| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-sdks - 2017-03-02 | |||
| 06:16:33 | ankur-gupta-f4 | but the unset is a valid point. | |
| 06:17:00 | reedip_ | That is taken care in the UT | |
| 06:17:13 | reedip_ | assertNone does that, I guess | |
| 06:19:45 | ankur-gupta-f4 | given the complexity of removing/deleting resources (as seen with how many times the test has been rewritten order wise). I think it instead of 'floating ip unset port' coming in in the cleanUp, it should be done like the the other commands where it is run, a fip or port show is done and confirmed that the attachment no longer exists. But that is just an | |
| 06:19:45 | ankur-gupta-f4 | opinion. I will remove the 01 | |
| 06:19:47 | ankur-gupta-f4 | -1 | |
| 06:20:01 | ankur-gupta-f4 | thanks for the answers | |
| 06:22:53 | ankur-gupta-f4 | reedip_: headed to bed. talk to you in morning | |
| 06:23:22 | reedip_ | ankur-gupta-f4 : thanks :) | |
| 06:24:39 | openstackgerrit | jiahui.qiang proposed openstack/python-openstackclient master: Functional test for subnet_pool https://review.openstack.org/440150 | |
| 09:13:47 | openstackgerrit | Rico Lin proposed openstack/keystoneauth master: Update test requirement https://review.openstack.org/440236 | |
| 09:20:13 | openstackgerrit | Merged openstack/python-openstackclient master: Nit: Trivial doc formatting fix for network flavor https://review.openstack.org/439965 | |
| 12:23:08 | openstackgerrit | Huanxuan Ao proposed openstack/python-openstackclient master: Fix NoneType error for volume snapshot create command https://review.openstack.org/440497 | |
| 13:24:23 | openstackgerrit | Huanxuan Ao proposed openstack/python-openstackclient master: Remove deprecated "ip floating *" commands https://review.openstack.org/440550 | |
| 13:30:00 | openstackgerrit | Merged openstack/python-openstacksdk master: Updated from global requirements https://review.openstack.org/440103 | |
| 15:01:59 | reedip_1 | o/ | |
| 15:05:20 | openstackgerrit | Steve Martinelli proposed openstack/python-openstackclient master: Remove deprecated "ip floating *" commands https://review.openstack.org/440550 | |
| 15:45:00 | openstackgerrit | Nakul Dahiwade proposed openstack/python-openstackclient master: OSC Network Flavor Profile https://review.openstack.org/396783 | |
| 15:53:31 | openstackgerrit | Shashank Kumar Shankar proposed openstack/python-openstackclient master: Introduce Neutron DHCP agent commands to OSC https://review.openstack.org/387611 | |
| 16:22:23 | openstackgerrit | Merged openstack/api-wg master: Fix typo in the doc https://review.openstack.org/436865 | |
| 16:39:08 | dims | stevemar : dtroyer : does this ring a bell? | |
| 16:39:10 | dims | 2017-03-02 16:31:35.234 | + /opt/stack/new/grenade/projects/70_cinder/resources.sh:destroy:205 : openstack floating ip delete 0759cfd6-2e42-410b-ab70-7ff81d19562c | |
| 16:39:10 | dims | 2017-03-02 16:31:36.894 | Failed to delete floating_ip with name or ID '0759cfd6-2e42-410b-ab70-7ff81d19562c': 'Proxy' object has no attribute 'session' | |
| 16:40:12 | dims | could it be from openstacksdk===0.9.14? | |
| 16:44:38 | openstackgerrit | Nakul Dahiwade proposed openstack/python-openstackclient master: Jsonify meter and meter rule functional tests https://review.openstack.org/439854 | |
| 16:58:20 | reedip_1 | dims : this issue was occurring with the older version of devstack | |
| 16:58:36 | reedip_1 | I think there was a new patch which fixed it | |
| 16:58:41 | reedip_1 | dont recall it , sorry | |
| 17:01:00 | dims | reedip_1 : ack thanks | |
| 17:10:48 | dtroyer | dims: yes it is from sdk 0.9.14, fixed in an unreleased OSC | |
| 17:10:52 | dtroyer | time for a release? | |
| 17:11:52 | openstackgerrit | Anindita Das proposed openstack/python-openstackclient master: Jsonify network flavor functional tests https://review.openstack.org/440710 | |
| 17:16:41 | dims | ack dtroyer, early next week is fine | |
| 17:26:59 | ssbarnea | hi! Am i the first one that is surprised that I cannot delete resources like networks, routers using wildcards/regex? | |
| 17:28:21 | reedip_1 | ssbarnea : can u elaborate ? | |
| 17:28:55 | ssbarnea | reedip_1: I was hoping that I could do:> router delete pref* | |
| 17:29:30 | ssbarnea | glob or regexp patterns, any would have been ok. | |
| 17:30:20 | reedip_1 | umm, no , you cannot use regex for name/id parameter AFAIK | |
