| openstackgerrit | Ghanshyam Mann proposed openstack/senlin: Fix Senlin tempest plugin https://review.openstack.org/412658 | 00:17 |
|---|---|---|
| *** guoshan has joined #senlin | 00:27 | |
| *** catintheroof has quit IRC | 00:29 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/senlin: Fix Senlin tempest plugin https://review.openstack.org/412658 | 00:30 |
| *** XueFeng has quit IRC | 00:42 | |
| *** XueFeng has joined #senlin | 00:43 | |
| *** xiangxinyong is now known as edisonxiang | 01:02 | |
| *** XueFeng has quit IRC | 01:03 | |
| *** guoshan has quit IRC | 01:05 | |
| *** XueFeng has joined #senlin | 01:06 | |
| *** guoshan has joined #senlin | 01:26 | |
| *** yanyanhu has joined #senlin | 01:34 | |
| openstackgerrit | Ha Van Tu proposed openstack/senlin: Fix typo in examples https://review.openstack.org/412699 | 02:24 |
| *** dinghh1 has joined #senlin | 02:55 | |
| *** dinghh has quit IRC | 02:56 | |
| *** chohoor has joined #senlin | 03:09 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/senlin: DNM: just testing https://review.openstack.org/412722 | 03:13 |
| openstackgerrit | Ghanshyam Mann proposed openstack/senlin: DNM: just testing https://review.openstack.org/412722 | 03:14 |
| openstackgerrit | RUIJIE YUAN proposed openstack/senlin: rework profile-get2/list2 https://review.openstack.org/412725 | 03:35 |
| openstackgerrit | Shan Guo proposed openstack/senlin: Add parameter in CLUSTER_DEL_NODES action https://review.openstack.org/412729 | 03:42 |
| *** guoshan has quit IRC | 04:46 | |
| *** gmann has joined #senlin | 05:24 | |
| gmann | Qiming: ping, anything changes for APi permission as it failing - https://review.openstack.org/#/c/412658/2 | 05:24 |
| Qiming | gmann, nothing I'm aware of | 05:36 |
| Qiming | yanyanhu, can you help look into this as well? | 05:37 |
| Qiming | recent changes at tempest side is breaking our api/functional/integration tests ... | 05:37 |
| gmann | Qiming: yea, but with fix it gives 403 error even on tempest revet patch also - https://review.openstack.org/#/c/412722/ | 05:40 |
| yanyanhu | Qiming, sure, will check it | 05:40 |
| gmann | even it works on my local devstack setup | 05:40 |
| gmann | yanyanhu: this also have same 403 issue. this is depends-on tempest revert one - https://review.openstack.org/#/c/412722/ | 05:41 |
| yanyanhu | gmann, ok, let me have a look | 05:42 |
| *** guoshan has joined #senlin | 05:46 | |
| *** guoshan has quit IRC | 05:51 | |
| yanyanhu | gmann, Qiming, that is weird, senlin-api service can not bind to port 8778 when it is started | 05:51 |
| yanyanhu | http://logs.openstack.org/22/412722/2/check/gate-senlin-dsvm-tempest-api/db47999/logs/screen-sl-api.txt.gz | 05:51 |
| yanyanhu | it guess this error is not related to change on tempest | 05:51 |
| yanyanhu | let me check the test log of other failed patches | 05:52 |
| Qiming | okay | 05:52 |
| gmann | this too same issue - http://logs.openstack.org/58/412658/2/check/gate-senlin-dsvm-tempest-api/03d369c/logs/screen-sl-api.txt.gz | 05:54 |
| yanyanhu | looks like 8778 port is occupied by another service. that why 403 exception was raised | 05:57 |
| yanyanhu | and when this patch was proposed and verified in gate, this error haven't happened yet | 05:58 |
| yanyanhu | https://review.openstack.org/#/c/412460/ | 05:58 |
| gmann | yea placement one - https://github.com/openstack-dev/devstack/blob/51a225c5d7f7d8655be62cab0e4d677b6839f35a/lib/placement#L50 | 05:59 |
| yanyanhu | ah | 06:00 |
| yanyanhu | I guess we need a better way to manage those ports consumed by different services :) | 06:01 |
