org/53457606:31
*** jtomasek has joined #heat06:38
*** pengdake has quit IRC06:41
*** AlexeyAbashkin has joined #heat06:42
ricolinkazsh_, I think it will be better to at least leave some descriptions in resource to demonstrate why we think leave headle_update is a better idea06:51
*** hoonetorg has quit IRC07:01
*** pengdake has joined #heat07:03
*** AlexeyAbashkin has quit IRC07:04
kazsh_ricolin: yeah, you are super correct, will do that :)07:06
ricolinkazsh_, thx:)07:06
*** AlexeyAbashkin has joined #heat07:10
*** rcernin|lunch has quit IRC07:13
*** gkadam has joined #heat07:15
*** lifeless has quit IRC07:39
*** lifeless has joined #heat07:39
*** serlex has joined #heat07:40
*** hoonetorg has joined #heat07:40
*** shardy has joined #heat07:51
strigaziricolin: you there?07:54
ricolinstrigazi, hi07:54
strigaziricolin: quick question for stack update07:55
strigaziricolin: I'm doing an update on a RG with one server07:55
strigaziricolin: changing only user_data and expecting server replace07:55
strigaziricolin: instread I get two servers with the same name and the new one has no network07:56
strigaziricolin: when I say not network, I mean no nics attached07:56
ricolinstrigazi, is the stack completed?07:57
strigaziricolin: no07:57
strigaziricolin: it is stuck in UPDATE_IN_PROGRESS07:58
ricolinstrigazi, usually we will create a new server, detach network from old, and add it to new one, and than delete old server07:58
strigaziricolin the thing is that the server has software deployments07:58
strigaziricolin: and without network it won't manage to deploy anything07:59
*** ysandeep has joined #heat08:00
ricolinAnything stuck in In_progress is a bug(story):/08:02
ricolinis the software deployment resource depends on server resource?08:02
strigaziricolin the SD is to applied to that resource08:07
strigaziricolin: so yes, the SD depends on the resource08:07
ricolinstrigazi, Okay, just try to make sure, because if you directly input an server id without implicit or explicit reference to resource it won't depend on it.08:12
ricolinIs there any more exception happens in log?08:12
strigaziricolin: https://github.com/openstack/magnum/blob/master/magnum/drivers/k8s_fedora_atomic_v1/templates/kubemaster.yaml#L66408:13
ricolinyep, an implicit depend08:15
strigaziricolin: would it be better if we move the SDs outside the RG?08:16
*** derekh has joined #heat08:17
strigaziricolin: and apply the SD on the RG after?08:18
strigaziricolin: it is a big a change but what we have now it a bit weird08:18
strigaziricolin: it is a big a change but what we have now is a bit weird08:18
strigaziricolin: how would an explicit depend look like?08:18
*** ysandeep_ has joined #heat08:19
strigaziricolin: Do I need to set the dependency  on network too?08:21
ricolinI don't think move SD out will help in this case08:27
ricolinstrigazi, a explicit will be to add `depends_on` in resource08:27
ricolinbut implicit should be enough08:27
strigaziricolin: it seeds it isn't08:28
strigaziricolin: I added the dependency explicitly08:28
strigaziricolin: first port, server and then SD08:29
ricolinyou mean it still will stuck in Update in progress?08:30
strigaziricolin: I deleted the stack and now I'm waiting for new one with the deps08:30
ricolinIs there any exception during update?08:31
ricolinthe interesting part is in magnum's SD resource here https://github.com/openstack/magnum/blob/master/magnum/drivers/k8s_fedora_atomic_v1/templates/kubemaster.yaml#L67008:34
ricolinsaid only apply config when create resource08:34
ricolinWhy that SD working for update?08:35
ramishraricolin: replacing the server would replace the SD. There should be some error in nova logs if you end up with 2 servers with one having no interfaces..I assume something to do with interface detach/attach08:41
openstackgerritXiaojueGuan proposed openstack/heat-dashboard master: Update links in README  https://review.openstack.org/57464908:42
