| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-sdks - 2017-03-28 | |||
| 15:55:52 | sfinucan | dtroyer: ^ | |
| 15:56:09 | sfinucan | reedip: Shh! No spoilers! | |
| 15:56:37 | sfinucan | (Ghost in the Shell isn't out in Ireland yet :)) | |
| 15:56:47 | reedip | sfinucan : Oh you didnt see the 1995 version | |
| 15:57:00 | reedip | the original Japanese Magna | |
| 15:57:16 | reedip | There are 2 versions | |
| 15:57:18 | reedip | one 1995 | |
| 15:57:20 | reedip | one 2003 | |
| 15:57:28 | reedip | the 2003 one is of the Laughing Man | |
| 15:57:41 | reedip | the 1995 is of the Puppetier | |
| 15:57:52 | reedip | the 2003 one is taken up for the Live Action movie :) | |
| 15:58:34 | reedip | Well .. thats all I will say ... | |
| 15:58:45 | reedip | I have to also see the Ghost in the Shell | |
| 15:59:04 | reedip | it will release a month later, but this is the official release notes of the movie :D | |
| 15:59:42 | openstackgerrit | Monty Taylor proposed openstack/os-client-config master: Remove out of date comment https://review.openstack.org/450871 | |
| 15:59:48 | dtroyer | sfinucan: thanks, I should get to that after the noon meetings | |
| 17:21:45 | openstackgerrit | Merged openstack/os-client-config master: Stop special-casing idenity catalog lookups https://review.openstack.org/450259 | |
| 17:33:26 | openstackgerrit | Merged openstack/python-openstackclient master: Fix volume transfers request commands https://review.openstack.org/386770 | |
| 17:34:18 | efried | Greetings! | |
| 17:34:24 | efried | How do I create a VM with no network? Used to be I could say --nick none (which the usage statement & help claim is fine) - but that's now telling me "nics must be a list". | |
| 17:34:43 | efried | (via openstack CLI, btw) | |
| 17:38:37 | ankur-gupta-f4 | efried: looking into it. You might have hit a bug since you should be able to specify --nic none | |
| 17:39:01 | efried | ankur-gupta-f4 Okay, thanks. Let me know if I can help. | |
| 17:39:25 | efried | ankur-gupta-f4 Where can I clone the source from? (Apparently it's not https://review.openstack.org/#/admin/projects/openstack/openstackclient - that guy is pretty empty.) | |
| 17:39:46 | ankur-gupta-f4 | github.com/openstack/python-openstackclient | |
| 17:40:01 | efried | thanks | |
| 17:45:47 | ankur-gupta-f4 | https://github.com/openstack/python-novaclient/blob/master/novaclient/v2/servers.py#L1295 | |
| 17:45:51 | ankur-gupta-f4 | efried: ^^^ | |
| 17:46:04 | efried | Yup, just found that. | |
| 17:49:28 | efried | Looks like _validate_create_nics should be tolerating a value of 'none' (str); or that should be getting cast to an actual None and _validate_create_nics should be tolerating *that* value. | |
| 17:51:55 | ankur-gupta-f4 | should be. May need to wrangle some Nova folk. Regardless its a bug and if Nova no longer accepts server creation with no NIC then it needs to be noted in the OpenStack Client. | |
| 17:53:15 | efried | ankur-gupta-f4 The docstring in the caller of that method clearly states that 'none' should be allowed, so I suspect it's a functional (rather than help text) bug. | |
| 17:53:23 | efried | Thanks for the help; I'll follow up with mriedem in #openstack-nova. | |
| 17:53:44 | ankur-gupta-f4 | good luck | |
| 19:01:27 | efried | ankur-gupta-f4 Turns out I needed to specify --os-compute-api-version 2.37 | |
| 19:05:32 | ankur-gupta-f4 | Ah okay. Good to know | |
| 19:05:52 | ankur-gupta-f4 | Altho i think there should be some doc note so future users know | |
| 19:07:06 | efried | ankur-gupta-f4 Well, the help text does mention that 'none' is for 2.37+. I just didn't have any idea what that meant. | |
| 19:07:36 | efried | --nic <net-id=net-uuid,v4-fixed-ip=ip-addr,v6-fixed-ip=ip-addr,port-id=port-uuid,auto,none> | |
| 19:07:37 | efried | with this UUID, v4-fixed-ip: IPv4 fixed address for | |
| 19:07:37 | efried | to network with this UUID, port-id: attach NIC to port | |
| 19:07:37 | efried | id must be provided, but not both. net-id: attach NIC | |
| 19:07:37 | efried | times to create multiple NICs. Either net-id or port- | |
| 19:07:37 | efried | Create a NIC on the server. Specify option multiple | |
| 19:07:39 | efried | NIC (optional), v6-fixed-ip: IPv6 fixed address for | |
| 19:07:41 | efried | NIC (optional), none: (v2.37+) no network is attached, | |
| 19:07:43 | efried | auto: (v2.37+) the compute service will automatically | |
| 19:07:45 | efried | allocate a network. Specifying a --nic of auto or none | |
| 19:07:47 | efried | cannot be used with any other --nic value. | |
| 19:09:34 | ankur-gupta-f4 | Ah okay. I didnt see it | |
| 19:10:36 | dtroyer | So in the case of none or auto, we need to attempt to do the microversion thing and just handle that if possible, otherwise throw an error. | |
| 19:40:45 | efried | bleh, now it's not working again. | |
| 19:41:55 | efried | nope, cause I'm a doofus. I was spelling 2.37 as 2.7. | |
| 19:51:08 | efried | ankur-gupta-f4 dtroyer Apparently novaclient will detect and use the latest possible version. But openstackclient will default to 2.1 if not otherwise specified via OS_COMPUTE_API_VERSION/--os-compute-api-version: https://github.com/openstack/python-openstackclient/blob/master/openstackclient/compute/client.py#L26 | |