| gmann | and this one was enabled for multinode here - https://review.openstack.org/#/c/411510/ | 06:01 |
| gmann | yanyanhu: i think we have but not for all services | 06:01 |
| gmann | let me find wiki i looked last week | 06:01 |
| yanyanhu | gmann, so any way to handle this conflict? | 06:01 |
| yanyanhu | gmann, thanks :) | 06:01 |
| Qiming | it is at gate side ... | 06:02 |
| gmann | http://docs.openstack.org/newton/config-reference/firewalls-default-ports.html | 06:03 |
| gmann | but not upto date seems :) | 06:03 |
| gmann | it should be via this wiki | 06:03 |
| gmann | let me talk to sean on this | 06:03 |
| yanyanhu | gmann, yes... it not uptodate | 06:03 |
| yanyanhu | gmann, great, thank you so much, please just leave us message if there is any action we need to take | 06:04 |
| gmann | may be you can add for senlin and then palcement API we can change as it was later one | 06:04 |
| gmann | sure | 06:04 |
| yanyanhu | gmann, yes, will try update the wiki | 06:04 |
| gmann | let me change placement API port on devstack | 06:05 |
| yanyanhu | thanks | 06:05 |
| *** guoshan has joined #senlin | 06:15 | |
| Qiming | alright, gmann is right, it looks like placement service is 'robbing' port 8778 from us, :) | 06:30 |
| Qiming | http://logs.openstack.org/49/412449/1/check/gate-senlin-dsvm-tempest-api/e84280d/logs/devstacklog.txt.gz#_2016-12-19_13_16_45_139 | 06:30 |
| gmann | Qiming: yea, changing here hope 8780 not being used - https://review.openstack.org/#/c/412770/ | 06:31 |
| Qiming | I'm proposing a patch to the openstack-manuals project so that people will know that 8778 is used | 06:37 |
| yanyanhu | Qiming, great | 06:38 |
| yanyanhu | I'm trying download the openstack-manual repo, it has costed 30mins and it still in progress... | 06:38 |
| Qiming | it is BIG | 06:38 |
| yanyanhu | yes... need another 30mins I guess | 06:39 |
| *** dinghh has joined #senlin | 07:11 | |
| *** dinghh1 has quit IRC | 07:13 | |
| openstackgerrit | miaohb proposed openstack/senlin: Trivial: Arrange the imported objects in alphabetical order https://review.openstack.org/412843 | 07:43 |
| openstackgerrit | Shan Guo proposed openstack/senlin: Add parameter in CLUSTER_DEL_NODES action https://review.openstack.org/412729 | 08:12 |
| openstackgerrit | Shan Guo proposed openstack/senlin: Add parameter in CLUSTER_DEL_NODES action https://review.openstack.org/412729 | 08:13 |
| openstackgerrit | Shan Guo proposed openstack/senlin: Add parameter in CLUSTER_DEL_NODES action https://review.openstack.org/412729 | 08:15 |
| *** shu-mutou has joined #senlin | 08:34 | |
| yanyanhu | hi, Qiming, around? | 08:37 |
| yanyanhu | do you know what is placement service for? Didn't find this project in project list or git.openstack.org. Is it a sub proj of Nova? | 08:39 |
| openstackgerrit | miaohb proposed openstack/senlin: Trivial: Arrange the imported objects in alphabetical order https://review.openstack.org/412843 | 08:39 |
| Qiming | not quite sure | 08:40 |
| Qiming | could be an extension of nova | 08:40 |
| shu-mutou | Hi! Does API for updating resources register timestamp into updated_at? | 08:43 |
| yanyanhu | I saw -1 Sylvain Bauza | 08:44 |
| yanyanhu | hi, shu-mutou, I think so | 08:44 |
| Qiming | see this: https://github.com/openstack-dev/devstack/blob/51a225c5d7f7d8655be62cab0e4d677b6839f35a/lib/placement#L50 | 08:44 |
| yanyanhu | the timestamp property of updated resource should be refreshed | 08:44 |
| Qiming | in the first few lines | 08:44 |
| yanyanhu | Qiming, yes, but still not quite sure what it is for | 08:45 |
