| *** lixinhui_ has quit IRC | 00:02 | |
| *** lixinhui_ has joined #senlin | 00:10 | |
| *** pratikma_ has quit IRC | 00:12 | |
| *** lixinhui_ has quit IRC | 00:32 | |
| *** xuhaiwei has quit IRC | 01:02 | |
| *** zhenguo has joined #senlin | 01:23 | |
| *** jruano has quit IRC | 01:32 | |
| *** xuhaiwei has joined #senlin | 01:32 | |
| *** Yanyanhu has joined #senlin | 01:38 | |
| *** pratikmallya has joined #senlin | 01:47 | |
| Yanyanhu | morning, guys | 01:56 |
|---|---|---|
| xuhaiwei | morning | 01:57 |
| *** yuanying_ has joined #senlin | 02:02 | |
| *** yuanying has quit IRC | 02:05 | |
| *** pratikmallya has quit IRC | 02:11 | |
| *** yuanying_ has quit IRC | 02:11 | |
| *** elynn has joined #senlin | 02:13 | |
| elynn | morning | 02:15 |
| xuhaiwei | morning, elynn | 02:16 |
| elynn | Big news to senlin becoming official project. congratuate you guys! | 02:17 |
| xuhaiwei | the same to you | 02:17 |
| xuhaiwei | about your patch https://review.openstack.org/#/c/239591/, want to ask you about it | 02:17 |
| xuhaiwei | are you free now ? elynn | 02:21 |
| elynn | yes | 02:21 |
| xuhaiwei | is this patch needed by Heat? | 02:21 |
| elynn | Hmm, how to say that... | 02:22 |
| elynn | For now, without this patch will not block contributing senlin resources to heat. | 02:23 |
| elynn | I would say we can just hold on for this patch. | 02:23 |
| xuhaiwei | I am not quite familiar with heat template, for heat template, both ' template: tem.yaml and template: get_file: tem.yaml ' are ok? | 02:24 |
| elynn | No, only the last one will work for heat. | 02:24 |
| xuhaiwei | ok | 02:25 |
| xuhaiwei | so I think maybe senlin should do the same | 02:25 |
| xuhaiwei | currently we support the first style | 02:26 |
| elynn | yes, if without this patch, we can not reference another template file in spec. | 02:26 |
| elynn | The spec used for heat will like this https://github.com/lynic/templates/blob/master/senlin/heat_resource/template/senlin-profile.yaml | 02:27 |
| elynn | The template content need to put in spec file. | 02:27 |
| xuhaiwei | yes | 02:28 |
| xuhaiwei | wait a minute | 02:30 |
| *** pratikmallya has joined #senlin | 02:30 | |
| xuhaiwei | you can do it like this https://github.com/openstack/senlin/blob/master/examples/profiles/heat_stack_random_string.yaml | 02:31 |
| xuhaiwei | elynn | 02:31 |
| elynn | No, for now senlin resource can not support that spec. | 02:32 |
| elynn | I haven't figure out how to support that yet. | 02:33 |
| xuhaiwei | about your patch, it seems both types templates are supported with 'get_file' or without it | 02:34 |
| elynn | senlinclient collect all files when parse this spec https://github.com/openstack/senlin/blob/master/examples/profiles/heat_stack_random_string.yaml , it's not easy to implement similiar codes in heat-engine. Needs heatclient and senlinclient and heat-engine to cooperate. | 02:35 |
| xuhaiwei | yes | 02:37 |
| elynn | Anyway , we can hold on to approve this patch. | 02:43 |
| xuhaiwei | ok, I will try to understand what is different in senlin template and heat template | 02:45 |
| xuhaiwei | thanks elynn | 02:45 |
| elynn | np | 02:47 |
| *** rebase has quit IRC | 02:48 | |
| *** yuanying has joined #senlin | 02:52 | |
| *** lkarm has joined #senlin | 03:16 | |
| openstackgerrit | Julio Ruano proposed openstack/senlin: Update TODO work items https://review.openstack.org/243903 | 03:20 |
| *** lkarm has quit IRC | 03:20 | |
| *** hightall has joined #senlin | 03:22 | |
| *** yuanying has quit IRC | 03:23 | |
| *** pratikmallya has quit IRC | 03:40 | |
| *** hightall has quit IRC | 03:50 | |
| *** shu has joined #senlin | 03:50 | |
| *** yuanying has joined #senlin | 04:07 | |
