| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2021-05-21 | |||
| 10:42:00 | bauzas | ? | |
| 10:42:00 | bauzas | ? | |
| 10:42:06 | sean-k-mooney | so we could maybe model this as a compute capablity trait that was conditional on the neutron extenion being presnt | |
| 10:42:06 | sean-k-mooney | so we could maybe model this as a compute capablity trait that was conditional on the neutron extenion being presnt | |
| 10:42:19 | bauzas | wait wait | |
| 10:42:19 | bauzas | wait wait | |
| 10:42:28 | sean-k-mooney | but i dont think this should be a microverion sicne that on its own woudl not tell you it will work | |
| 10:42:28 | sean-k-mooney | but i dont think this should be a microverion sicne that on its own woudl not tell you it will work | |
| 10:42:46 | sean-k-mooney | bauzas: i was just trying to come up with any signel that could be useful | |
| 10:42:46 | sean-k-mooney | bauzas: i was just trying to come up with any signel that could be useful | |
| 10:42:58 | sean-k-mooney | bauzas: i dont think we actully need that | |
| 10:42:58 | sean-k-mooney | bauzas: i dont think we actully need that | |
| 10:43:14 | bauzas | I think we have a loose contract on the features a cloud supports | |
| 10:43:14 | bauzas | I think we have a loose contract on the features a cloud supports | |
| 10:43:55 | sean-k-mooney | gibi you just wanted a way to tell form the api if you could use this yes? | |
| 10:43:55 | sean-k-mooney | gibi you just wanted a way to tell form the api if you could use this yes? | |
| 10:44:02 | bauzas | as an end user, especially when you wanna use scheduler hints, you're absolutely blind whether the API endpoint will support it | |
| 10:44:02 | bauzas | as an end user, especially when you wanna use scheduler hints, you're absolutely blind whether the API endpoint will support it | |
| 10:44:06 | gibi | especially as this is an optinal feature as the admin needs to configure pps resources | |
| 10:44:06 | gibi | especially as this is an optinal feature as the admin needs to configure pps resources | |
| 10:44:07 | sean-k-mooney | gibi: is the presnce of the QOS policy created by the admin not enough | |
| 10:44:07 | sean-k-mooney | gibi: is the presnce of the QOS policy created by the admin not enough | |
| 10:44:37 | bauzas | gibi: why can't the user request the neutron extension to see whether this is configured ? | |
| 10:44:37 | bauzas | gibi: why can't the user request the neutron extension to see whether this is configured ? | |
| 10:45:07 | sean-k-mooney | bauzas: they could but they would not know if nova was new enough | |
| 10:45:07 | sean-k-mooney | bauzas: they could but they would not know if nova was new enough | |
| 10:45:16 | sean-k-mooney | but since qos policy create is admin only | |
| 10:45:16 | sean-k-mooney | but since qos policy create is admin only | |
| 10:45:27 | sean-k-mooney | listing the qos polices seams like a better approch | |
| 10:45:27 | sean-k-mooney | listing the qos polices seams like a better approch | |
| 10:45:33 | gibi | bauzas: from API perspective both the extension list and the nova microversion is needed to guarantee that this works | |
| 10:45:34 | gibi | bauzas: from API perspective both the extension list and the nova microversion is needed to guarantee that this works | |
| 10:45:58 | bauzas | gibi: remind me, have you written a new API microversion for bw ? | |
| 10:45:58 | bauzas | gibi: remind me, have you written a new API microversion for bw ? | |
| 10:46:00 | gibi | sean-k-mooney: yes, if the admin created a new pps QoS policy then the user should assume that it is usable | |
| 10:46:00 | gibi | sean-k-mooney: yes, if the admin created a new pps QoS policy then the user should assume that it is usable | |
| 10:46:11 | gibi | bauzas: for the first step yes, | |
| 10:46:11 | gibi | bauzas: for the first step yes, | |
| 10:46:25 | gibi | bauzas: then all the later support for lifecycle operations we did not bump | |
| 10:46:25 | gibi | bauzas: then all the later support for lifecycle operations we did not bump | |
| 10:46:28 | sean-k-mooney | gibi: right and since a user cant create the policy i dont think they need a way to check if the cloud could supprot it | |
| 10:46:28 | sean-k-mooney | gibi: right and since a user cant create the policy i dont think they need a way to check if the cloud could supprot it | |
| 10:46:48 | sean-k-mooney | ah your right | |
| 10:46:48 | sean-k-mooney | ah your right | |
| 10:46:59 | bauzas | gibi: that's what I recall, then | |
| 10:47:00 | bauzas | gibi: that's what I recall, then | |
| 10:47:01 | sean-k-mooney | we did add an inital bump | |
| 10:47:01 | sean-k-mooney | we did add an inital bump | |