| yanyanhu | I saw Sylvain Bauza's comment in that patch. Anyway, we can consider to disable placement service if we don't need it in our test. Although I feel unhappy that a service is enabled in devstack core lib and breaks our job without any hint. | 08:45 |
| shu-mutou | yanyanhu: thanks!! | 08:46 |
| yanyanhu | shu-mutou, you're welcome :) | 08:46 |
| Qiming | something related to this: http://specs.openstack.org/openstack/nova-specs/specs/newton/implemented/generic-resource-pools.html | 08:47 |
| yanyanhu | will wait for more feedback on that patch | 08:47 |
| gmann | Qiming: yanyanhu can you join openstack-dev and ping bauzas for port discussion | 08:56 |
| yanyanhu | gmann, sure | 08:56 |
| openstackgerrit | Sylvain Bauza proposed openstack/senlin: DO NOT MERGE: Testing the gate https://review.openstack.org/412930 | 09:03 |
| *** xuhaiwei has joined #senlin | 09:04 | |
| openstackgerrit | yunfeng zhou proposed openstack/senlin-dashboard: add CONTRIBUTING.rst https://review.openstack.org/412966 | 09:22 |
| openstackgerrit | Shu Muto proposed openstack/senlin-dashboard: Add node create/update actions https://review.openstack.org/396975 | 09:36 |
| *** shu-mutou is now known as shu-mutou-AWAY | 09:44 | |
| *** yanyanhu has quit IRC | 09:54 | |
| *** guoshan has quit IRC | 10:24 | |
| *** chohoor has quit IRC | 11:00 | |
| *** guoshan has joined #senlin | 11:25 | |
| *** guoshan has quit IRC | 11:29 | |
| openstackgerrit | miaohb proposed openstack/senlin: [WIP] Versioned request and engine support for credential_update https://review.openstack.org/413050 | 12:14 |
| *** guoshan has joined #senlin | 12:26 | |
| *** guoshan has quit IRC | 12:30 | |
| *** catintheroof has joined #senlin | 12:37 | |
| *** yanyanhu has joined #senlin | 12:57 | |
| yanyanhu | hi, guys, meeting will start in openstack-meeting channel in minutes | 12:58 |
| yanyanhu | hi, meeting has started | 13:00 |
| *** elynn has joined #senlin | 13:00 | |
| *** lxinhui has joined #senlin | 13:13 | |
| *** yanyan has joined #senlin | 13:23 | |
| yanyan | hello, am I still online? | 13:24 |
| *** yanyanhu has quit IRC | 13:26 | |
| *** guoshan has joined #senlin | 13:26 | |
| *** guoshan has quit IRC | 13:31 | |
| *** Ruijie_ has joined #senlin | 14:00 | |
| yanyan | hi, just a quick update, our gate is broken.... | 14:01 |
| yanyan | from this morning I think | 14:01 |
| Qiming | last night | 14:01 |
| Qiming | 9pm or so | 14:01 |
| yanyan | yea, last night | 14:01 |
| Qiming | before some patches and after some patches | 14:01 |
| yanyan | that is because a new added service "placement" is enabled by default in devstack lib | 14:02 |
| Qiming | suddenly ... tempest jobs all fail | 14:02 |
| elynn | placement is what... | 14:02 |
| yanyan | and it uses 8778 port as senlin-api service | 14:02 |
| yanyan | elynn, it part of nova currently | 14:02 |
| Qiming | hopefully, it will be all okay again tomorrow | 14:03 |
| Qiming | [21:39] <sdague> bauzas: ok, cool, just let me know when you have stuff up for review, will ensure we get that in today | 14:03 |
| yanyan | and will be splitted as an invidual API in future based on my understanding | 14:03 |
| lxinhui | Qiming, yanyan | 14:03 |
| yanyan | Qiming, yes, hope it will receover tomorrow | 14:03 |
| Qiming | [21:40] <bauzas> I need to cut https://github.com/openstack-dev/devstack/blob/master/files/apache-placement-api.template#L1-L16 and put some directives in the location directive itself | 14:03 |
| Qiming | [21:40] <jroll> ah, I assumed if you only had the location directive it wouldn't be needed | 14:03 |
| Qiming | [21:40] <bauzas> yeah, we're honestly blocking senlin | 14:03 |