ricolinramishra, right, forgot it's a create and delete case08:43
strigaziricolin: with explicit deps it works08:45
strigaziricolin: ramishra One more, is it possible to delete the first server before booting the new one08:46
csatariIs there any hard limit for the number of resources in a stack?08:46
strigazicsatari: a stack might have other stacks, so you can heat that limit, stack_per_tenant. I have hit this in the past08:47
strigazicsatari: also there is this in hea.conf max_resources_per_stack08:48
strigazi*heat.conf08:48
csatariThanks strigazi. Is there a maximum number I can set there?08:49
openstackgerritXiaojueGuan proposed openstack/heat-cfntools master: Update links in README  https://review.openstack.org/57465308:49
strigazicsatari: -1 :) means no limit08:49
strigazicsatari: from my experience with many stacks and resources the limit is the number of db connections08:50
strigazicsatari: still haven't figured that yet. You can remove the limits in heat but then heat will hammer the db08:51
csataristrigazi Okay :) We have some performance issues with really big stacks (more than 1500 resources).08:51
ricolincsatari, performance of heat service or performance in your other part of environment?08:53
csatariAnd we would like to figure out if 1) there is a maximum number what is tested and proved to work 2) we have some configuration issue somewhere causing the performance issues.08:53
strigazicsatari: we too, but when we have many parallel stack creations, we try to configure the db pool properly08:53
ramishrastrigazi: when a resource is replaced, we create the replacement and then delete the old one. I don't remember anyway you can work around that, what's your issue, capacity?08:53
csatariPerformance of heat service. Getting the resource list ends up in timeout kind of things.08:53
*** gfidente has joined #heat08:54
*** gfidente has quit IRC08:54
*** gfidente has joined #heat08:54
strigaziramishra one issue can be capacity, but the other issue is that two server with the same name are not working in 1) opemstack kubernetes cloud-provider08:54
*** AlexeyAbashkin has quit IRC08:54
strigazi2) cern or any cloud that the nova name is resolvable name08:55
ricolinstrigazi, I don't think it's a explicit/implicit problem, maybe worth to investigate on that environment08:55
*** priteau has joined #heat08:55
ricolinstrigazi, and if it's a explicit or implicit issue, means we need to fix it08:56
ricolincsatari, which release you're using?08:56
strigaziricolin: ramishra can I do anything to avoid having two servers with the same name at the same time?08:57
*** test has joined #heat08:58
*** test has quit IRC08:58
*** AlexeyAbashkin has joined #heat08:58
ricolinstrigazi, that's a protection for resource update, so if you don't want it, I guess you can only do it without updating stack09:00
*** ZoltanKo has joined #heat09:00
ramishrastrigazi: I mean you don't use the name and it's auto generated. but if it's resolvable names that you want to use for replaced server may be you can update it later09:00
strigaziramishra ricolin the name we set it explicitly, - [{ get_param: 'OS::stack_name' }, 'master', '%index%']09:03
*** dbecker has quit IRC09:11
ricolinstrigazi, if server replace complete one of the two servers will be removed, wondering why it stuck in update in progress09:11
*** dbecker has joined #heat09:11
*** nicolasbock has joined #heat09:18
openstackgerritKazunori Shinohara proposed openstack/heat master: Add Blazar support to Heat  https://review.openstack.org/53457609:29
*** Filippos has joined #heat09:30
*** flwang1 has joined #heat09:30
*** Christine has joined #heat09:30
*** Christine is now known as Guest4135509:31
*** Guest41355 has quit IRC09:33
Filipposhello09:35