| 10:47:16 | bauzas | gibi: I'd honestly consider mocking this approach | |
| 10:47:16 | bauzas | gibi: I'd honestly consider mocking this approach | |
| 10:47:39 | sean-k-mooney | i dont really like bumping the microverion if there is no api change though | |
| 10:47:39 | sean-k-mooney | i dont really like bumping the microverion if there is no api change though | |
| 10:47:50 | bauzas | agreed | |
| 10:47:50 | bauzas | agreed | |
| 10:47:51 | sean-k-mooney | i mean do we really want to force user to use the new microverion | |
| 10:47:51 | sean-k-mooney | i mean do we really want to force user to use the new microverion | |
| 10:47:59 | sean-k-mooney | wehn they do a server create | |
| 10:47:59 | sean-k-mooney | wehn they do a server create | |
| 10:49:06 | gibi | it don't see why a user would use an old microversion if we bumped one for this. | |
| 10:49:06 | gibi | it don't see why a user would use an old microversion if we bumped one for this. | |
| 10:49:37 | sean-k-mooney | well the process is openstack server create --port <uuid> <name> | |
| 10:49:37 | sean-k-mooney | well the process is openstack server create --port <uuid> <name> | |
| 10:50:04 | sean-k-mooney | i dont think they shoudl have to specify a new microverion just becasue the port has a pps qos policy | |
| 10:50:04 | sean-k-mooney | i dont think they shoudl have to specify a new microverion just becasue the port has a pps qos policy | |
| 10:50:21 | gibi | me neither | |
| 10:50:21 | gibi | me neither | |
| 10:50:41 | gibi | but we have a history of bumping for bw and also we bumped for volume multiattach too. So I had to ask | |
| 10:50:41 | gibi | but we have a history of bumping for bw and also we bumped for volume multiattach too. So I had to ask | |
| 10:50:49 | kashyap | sean-k-mooney: dtantsur: You both fell for the "flame-bait" on the IRC topic from Artem Goncharov | |
| 10:50:49 | kashyap | sean-k-mooney: dtantsur: You both fell for the "flame-bait" on the IRC topic from Artem Goncharov | |
| 10:50:56 | sean-k-mooney | volume multi attach had a chagne to the api | |
| 10:50:56 | sean-k-mooney | volume multi attach had a chagne to the api | |
| 10:50:59 | kashyap | He successfully distracted you all. | |
| 10:50:59 | kashyap | He successfully distracted you all. | |
| 10:51:16 | sean-k-mooney | kashyap: not really | |
| 10:51:16 | sean-k-mooney | kashyap: not really | |
| 10:51:27 | gibi | sean-k-mooney: I don't think we changed anyithing in the nova api | |
| 10:51:27 | gibi | sean-k-mooney: I don't think we changed anyithing in the nova api | |
| 10:51:46 | gibi | bauzas: sean-k-mooney: but anyhow then I will not propose a microversion bump for pps | |
| 10:51:46 | gibi | bauzas: sean-k-mooney: but anyhow then I will not propose a microversion bump for pps | |
| 10:51:49 | sean-k-mooney | for multi atach did we not add a flag to the bdms? | |
| 10:51:49 | sean-k-mooney | for multi atach did we not add a flag to the bdms? | |
| 10:51:55 | sean-k-mooney | or was that all cinder side | |
| 10:51:55 | sean-k-mooney | or was that all cinder side | |
| 10:52:05 | gibi | sean-k-mooney: at least our api history does not mention that flag | |
| 10:52:05 | gibi | sean-k-mooney: at least our api history does not mention that flag | |
| 10:52:20 | gibi | https://docs.openstack.org/nova/latest/reference/api-microversion-history.html#maximum-in-queens | |
| 10:52:20 | gibi | https://docs.openstack.org/nova/latest/reference/api-microversion-history.html#maximum-in-queens | |
| 10:52:25 | sean-k-mooney | ah ok i tought there was one for read only vs read write monthing | |
| 10:52:25 | sean-k-mooney | ah ok i tought there was one for read only vs read write monthing | |
| 10:53:03 | gibi | sean-k-mooney, bauzas: thanks again for talking this through. I will update the spec later today, after my fancy friday calls | |
| 10:53:03 | gibi | sean-k-mooney, bauzas: thanks again for talking this through. I will update the spec later today, after my fancy friday calls | |
| 10:53:43 | bauzas | gibi: can you recall *why* you had to create a microversion for bw ? | |
| 10:53:43 | bauzas | gibi: can you recall *why* you had to create a microversion for bw ? | |
| 10:54:05 | sean-k-mooney | i gues not https://github.com/openstack/nova/commit/7e6ae9afd9f6f6bb1bf4289c598fbc95dfd52612 | |
| 10:54:05 | sean-k-mooney | i gues not https://github.com/openstack/nova/commit/7e6ae9afd9f6f6bb1bf4289c598fbc95dfd52612 | |
| 10:58:40 | sean-k-mooney | bauzas: i suspect discoverablity | |
| 10:58:40 | sean-k-mooney | bauzas: i suspect discoverablity | |
| 10:58:50 | sean-k-mooney | likely the same reason for multi attach | |
| 10:58:50 | sean-k-mooney | likely the same reason for multi attach | |