| openstackgerrit | xu-haiwei proposed openstack/python-senlinclient: Add test case for v1/shell module - part 1 https://review.openstack.org/243407 | 04:10 |
| openstackgerrit | xu-haiwei proposed openstack/python-senlinclient: Add test case for v1/shell module - part 1 https://review.openstack.org/243407 | 04:11 |
| *** gongysh_ has joined #senlin | 04:31 | |
| *** hightall has joined #senlin | 04:39 | |
| *** pratikmallya has joined #senlin | 04:49 | |
| *** jdandrea has quit IRC | 04:53 | |
| *** pratikmallya has quit IRC | 04:55 | |
| openstackgerrit | xu-haiwei proposed openstack/senlin: Fix SDK exception message parse https://review.openstack.org/243959 | 05:33 |
| *** lkarm has joined #senlin | 05:43 | |
| *** lkarm has quit IRC | 05:47 | |
| openstackgerrit | Yanyan Hu proposed openstack/senlin: Add a FEATURE item https://review.openstack.org/243962 | 05:55 |
| *** xuhaiwei has quit IRC | 06:09 | |
| *** lkarm has joined #senlin | 06:15 | |
| openstackgerrit | Yanyan Hu proposed openstack/senlin: Add a FEATURE item https://review.openstack.org/243962 | 06:19 |
| *** lkarm has quit IRC | 06:19 | |
| openstackgerrit | Yanyan Hu proposed openstack/senlin: Add a FEATURE item https://review.openstack.org/243962 | 06:20 |
| elynn | Yanyanhu, just update senlin resource and add attributes, https://github.com/lynic/templates/blob/master/senlin/heat_resource/template/senlin-profile.yaml | 06:26 |
| Yanyanhu | elynn, got it, will check it | 06:26 |
| *** elynn_ has joined #senlin | 06:29 | |
| Yanyanhu | oh, elynn, I suddenly feel maybe we should directly use 'desired_capacity' rather than 'current_size' to name that attribute of cluster | 06:30 |
| *** xuhaiwei has joined #senlin | 06:31 | |
| elynn_ | yes, will change it. | 06:33 |
| *** elynn has quit IRC | 06:33 | |
| Yanyanhu | Thanks. These two concepts are different and we only have a clear definition of 'desired_capacity' now in Senlin. | 06:34 |
| Yanyanhu | we may need further discussion to decide how to define the concept of 'current_size' in Senlin and whether we want to expose it to user | 06:35 |
| Yanyanhu | it should be related to the health management design | 06:36 |
| *** pratikmallya has joined #senlin | 06:39 | |
| elynn_ | the current_size you mean the desired_capacity? | 06:46 |
| elynn_ | What's the difference between them? | 06:46 |
| *** lvdongbing has joined #senlin | 06:50 | |
| openstackgerrit | Merged openstack/senlin: Updated from global requirements https://review.openstack.org/243940 | 06:54 |
| Yanyanhu | hi, elynn_, desired_capacity now has clear definition in Senlin. It refers to the size of cluster that user expected when performing cluster create/scale/resize | 07:01 |
| Yanyanhu | e.g. when desired_capacity is set to 10, user is expecting a cluster composed by 10 health nodes | 07:02 |
| Yanyanhu | but I guess we are still not very sure about the definition of 'current_size' in Senlin. Maybe it means the 'actual number' of health nodes? | 07:04 |
| elynn_ | got it. | 07:05 |
| Yanyanhu | but that really depends on our health management and related policy design | 07:05 |
| elynn_ | let's stick on desired_capacity for now. | 07:05 |
| Yanyanhu | yes, at least we can ensure it is a correct option for cluster attribute definition for now | 07:08 |
| Yanyanhu | maybe in future, after we have clear definition of 'current_size' as well, we can add it :) | 07:08 |
| elynn_ | yes, agree | 07:08 |
| Yanyanhu | before that, you can just treat these two concepts equally | 07:09 |
| openstackgerrit | Shu Muto proposed openstack/senlin-dashboard: Add link to Profile Detail from cluster table https://review.openstack.org/243984 | 07:26 |