| 20:00:49 | openstackgerrit | Merged openstack/os-client-config master: Remove out of date comment https://review.openstack.org/450871 | |
| 20:16:50 | efried | Philosophically, is there a reason openstackclient doesn't detect and use the latest? | |
| 20:18:04 | dtroyer | no, it should, but there is some trickiness with microversions, and with the upcoming novaclient 8.0 release that'll break us like it breaks Horizon. | |
| 20:20:29 | efried | Breaks who? The openstackclient... uh, client? | |
| 20:21:24 | dtroyer | yup, we'll be supporting nova-net for a lot longer than Novaclient | |
| #openstack-sdks - 2017-03-29 | |||
| 01:51:56 | openstackgerrit | Ethan Lynn proposed openstack/python-openstacksdk master: Add vlan_transparent property to network resource https://review.openstack.org/450080 | |
| 02:44:55 | openstackgerrit | Merged openstack/python-openstacksdk master: Add vlan_transparent property to network resource https://review.openstack.org/450080 | |
| 03:05:11 | reedip | dtroyer, stevemar there / | |
| 03:16:15 | openstackgerrit | Merged openstack/python-openstacksdk master: Find floating ip by ip address https://review.openstack.org/449879 | |
| 03:19:03 | openstackgerrit | Takashi NATSUME proposed openstack/python-openstackclient master: Add '--no-live' and '--force' option in 'server migrate' https://review.openstack.org/427565 | |
| 04:16:07 | openstackgerrit | Takashi NATSUME proposed openstack/python-openstackclient master: List/Show all server migration types https://review.openstack.org/450119 | |
| 06:53:16 | openstackgerrit | Takashi NATSUME proposed openstack/python-openstackclient master: List/Show all server migration types https://review.openstack.org/450119 | |
| 06:56:44 | rabel | good morning | |
| 07:14:50 | reedip | good afternoon | |
| 07:48:28 | rabel | reedip: ;) | |
| 07:49:05 | reedip | :) | |
| 07:52:24 | rabel | could someone have a look at https://review.openstack.org/#/c/444924/ ? It is not of high priority, but it lies idle for a week now. | |
| 07:57:42 | reedip | rabel hi | |
| 07:57:55 | reedip | i had a patch for smething you are doing in Line#614 | |
| 07:58:05 | reedip | but didnt push it again ... need to retake it | |
| 07:58:59 | reedip | the split on , and then dict key value pair logic | |
| 07:58:59 | reedip | you can put this logic in osc-lib | |
| 07:59:34 | reedip | https://review.openstack.org/#/c/425515/ | |
| 08:00:08 | openstackgerrit | Reedip proposed openstack/osc-lib master: [WIP]Add MultiKeyValueCommaAction to osc-lib https://review.openstack.org/425515 | |
| 08:00:34 | reedip | rabel : can you check this patch , it would avoid you repeating the same code ( probably ) | |
| 08:03:58 | rabel | reedip: i will have a look at it. it would make sense to use your class for my patch once it is merged. | |
| 08:04:29 | reedip | rabel yes . though I wont like to bliock you , so you can add a ToDo for yourself | |
| 08:04:51 | reedip | that ways, everything is perfect, I guess | |
| 08:05:54 | rabel | reedip: it's true that my patch could already be merged without yours. i use the same logic as the --nic implementation. so we could change both to use the new class in a separate patch. | |
| 08:06:13 | reedip | rabel : sounds good ... lets race :D | |
| 08:07:31 | reedip | just kidding, I will push mine a bit slower , but otherwise the code looks good , I will look at it once more though in the evening | |
| 08:09:09 | rabel | :D i added me as a reviewer for your patch. i will have a look at it now and once you finished it, i will review it in detail | |
| 08:13:55 | reedip | :D | |
| 08:32:45 | rabel | reedip: let's see if i understand it right: the difference between MultiKeyValueAction and MultiKeyValueCommaAction is the latter supports something like "--my_param key1=val,ue1,key2=val,ue3", is that right? | |
| 08:33:06 | reedip | something like that | |
| 08:33:15 | reedip | actually exatly liek that | |
| 08:35:56 | rabel | than i think --nic, --network and --port should use MultiKeyValueAction, since we don't have values containing commas or multiple values for a key, i think. | |
| 08:36:14 | rabel | by the way: your patch looks good to me | |
| 08:43:38 | reedip | --network can have IPV4 address, IPv6 address and Network ID | |
| 08:44:18 | reedip | as the addresses are one-to-one, so I guess yes, the multikeyvaluecomma action may not be required | |
| 09:19:39 | RuiChen | hi dtroyer: I add comments in your patch https://review.openstack.org/#/c/450452/ | |
| 12:08:03 | dtroyer | RuiChen: thanks. FWIW, if we have not already set neutron to be the default for help we should do that so we can still do most help wihtout auth. | |
| 12:09:40 | dtroyer | it looks like that is the case… with no auth info, openstack help lists neutron commands now | |
| 14:23:46 | openstackgerrit | Jens Rosenboom proposed openstack/python-openstackclient master: Fix block-device-mapping when volume_size is empty https://review.openstack.org/451432 | |
| 14:32:50 | reedip | o/ | |
| 14:43:11 | frickler | why does e.g. "openstack server show" use compute API v2.1 by default and not something newer? | |