| 17:30:42 | ssbarnea | some applies to instances, networks and routers. Now I don't know if this was not implemented for a special reason or just there was no need for it. | |
| 17:31:10 | ssbarnea | reedip_: if nobody cared about that, I would not mind trying to implement it myself. | |
| 17:37:11 | reedip_1 | ssbarnea : do you think that there is a high requirement for regex ? | |
| 17:39:02 | ssbarnea | reedip_: i don't know, in this case I am only speaking about myself: in order to cleanup remaining resources from broken CI jobs, I do want to use a prefix pattern for removing them. We use a naming convention "job-name-number-" which makes easier to do cleanups. | |
| 17:39:29 | ssbarnea | reedip_: ideally jobs do take care of cleanup themselves, but as you know things can always go wrong. | |
| 17:39:59 | reedip_1 | I know , have experienced it :) | |
| 17:40:22 | reedip_1 | ssbarnea : but I do not think all users follow the same patern | |
| 17:40:47 | ankur-gupta-f4 | ssbarnea: would purge satisfy your requirements? Because that is in works | |
| 17:40:51 | reedip_1 | I mean, not all users use some prefix for naming their resources | |
| 17:42:15 | ssbarnea | reedip_1: indeed, but the idea of being able to use a regex or glob pattern seems live a valid use-case. Now I don't know which limitations we have on job names because if people can use "*" in their names glob will not work. but if the names are close to filename patters blog would sound like a reasonable way. | |
| 17:42:18 | reedip_1 | ankur-gupta-f4 : purge is more like a nuclear warhead for a tenant , it clears all ists resources. I think ssbarnea needs something more fine tuned, like one which removes all routers or networks etc . of a specific tenant | |
| 17:43:23 | reedip_1 | ssbarnea : I agree with you , but then again its not a pretty big requirement , IIUC. And it can be easily scripted at user end if required | |
| 17:43:35 | ssbarnea | indeed, purge would not work, there is a must to be able to specify a pattern. Still, this may be an option of purge itself. I am not sure which is the best place the os-client or the purge utility? | |
| 17:43:41 | dtroyer | ssbarnea: there are a couple of reasons globbing resource names is not implemented, not necessarily in any order: a) not all users feel comfortable allowing that (rm -rf / like fear), even with an option; b) it has not been a priority; c) error handling gets tricky (mostly solved in OSC now with the multiple hand-listed deletes) | |
| 17:44:31 | dtroyer | your closest work-around right now is to list resources using a filter, possibly piping thought more filtering, then piping in to a delete command | |
| 17:44:35 | dtroyer | you know, UNIX-style | |
| 17:44:44 | ssbarnea | dtroyer: maybe you are right, i could easily make a bash script that does this. | |
| 17:44:52 | ankur-gupta-f4 | +1 | |
| 17:44:58 | reedip_1 | +1 | |
| 17:45:03 | dtroyer | the thing I really want here is the ability to do ith in interactive mode | |
| 17:45:18 | dtroyer | so that takes shell out of the option list | |
| 17:45:53 | dtroyer | also, piping lists of commands into interactive mode is fun and definitely a step forward | |
| 18:30:26 | ssbarnea | dtroyer: I had a good progress writing a cleanup script but hit one problem: network delete may fail due to existing ports being assigned. The bad part is that it seems that not all ports are named so I cannot cleanup them using the naming pattern. Is there a way to remove networks with "cascade"? | |
| 18:36:40 | dtroyer | that's what the purge command is intended to do, but at the project level | |
| 18:38:29 | ssbarnea | dtroyer: that's the problem: at project level is useless because it would wipe everything. | |
| 18:39:35 | dtroyer | not useless, but not what you want... | |
| 18:40:10 | dtroyer | etiher way the same tasks need to be done, whether in a script or inside OSC, walk the resources and delete everything required by Neutron for a specific resource | |
| 18:51:18 | openstackgerrit | Merged openstack/os-client-config master: [Fix gate]Update test requirement https://review.openstack.org/440597 | |
| 18:52:10 | ssbarnea | dtroyer: well, i will do it in the end and publish the script in came someone else is interested. | |