| openstackgerrit | Yanyan Hu proposed openstack/senlin: Pick up a TODO item https://review.openstack.org/243987 | 07:27 |
| elynn_ | Hi yanyan, do you know why we add a validation for desired_capacity? https://github.com/openstack/senlin/blob/master/senlin/api/openstack/v1/clusters.py#L74-L76 | 07:31 |
| *** pratikmallya has quit IRC | 07:32 | |
| Yanyanhu | hi, elynn_ just came back from coffee break :) | 07:47 |
| Yanyanhu | let me check it | 07:47 |
| *** Liuqing has joined #senlin | 07:47 | |
| elynn_ | Yanyanhu: thanks , I understand that now. | 07:47 |
| Yanyanhu | ok :) | 07:47 |
| elynn_ | desired_capacity can not be null for cluster create, but can be null for cluster update | 07:48 |
| Yanyanhu | yep | 07:48 |
| elynn_ | Why it can not be null for cluster create? | 07:53 |
| Yanyanhu | elynn_, since we want user to define the cluster size they expect explicitly | 07:54 |
| elynn_ | But in senlin client, if user don't specify desired_capacity, it will replace with 0. That doesn't make sense. | 07:56 |
| Yanyanhu | hmm, you mean there is conflict between these two logics? | 07:57 |
| elynn_ | The behavior of senlinclient and API doesn't consistent. | 07:57 |
| elynn_ | Yes | 07:57 |
| Yanyanhu | I think the default value 0 defined in clientside is just for convinence | 07:58 |
| Yanyanhu | convenience | 07:58 |
| Yanyanhu | I understand your concern | 07:59 |
| Yanyanhu | so you expected an exception from senlinclient when user didn't provide desired_capacity when creating cluster | 07:59 |
| elynn_ | I met a bug, but not about desired_capacity, it's about min_size and max_size | 08:00 |
| Yanyanhu | is the default value provided by interface defined in client.py? | 08:00 |
| elynn_ | They are allowed to be null. | 08:00 |
| Yanyanhu | yes | 08:00 |
| Yanyanhu | this is supported from service side | 08:01 |
| elynn_ | I create a cluster through rest API with min_size is null. | 08:01 |
| elynn_ | then when I use senlinclient to show it, I got a NoneType error. | 08:02 |
| Yanyanhu | let me check the code | 08:02 |
| elynn_ | I just wanna to know it's better to fix it in senlinclient or senlin server. | 08:02 |
| elynn_ | I met several bugs about null value problem, it might cause problems when doing cluster-scale-out/in. | 08:03 |
| Yanyanhu | http://git.openstack.org/cgit/openstack/senlin/tree/senlin/engine/cluster.py#n70 | 08:03 |
| elynn_ | Seems the logic is wrong there. | 08:04 |
| Yanyanhu | so I guess after cluster object is created, min_size, max_size should not be None even user didn't give these two parameters when creating cluster | 08:04 |
| Yanyanhu | you mean this code? | 08:05 |
| elynn_ | I pass with min_size, but its value is Null. | 08:05 |
| Yanyanhu | http://git.openstack.org/cgit/openstack/senlin/tree/senlin/engine/cluster.py#n70 | 08:05 |
| elynn_ | yes | 08:05 |
| Yanyanhu | Null? | 08:05 |
| Yanyanhu | string | 08:05 |
| xuhaiwei | elynn_, you do it by curl ? | 08:05 |
| elynn_ | means None when translate to dict I think. | 08:06 |
| elynn_ | xuhaiwei: I use rest client. | 08:06 |
| xuhaiwei | not by senlin command? | 08:07 |
| Yanyanhu | I guess this could be client's problem since we have the following check in engine service http://git.openstack.org/cgit/openstack/senlin/tree/senlin/engine/service.py#n494 | 08:07 |
| elynn_ | No, I'm not using senlinclient. | 08:07 |
| Yanyanhu | if the value of min_size/max_size parameter is not integer, exception should be raised here | 08:08 |
| elynn_ | Let me check | 08:08 |
| elynn_ | I don't think it will reach L494, L493 will check if it's none or not. | 08:09 |
| Yanyanhu | yes, 494 is for checking the value type | 08:09 |
| Yanyanhu | if the value is None, the workflow will go to http://git.openstack.org/cgit/openstack/senlin/tree/senlin/engine/cluster.py#n70 | 08:09 |