| lxinhui | please help to explain nowdays | 14:03 |
| lxinhui | senlin capcaity management fucntions | 14:03 |
| elynn | okay too many projects flying in openstack world now | 14:03 |
| yanyan | https://review.openstack.org/#/c/412770/ | 14:03 |
| Qiming | yes, they are working hard on that | 14:03 |
| lxinhui | I remember we need to define the numbers | 14:04 |
| yanyan | https://review.openstack.org/#/c/412915/ | 14:04 |
| Qiming | placement is a boy from nova | 14:04 |
| lxinhui | about the | 14:04 |
| lxinhui | basic | 14:04 |
| yanyan | here are the efforts to address this issue | 14:04 |
| lxinhui | cluster size | 14:04 |
| lxinhui | or something | 14:04 |
| lxinhui | max, min | 14:04 |
| lxinhui | desired | 14:04 |
| Qiming | yes? desired_capacity? | 14:04 |
| lxinhui | yes | 14:04 |
| lxinhui | desired_capacity? | 14:05 |
| lxinhui | please share more about the design of desired_capacity | 14:05 |
| *** zzxwill has joined #senlin | 14:05 | |
| yanyan | lxinhui, we do make a little change on it actually | 14:05 |
| lxinhui | that means the first size | 14:05 |
| yanyan | about the conception of "desired" capacity | 14:05 |
| lxinhui | when create | 14:05 |
| yanyan | maybe Qiming can give you a more clear and accurate definition on it | 14:06 |
| *** catinthe_ has joined #senlin | 14:06 | |
| lxinhui | can I use it for reserve purpose? | 14:06 |
| lxinhui | :) | 14:06 |
| yanyan | we discussed on it a lot before reaching consensus | 14:06 |
| Qiming | yes, later, it will be increased or decreased when you scale a cluster | 14:06 |
| lxinhui | Can I use senlin to reserve some reasurces for some tenant? | 14:07 |
| Qiming | no, xinhui, its current meaning is: user desired capacity of a cluster, which may and may not be the actual capacity | 14:07 |
| Qiming | yes, you can create a separate cluster | 14:07 |
| lxinhui | ... | 14:07 |
| lxinhui | oh | 14:07 |
| Qiming | senlin gives you a tool, we don't mandate on the way you use it | 14:08 |
| Qiming | we considered that use case of reservation | 14:08 |
| lxinhui | but little bit upgly | 14:08 |
| lxinhui | ugly | 14:08 |
| Qiming | you can create a standby cluster | 14:08 |
| lxinhui | yes, I see | 14:08 |
| Qiming | shared by many active clusters | 14:08 |
| yanyan | yes, as Qiming said, that all depends on how you use it | 14:08 |
| *** catintheroof has quit IRC | 14:08 | |
| lxinhui | but you know for the resource pool management purpose | 14:08 |
| yanyan | how you use senlin to manage resource | 14:08 |
| Qiming | when you want to scale out one of them, you can migrate one out of the standby cluster | 14:09 |
| lxinhui | there are several numbers user need to set | 14:09 |
| lxinhui | one the max | 14:09 |
| lxinhui | the other is the mini | 14:09 |
| yanyan | we provides abundant features for cluster membership management | 14:09 |
| lxinhui | the third one is the resesered | 14:09 |
| lxinhui | reserved | 14:09 |
| Qiming | yes, you can omit them | 14:09 |
| Qiming | the 3rd is desired_capacity | 14:09 |
| lxinhui | the reserved should be less than the max | 14:09 |
| Qiming | reserved will become the 4th | 14:10 |
| lxinhui | and larger than mini | 14:10 |
| lxinhui | but have keep the resource can not be used by others | 14:10 |
| Qiming | why should I reserve some nodes in a cluster if I'm not using them | 14:10 |
| lxinhui | I say standy solution is ugly is becuase | 14:10 |
| Qiming | for reservation, you can create a standby one | 14:10 |
| yanyan | lxinhui, I think we need to define "reserve" first :) | 14:11 |
| lxinhui | it waste much before someone really wanna it | 14:11 |