*** nicolasbock has quit IRC09:37
*** Filippos has quit IRC09:37
*** Alexey_Abashkin has joined #heat09:40
*** AlexeyAbashkin has quit IRC09:40
*** Alexey_Abashkin is now known as AlexeyAbashkin09:40
*** pbourke has quit IRC09:43
*** pbourke has joined #heat09:43
*** ricolin has quit IRC09:57
*** lifeless has quit IRC09:59
*** AlexeyAbashkin has quit IRC10:28
*** pengdake has quit IRC10:30
*** mdnadeem has quit IRC10:51
*** gfidente has quit IRC11:02
*** ysandeep_ has quit IRC11:03
*** lifeless has joined #heat11:03
*** lifeless has quit IRC11:13
*** lifeless has joined #heat11:13
openstackgerritchenxiangui proposed openstack/heat master: Add wiki and release notes link to README  https://review.openstack.org/57469611:14
*** pengdake has joined #heat11:18
openstackgerritPavlo Shchelokovskyy proposed openstack/heat master: Fix versions Controller for Py3  https://review.openstack.org/57433511:28
openstackgerritPavlo Shchelokovskyy proposed openstack/heat master: Restore auth-less version negotiation  https://review.openstack.org/57469711:28
openstackgerritPavlo Shchelokovskyy proposed openstack/heat master: Use application_url in API version document  https://review.openstack.org/57469811:28
openstackgerritPavlo Shchelokovskyy proposed openstack/heat master: Add functional test for version negotiation  https://review.openstack.org/57469911:28
pas-hazaneb: ^ for version negotiation.. could've squashed them... but seems more atomic this way11:29
*** nicolasbock has joined #heat11:33
*** gfidente has joined #heat11:39
*** gfidente has quit IRC11:39
*** gfidente has joined #heat11:39
*** AlexeyAbashkin has joined #heat11:45
*** Alexey_Abashkin has joined #heat11:48
*** AlexeyAbashkin has quit IRC11:50
*** Alexey_Abashkin is now known as AlexeyAbashkin11:50
*** AlexeyAbashkin has quit IRC11:52
*** AlexeyAbashkin has joined #heat11:53
*** edmondsw has joined #heat12:03
*** pengdake has quit IRC12:46
*** pengdake has joined #heat12:46
*** dyasny has quit IRC12:47
*** dyasny has joined #heat12:48
*** skramaja has quit IRC13:13
*** felipemonteiro_ has joined #heat13:23
*** AlexeyAbashkin has quit IRC13:24
*** pengdake has quit IRC13:26
*** mdnadeem has joined #heat13:28
*** felipemonteiro_ has quit IRC13:29
*** AlexeyAbashkin has joined #heat13:32
*** ramishra has quit IRC14:32
*** crose has joined #heat14:36
*** hongbin has joined #heat14:43
*** serlex has quit IRC14:49
*** mdnadeem has quit IRC14:49
*** mordred has quit IRC14:55
*** jmlowe has joined #heat14:57
zanebpas-ha: I take it http_proxy_to_wsgi has to go in front of versionnegotiation?14:58
pas-haI think so, as it rewrites the headers and we need them to present a correct URL14:59
pas-hae.g. when SSL was properly terminated (with added Proto headers) on a service proxying to heat15:00
zanebmaybe we should just get rid of all the debug logs from the versionnegotiation middleware, since that's the one that is logging with the wrong request ID15:05
zanebhttp://logs.openstack.org/97/574697/1/check/heat-functional-convg-mysql-lbaasv2/146c952/logs/screen-h-api.txt.gz#_Jun_12_12_20_54_60012515:06
openstackgerritAkihiro Motoki proposed openstack/heat-dashboard master: Add python3 django 1.11 job instead of django 2.0 job  https://review.openstack.org/57476215:06
zaneb(the [-] in that ^ link, because this is the first request and it doesn't have an id yet)15:06
*** dyasny has quit IRC15:07
*** dyasny has joined #heat15:08
*** mordred has joined #heat15:09
*** gfidente has quit IRC15:20
*** kiennt26 has joined #heat15:22
*** dyasny_ has joined #heat15:31
*** dyasny has quit IRC15:33
openstackgerritHarald Jensås proposed openstack/heat master: OS::Nova::Server: Extend addresses attr to include subnets  https://review.openstack.org/56896015:35
*** jtomasek is now known as jtomasek|bbl15:37
*** dyasny__ has joined #heat15:43
*** dyasny_ has quit IRC15:45
*** kiennt26 has quit IRC15:47
*** mordred has quit IRC15:48
*** ZoltanKo has quit IRC15:49
openstackgerritHarald Jensås proposed openstack/heat master: Allow updating the segment property of OS::Neutron::Subnet  https://review.openstack.org/56720615:51