| elynn_ | I will figure out to propose a patch for it... | 08:10 |
| Yanyanhu | thanks a lot. Or if you have log detail, you can also paste it out. We can help to check it :) | 08:10 |
| elynn_ | Will open a bug first, so you will know what happened. | 08:10 |
| Yanyanhu | ok | 08:11 |
| *** lvdongbing has quit IRC | 08:37 | |
| *** pratikmallya has joined #senlin | 08:38 | |
| *** lvdongbing has joined #senlin | 08:41 | |
| *** lkarm has joined #senlin | 08:42 | |
| *** lkarm has quit IRC | 08:46 | |
| *** pratikmallya has quit IRC | 08:58 | |
| *** pratikmallya has joined #senlin | 08:59 | |
| *** pratikmallya has quit IRC | 09:00 | |
| *** yangyapeng has joined #senlin | 09:11 | |
| *** hightall has quit IRC | 09:30 | |
| *** Liuqing has quit IRC | 09:42 | |
| *** lvdongbing has quit IRC | 09:43 | |
| *** gongysh_ has quit IRC | 09:44 | |
| *** lvdongbing has joined #senlin | 09:45 | |
| *** lvdongbing has quit IRC | 09:46 | |
| *** elynn_ has quit IRC | 09:49 | |
| openstackgerrit | Yanyan Hu proposed openstack/senlin: [WIP]Enable initiative action selecting in start_action https://review.openstack.org/244026 | 09:50 |
| *** zhenguo has quit IRC | 09:56 | |
| *** Yanyanhu has quit IRC | 09:59 | |
| *** Liuqing has joined #senlin | 10:09 | |
| openstackgerrit | xu-haiwei proposed openstack/senlin: Fix SDK exception message parse https://review.openstack.org/243959 | 10:44 |
| *** yangyapeng has quit IRC | 11:07 | |
| *** elynn_ has joined #senlin | 11:42 | |
| *** elynn_ has quit IRC | 12:06 | |
| *** openstackgerrit has quit IRC | 12:31 | |
| *** openstackgerrit has joined #senlin | 12:32 | |
| *** pratikmallya has joined #senlin | 12:39 | |
| *** pratikmallya has quit IRC | 12:42 | |
| openstackgerrit | Julio Ruano proposed openstack/senlin: Update TODO work items https://review.openstack.org/243903 | 12:51 |
| *** lvdongbing has joined #senlin | 12:56 | |
| *** pratikmallya has joined #senlin | 12:58 | |
| *** lvdongbing has quit IRC | 13:43 | |
| *** pratikmallya has quit IRC | 13:44 | |
| *** lkarm has joined #senlin | 14:03 | |
| *** pratikmallya has joined #senlin | 14:05 | |
| *** pratikma_ has joined #senlin | 14:08 | |
| *** pratikmallya has quit IRC | 14:10 | |
| *** Liuqing has quit IRC | 14:14 | |
| *** rebase has joined #senlin | 14:30 | |
| *** jdandrea has joined #senlin | 15:17 | |
| *** jdandrea has quit IRC | 15:18 | |
| *** jdandrea has joined #senlin | 15:18 | |
| *** openstackgerrit has quit IRC | 16:02 | |
| *** openstackgerrit has joined #senlin | 16:02 | |
| *** lkarm has quit IRC | 16:55 | |
| *** pratikma_ has quit IRC | 17:28 | |
| *** pratikmallya has joined #senlin | 17:51 | |
| *** pratikma_ has joined #senlin | 18:20 | |
| *** pratikmallya has quit IRC | 18:23 | |
| *** Jezogwza has joined #senlin | 18:57 | |
| *** lkarm has joined #senlin | 19:04 | |
| *** rebase has quit IRC | 19:11 | |
| *** rebase has joined #senlin | 19:36 | |
| *** pratikma_ has quit IRC | 20:28 | |
| *** pratikmallya has joined #senlin | 20:49 | |
| *** pratikmallya has quit IRC | 20:50 | |
| *** pratikmallya has joined #senlin | 20:51 | |
| *** pratikma_ has joined #senlin | 21:18 | |
| *** pratikmallya has quit IRC | 21:21 | |
| *** pratikma_ has quit IRC | 21:23 | |
| *** pratikmallya has joined #senlin | 21:51 | |
| *** pratikma_ has joined #senlin | 21:54 | |
| *** pratikmallya has quit IRC | 21:57 | |
| *** pratikmallya has joined #senlin | 22:17 | |
| *** pratikma_ has quit IRC | 22:19 | |
| *** pratikmallya has quit IRC | 22:49 | |
| *** Jezogwza has quit IRC | 23:06 | |
| *** rebase has quit IRC | 23:10 | |
| *** rebase has joined #senlin | 23:39 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!