| *** esker has joined #refstack | 00:35 | |
| *** esker has quit IRC | 00:35 | |
| *** esker has joined #refstack | 00:35 | |
| *** cody-somerville has joined #refstack | 01:17 | |
| *** pvaneck has quit IRC | 01:38 | |
| *** dwalleck has joined #refstack | 01:51 | |
| *** edmondsw has quit IRC | 02:04 | |
| *** cody-somerville has quit IRC | 02:09 | |
| *** dwalleck has quit IRC | 02:14 | |
| *** esker has quit IRC | 03:25 | |
| openstackgerrit | David Liu proposed openstack/refstack: RefStack test results page paragraph should change accordingly when user is login. https://review.openstack.org/243904 | 03:26 |
|---|---|---|
| *** esker has joined #refstack | 03:26 | |
| *** esker has quit IRC | 03:27 | |
| openstackgerrit | David Liu proposed openstack/refstack: Set test results paragraph messages. https://review.openstack.org/243904 | 03:32 |
| *** hogepodge has quit IRC | 04:35 | |
| *** dwalleck has joined #refstack | 04:54 | |
| *** dwalleck has quit IRC | 07:05 | |
| *** catherineD has quit IRC | 07:43 | |
| *** catherineD has joined #refstack | 07:44 | |
| *** nijaba has quit IRC | 08:28 | |
| openstackgerrit | David Liu proposed openstack/refstack-client: Add guard checks in _get_cpid_from_keystone. https://review.openstack.org/236780 | 09:59 |
| *** nijaba has joined #refstack | 10:52 | |
| openstackgerrit | David Liu proposed openstack/refstack-client: Add guard checks in _get_cpid_from_keystone. https://review.openstack.org/236780 | 11:34 |
| *** openstackgerrit has quit IRC | 12:31 | |
| *** openstackgerrit has joined #refstack | 12:31 | |
| *** edmondsw has joined #refstack | 13:31 | |
| *** dwalleck has joined #refstack | 16:01 | |
| *** openstackgerrit has quit IRC | 16:02 | |
| *** openstackgerrit has joined #refstack | 16:02 | |
| *** hogepodge has joined #refstack | 16:03 | |
| *** nijaba_ has joined #refstack | 16:13 | |
| *** nijaba_ has quit IRC | 16:13 | |
| *** nijaba_ has joined #refstack | 16:13 | |
| *** nijaba has quit IRC | 16:17 | |
| *** zehicle has joined #refstack | 16:28 | |
| *** esker has joined #refstack | 16:49 | |
| *** dwalleck has quit IRC | 17:03 | |
| *** zehicle has quit IRC | 17:26 | |
| *** esker has quit IRC | 17:28 | |
| *** esker has joined #refstack | 18:32 | |
| *** zackf has joined #refstack | 18:45 | |
| *** pvaneck has joined #refstack | 19:31 | |
| *** esker has quit IRC | 20:12 | |
| openstackgerrit | Paul Van Eck proposed openstack/refstack: Add support for defcore schema 1.4 https://review.openstack.org/244295 | 20:22 |
| *** esker has joined #refstack | 20:32 | |
| *** esker has quit IRC | 20:37 | |
| *** edmondsw has quit IRC | 22:23 | |
| openstackgerrit | Paul Van Eck proposed openstack/refstack: Add support for defcore schema 1.4 https://review.openstack.org/244295 | 22:35 |
| hogepodge | catherineD: pvaneck: When I was running against DH I was running into the cpid problem. | 22:36 |
| hogepodge | catherineD: pvaneck: I had to modify refstack-client to fake the cpid (and duration) and upload existing subunit results | 22:37 |
| pvaneck | hogepodge: for cpid, did you just generate a random hash? | 22:40 |
| pvaneck | thinking we need something like that for a fallback | 22:41 |
| hogepodge | pvaneck: kind of? I just mashed on the keyboard | 22:42 |
| pvaneck | haha okay. in any case, it seems like the service catalog where we get the identity id is not exposed on some clouds. | 22:44 |
| pvaneck | catherineD: so if this is the case, we have to just randomly generate one so as not to block or prompt the user for a cpid entry | 22:45 |
| pvaneck | i know we were talking about taking in cpid as input from the user | 22:46 |
| catherineD | yup it is our top priority to implement in refstack-client see the Mitaka cycle priority list in https://etherpad.openstack.org/p/refstack-tokyo-summit | 22:48 |
| pvaneck | but then mapping results to the same cloud would be harder as multiple users could put in 'my-cloud' or something when referring to completely different clouds. | 22:48 |
| catherineD | that is the possibility and the reason why we use keystone service id as the cpid ... | 22:50 |
| catherineD | pvaneck: we have no idea whether the user is sending data for the same cloud ... by randomly generating the cpid by refstack-client at upload, we would treat each upload without cpid as a new cloud ... | 22:57 |
| pvaneck | yes unfortunately, but only as a last resort | 22:59 |
| hogepodge | catherineD: pvaneck: there are possible workarounds. Parsing tempest.conf and storing a file that matches the endpoint with a random cpid? | 23:02 |
| pvaneck | hogepodge: yea i think something like that will be the best bet for at least keeping the cpid consistent between test runs from the same user | 23:03 |
| hogepodge | hash the endpoint? | 23:04 |
| catherineD | hogepodge: where is the file stored? | 23:04 |
| hogepodge | wherever the client is run from | 23:04 |
| pvaneck | just worried about private address space endpoints where two clouds could have the same endpoint ip | 23:05 |
| pvaneck | but as a fallback, just hashing the endpoint may not be so bad | 23:08 |
| pvaneck | seems mostly public clouds with the catalog not exposed, which would probably mean a public domain/ip for the endpoint | 23:13 |
| openstackgerrit | Merged openstack/refstack: Add support for defcore schema 1.4 https://review.openstack.org/244295 | 23:42 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!