*** mordred has joined #heat15:52
openstackgerritHarald Jensås proposed openstack/heat master: OS::Nova::Server: Extend addresses attr to include network  https://review.openstack.org/56896115:54
openstackgerritHarald Jensås proposed openstack/heat master: Add segments attribute to OS::Neutron::Net  https://review.openstack.org/56720715:55
*** AlexeyAbashkin has quit IRC15:57
*** AlexeyAbashkin has joined #heat16:01
*** hoonetorg has quit IRC16:02
*** jnettlet[m] has quit IRC16:08
*** jmlowe has quit IRC16:13
*** shardy has quit IRC16:14
*** cliffparsons has joined #heat16:19
*** hoonetorg has joined #heat16:23
*** Alexey_Abashkin has joined #heat16:33
*** Alexey_Abashkin has quit IRC16:35
*** AlexeyAbashkin has quit IRC16:35
*** jnettlet[m] has joined #heat16:36
*** flwang1 has quit IRC16:59
*** derekh has quit IRC17:00
*** hoonetorg has quit IRC17:03
*** gmoro_ has quit IRC17:05
*** flwang1 has joined #heat17:06
*** hoonetorg has joined #heat17:11
*** lifeless_ has joined #heat17:15
*** lifeless has quit IRC17:16
openstackgerritPavlo Shchelokovskyy proposed openstack/heat master: Run integration tests in parallel  https://review.openstack.org/57482817:18
*** hoonetorg has quit IRC17:19
pas-hain other news, it seems the last commit to heat-tempest-plugin kind sporadically breaks CI due to races with other tests / overoptimistic test logic , for now attempting to just run tests in parallel ^17:26
openstackgerritPavlo Shchelokovskyy proposed openstack/heat master: Run integration tests serially  https://review.openstack.org/57482817:26
pas-hawrong commit title :-)17:27
*** hoonetorg has joined #heat17:28
*** ysandeep has quit IRC17:30
*** hoonetorg has quit IRC17:47
*** hoonetorg has joined #heat17:49
*** priteau has quit IRC17:53
*** hoonetorg has quit IRC17:58
*** ysandeep has joined #heat18:12
*** hoonetorg has joined #heat18:16
*** hoonetorg has quit IRC18:21
*** spzala has joined #heat18:22
*** spzala has quit IRC18:22
*** ysandeep has quit IRC18:26
*** jtomasek|bbl is now known as jtomasek18:37
*** hoonetorg has joined #heat18:38
*** hoonetorg has quit IRC18:45
zanebpas-ha: I just approved ramishra's fix for it18:47
*** hoonetorg has joined #heat18:52
*** crose has quit IRC18:53
*** dyasny__ has quit IRC18:56
*** hoonetorg has quit IRC18:58
*** dyasny has joined #heat19:03
*** hoonetorg has joined #heat19:11
*** hoonetorg has quit IRC19:15
*** hoonetorg has joined #heat19:17
*** flwang1 has quit IRC19:25
*** hoonetorg has quit IRC19:30
*** lifeless_ has quit IRC19:31
*** lifeless has joined #heat19:49
*** AlexeyAbashkin has joined #heat20:05
*** Alexey_Abashkin has joined #heat20:08
*** AlexeyAbashkin has quit IRC20:09
*** Alexey_Abashkin is now known as AlexeyAbashkin20:09
*** Alexey_Abashkin has joined #heat20:15
*** AlexeyAbashkin has quit IRC20:16
*** Alexey_Abashkin is now known as AlexeyAbashkin20:16
openstackgerritMerged openstack/heat-tempest-plugin master: Fix software config/deployment api tests  https://review.openstack.org/57454220:26
*** gkadam has quit IRC20:42
*** boris_42_ has joined #heat20:47
*** jnettlet[m] has quit IRC20:49
*** AlexeyAbashkin has quit IRC20:52
*** flwang1 has joined #heat21:07
*** jnettlet[m] has joined #heat21:09
*** lifeless has quit IRC21:21
*** lifeless has joined #heat21:29
*** nicolasbock has quit IRC21:51
*** blomquisg is now known as blomquisg_gone22:20
*** edmondsw has quit IRC22:26
*** edmondsw has joined #heat22:26
*** edmondsw has quit IRC22:31
*** rcernin has joined #heat22:44
*** jmlowe has joined #heat22:55
*** hongbin has quit IRC23:15
*** jnettlet[m] has quit IRC23:16
*** jmlowe has quit IRC23:17
*** jmlowe has joined #heat23:18
*** jmlowe has quit IRC23:31
*** jmlowe has joined #heat23:35
*** jnettlet[m] has joined #heat23:36
*** mchlumsky has quit IRC23:59

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!