| Qiming | if it is shared, it may be not a waste | 14:11 |
| Qiming | on the contrary, if everyone "reserves" some capacity by themselves, that is a huge waste | 14:12 |
| yanyan | so I feel we just need to ensure users can get the resource promptly when they need it | 14:13 |
| lxinhui | that is what I mean | 14:13 |
| yanyan | but that is more about policy I feel | 14:13 |
| lxinhui | yanyan | 14:13 |
| lxinhui | that is not to find a solution | 14:14 |
| lxinhui | but wanna senlin to fill in some gap | 14:14 |
| yanyan | lxinhui, yes, agree | 14:14 |
| lxinhui | instead of just some mitigate | 14:14 |
| yanyan | we need to provide enough flexibility for users to support their resource management policy | 14:14 |
| yanyan | but we won't make decision for user | 14:15 |
| yanyan | about how to make that reservation | 14:15 |
| lxinhui | we do not | 14:15 |
| yanyan | yea | 14:15 |
| lxinhui | actaully we can not | 14:15 |
| yanyan | we let them to decide the strategy | 14:15 |
| lxinhui | provide support for waht I am discussed | 14:15 |
| lxinhui | not strategy | 14:15 |
| Qiming | tell me, if we have min=1, max=10, desired=5, reserve=7, current=6 | 14:16 |
| lxinhui | but resoruce calculation instead of real provisioing them | 14:16 |
| Qiming | if I'm scaling out the cluster by 2 nodes | 14:16 |
| yanyan | lxinhui, resource calculation means? | 14:16 |
| lxinhui | keep the source | 14:16 |
| Qiming | how should I calculate the new number? | 14:16 |
| lxinhui | for the user | 14:16 |
| lxinhui | instead of really create the vm | 14:16 |
| Qiming | the way to keep resource is to raise the 'min' value | 14:16 |
| lxinhui | Qiming, easy anwser, 7 can be ensured | 14:17 |
| lxinhui | but 8 can fail | 14:17 |
| Qiming | then you mean, max is useless? | 14:17 |
| lxinhui | that is max | 14:17 |
| lxinhui | just allow or not | 14:18 |
| lxinhui | but not ensure or not | 14:18 |
| Qiming | we cannot reserve without create ... | 14:18 |
| lxinhui | since virtual resource allow to over claim | 14:18 |
| lxinhui | Qiming, I see | 14:18 |
| yanyan | imo, occupying resource is the only way to really reserve resource :) | 14:18 |
| yanyan | reserve the word itself means "it is only for you" | 14:18 |
| yanyan | :) | 14:18 |
| Qiming | there have been some service proposal on that a long time ago | 14:19 |
| lxinhui | Yanyan, please refer to above mention about poor points of real creation | 14:19 |
| yanyan | otherwise, you can only say, we have higher priority to use it :) | 14:19 |
| Qiming | it is named climate or something | 14:19 |
| Qiming | doing resource reservation | 14:19 |
| lxinhui | why it failed | 14:19 |
| Qiming | and their have been proposals across community about quota | 14:20 |
| *** Ruijie__ has joined #senlin | 14:20 | |
| lxinhui | for the limits of resource? | 14:20 |
| Qiming | no one works on it | 14:20 |
| Qiming | yes | 14:20 |
| lxinhui | ha | 14:20 |
| lxinhui | or else | 14:20 |
| yanyan | poor points, you mean? | 14:20 |
| lxinhui | everyone still wanna vsphere | 14:20 |
| lxinhui | ... | 14:20 |
| Qiming | everyone brings in new idea and say: this is important, this is better | 14:20 |
| yanyan | disadvantage? | 14:20 |
| Qiming | then resources distracted | 14:20 |
| lxinhui | can not use senlin | 14:20 |
| Qiming | you can see projects blooming | 14:21 |
| lxinhui | but use vsphere directly | 14:21 |
| yanyan | resource redundancy, is one important prerequisite of HA :) | 14:21 |
| lxinhui | consider this in another way | 14:21 |
| Qiming | vsphere is good | 14:21 |
| lxinhui | yes, yanyan | 14:21 |