| 18:59:43 | dtroyer | ssbarnea: thanks. the logic and decision flow of that would be useful to re-use should it get brought into OSC | |
| 19:06:45 | openstackgerrit | Huanxuan Ao proposed openstack/python-openstackclient master: Fix NoneType error for volume snapshot create command https://review.openstack.org/440497 | |
| 19:26:46 | openstackgerrit | Huanxuan Ao proposed openstack/python-openstackclient master: Remove deprecated "ip floating (pool)/ip fixed *" commands https://review.openstack.org/440550 | |
| 19:54:30 | dtroyer | FWIW, OSC release note job failures are being fixed in the next reno release (2.1.2?) coming after https://review.openstack.org/#/c/440776/ merges. | |
| 19:54:52 | dtroyer | Please do not blindly recheck wihtout having some belief that the job might pass... | |
| 19:57:20 | ankur-gupta-f1 | I believeeee | |
| 20:14:40 | openstackgerrit | Justin A Wilson proposed openstack/python-openstackclient master: Add support for Cinder API 3.3/3.5 into OSC https://review.openstack.org/421585 | |
| 20:15:28 | openstackgerrit | Justin A Wilson proposed openstack/python-openstackclient master: Add support for Cinder API 3.3/3.5 into OSC https://review.openstack.org/421585 | |
| 20:46:26 | openstackgerrit | Brian Curtin proposed openstack/python-openstacksdk master: Enable streaming responses in download_image https://review.openstack.org/437588 | |
| 23:14:34 | openstackgerrit | Adrian Turjak proposed openstack/python-openstackclient master: Non-Admin can't list own projects https://review.openstack.org/376056 | |
| 23:15:11 | openstackgerrit | Adrian Turjak proposed openstack/python-openstackclient master: Non-Admin can't list own projects https://review.openstack.org/376056 | |
| 23:21:09 | openstackgerrit | Adrian Turjak proposed openstack/python-openstackclient master: Non-Admin can't list own projects https://review.openstack.org/376056 | |
| 23:35:41 | openstackgerrit | Adrian Turjak proposed openstack/python-openstackclient master: Remove ClientCache https://review.openstack.org/388232 | |
| #openstack-sdks - 2017-03-03 | |||
| 03:06:43 | reedip_1 | o/ | |
| 06:41:58 | Amit82 | Hi All, I have deployed Openstack using Openstack-Ansible. | |
| 06:42:13 | Amit82 | I am using Newton release from tag 14.0.8. My test environment is containing only Compute Node and Controller Node (Infra Node). | |
| 06:42:26 | Amit82 | When using Openstack Python SDK, I am getting following error while making connection to external_vib_lp_address (192.168.255.45) binded to port 5000. | |
| 06:42:43 | Amit82 | "openstack.exceptions.SDKException: Connection failure that may be retried." | |
| 06:43:06 | Amit82 | Any clue about this issue? More detailed traces are at: http://paste.openstack.org/show/601053/ | |
| 06:43:15 | Amit82 | Python script which I am using for this purpose is at: http://paste.openstack.org/show/601241/ | |
| 06:43:32 | Amit82 | I am running this python script on some external machine, I am successfully able to run Horizon UI on the same machine where this python script is run. | |
| 10:50:19 | Amit82 | Hi | |
| 10:50:53 | Amit82 | I am trying to use openstack python SDK | |
| 10:51:03 | Amit82 | to start with, I am trying to attempt a connection | |
| 10:51:15 | Amit82 | I have deployed Newton using Openstack-Ansible | |
| 10:51:43 | Amit82 | I tried the python script: http://paste.openstack.org/show/601268/ from utility container but it is giving me 400 | |
| 10:52:06 | Amit82 | These are detailed error logs: http://paste.openstack.org/show/601271/ | |
| 10:52:43 | Amit82 | Does these errors mean that some "domain" is expected while making connection | |
| 10:53:30 | Amit82 | Request at link: http://paste.openstack.org/show/601273/ is failing with 400. It is generated by python script | |
| 10:54:14 | Amit82 | Request at link: http://paste.openstack.org/show/601275/ is successful. It is generated by openstack command line client. | |
| 10:54:36 | Amit82 | You can observe the difference is related to "domain". These are excerpts from wireshark. | |
| 10:55:55 | Amit82 | Could any of you guide or point to some resource showing how to add domain related things in while attempting to connect | |
| 13:39:35 | dtroyer | amit213: You appear to be using v3 authentication, which yes requires a domain. By default the domain ID is "default" unless you have changed that in keystone.conf. | |