| lxinhui | exactly | 14:21 |
| Qiming | you got control of everything | 14:21 |
| lxinhui | many user rely on this for similar reasons | 14:21 |
| Qiming | openstack is not | 14:21 |
| lxinhui | I am asking for everything :) | 14:22 |
| yanyan | :) | 14:22 |
| lxinhui | I am not | 14:22 |
| lxinhui | I mean | 14:22 |
| lxinhui | just reserve | 14:22 |
| lxinhui | okay | 14:22 |
| lxinhui | Anyway, Thanks, Qiming and Yanyan | 14:22 |
| yanyan | lxinhui, you're welcome | 14:22 |
| Qiming | I mean vSphere can control everything | 14:22 |
| *** Ruijie_ has quit IRC | 14:22 | |
| lxinhui | I will read qiming's code and add more in the next week | 14:23 |
| yanyan | alway pleased to discuss :) | 14:23 |
| Qiming | but in OpenStack, no service can do that | 14:23 |
| lxinhui | Qiming, actaully not | 14:23 |
| Qiming | it is run by a single company | 14:23 |
| lxinhui | you need think it in a bigger stack | 14:23 |
| yanyan | hi, I have to leave. see u guys tomorrow. have a good night | 14:23 |
| lxinhui | so what I am asking actally is a kind of service | 14:23 |
| Qiming | you won't come up with two monitoring solutions: ceilometer and monasca | 14:23 |
| lxinhui | even only API | 14:23 |
| *** yanyan has quit IRC | 14:23 | |
| Qiming | you won't waste time on wsme, flask, pecan, wsgi all these variants | 14:24 |
| lxinhui | nono | 14:24 |
| lxinhui | what you are talking just naughty argue | 14:24 |
| lxinhui | We can find other time | 14:24 |
| Qiming | it IS a BIG problem of the community | 14:24 |
| lxinhui | discuss about API things | 14:24 |
| lxinhui | see you, have to run for another meeting | 14:25 |
| Qiming | ceilometer, when returns version info, does tthis: {'versions': {'values': []}} | 14:25 |
| lxinhui | it almost ending | 14:25 |
| lxinhui | ... | 14:25 |
| lxinhui | bye see you | 14:25 |
| *** lxinhui has quit IRC | 14:25 | |
| *** guoshan has joined #senlin | 14:27 | |
| *** Ruijie__ has quit IRC | 14:28 | |
| *** Ruijie__ has joined #senlin | 14:28 | |
| *** guoshan has quit IRC | 14:31 | |
| *** zzxwill has quit IRC | 14:32 | |
| *** Ruijie_ has joined #senlin | 14:50 | |
| *** Ruijie__ has quit IRC | 14:50 | |
| *** Ruijie_ has quit IRC | 15:00 | |
| *** zhouzhengxi has joined #senlin | 15:23 | |
| *** guoshan has joined #senlin | 15:28 | |
| *** guoshan has quit IRC | 15:32 | |
| *** zhouzhengxi has quit IRC | 15:58 | |
| *** Drago1 has joined #senlin | 16:22 | |
| *** guoshan has joined #senlin | 16:29 | |
| *** guoshan has quit IRC | 16:33 | |
| *** catintheroof has joined #senlin | 16:39 | |
| *** catinthe_ has quit IRC | 16:42 | |
| *** Drago1 has quit IRC | 17:01 | |
| *** guoshan has joined #senlin | 17:29 | |
| *** guoshan has quit IRC | 17:34 | |
| *** catinthe_ has joined #senlin | 18:03 | |
| *** catintheroof has quit IRC | 18:06 | |
| *** guoshan has joined #senlin | 18:30 | |
| *** guoshan has quit IRC | 18:35 | |
| *** elynn has quit IRC | 19:01 | |
| *** elynn has joined #senlin | 19:02 | |
| *** guoshan has joined #senlin | 19:31 | |
| *** guoshan has quit IRC | 19:35 | |
| *** catintheroof has joined #senlin | 19:37 | |
| *** catinthe_ has quit IRC | 19:40 | |
| *** guoshan has joined #senlin | 20:32 | |
| *** guoshan has quit IRC | 20:37 | |
| *** catintheroof has quit IRC | 20:47 | |
| *** Drago1 has joined #senlin | 21:32 | |
| *** guoshan has joined #senlin | 21:32 | |
| *** guoshan has quit IRC | 21:37 | |
| *** guoshan has joined #senlin | 22:33 | |
| *** guoshan has quit IRC | 22:38 | |
| *** guoshan has joined #senlin | 23:34 | |
| *** guoshan has quit IRC | 23:39 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!