| opendevreview | David proposed openstack/watcher-tempest-plugin master: Add tests for workload balance with real data https://review.opendev.org/c/openstack/watcher-tempest-plugin/+/950335 | 08:35 |
|---|---|---|
| opendevreview | Alfredo Moralejo proposed openstack/watcher master: Add debug message to report calculated metric for workload_balance https://review.opendev.org/c/openstack/watcher/+/951639 | 10:01 |
| opendevreview | David proposed openstack/watcher-tempest-plugin master: Add tests for workload_balance with injected data https://review.opendev.org/c/openstack/watcher-tempest-plugin/+/949722 | 10:05 |
| opendevreview | David proposed openstack/watcher-tempest-plugin master: Add tests for workload balance with real data https://review.opendev.org/c/openstack/watcher-tempest-plugin/+/950335 | 10:05 |
| opendevreview | Alfredo Moralejo proposed openstack/watcher-tempest-plugin master: Enforce compute host in create_one_instance_per_host https://review.opendev.org/c/openstack/watcher-tempest-plugin/+/950389 | 11:16 |
| opendevreview | Alfredo Moralejo proposed openstack/watcher-tempest-plugin master: Enforce compute host in create_one_instance_per_host https://review.opendev.org/c/openstack/watcher-tempest-plugin/+/950389 | 11:18 |
| opendevreview | David proposed openstack/watcher-tempest-plugin master: Add tests for workload balance with real data https://review.opendev.org/c/openstack/watcher-tempest-plugin/+/950335 | 12:28 |
| opendevreview | David proposed openstack/watcher-tempest-plugin master: Add tests for workload balance with real data https://review.opendev.org/c/openstack/watcher-tempest-plugin/+/950335 | 12:29 |
| opendevreview | David proposed openstack/watcher master: Adapt zuul.yaml strategies jobs to include tests with tag 'strategy' outside file test_execute_strategies https://review.opendev.org/c/openstack/watcher/+/951649 | 13:05 |
| opendevreview | David proposed openstack/watcher master: Adapt zuul.yaml strategies jobs to include tests with tag 'strategy' outside file test_execute_strategies https://review.opendev.org/c/openstack/watcher/+/951649 | 13:13 |
| opendevreview | David proposed openstack/watcher master: Adapt zuul.yaml strategies jobs to include tests with tag 'strategy' outside file test_execute_strategies https://review.opendev.org/c/openstack/watcher/+/951649 | 13:29 |
| opendevreview | David proposed openstack/watcher-tempest-plugin master: Add tests for workload balance with real data https://review.opendev.org/c/openstack/watcher-tempest-plugin/+/950335 | 13:45 |
| amoralej | hi https://review.opendev.org/c/openstack/watcher-tempest-plugin/+/950969 also has 2 +2, may we get it merged? i hope it will improve ci reliability | 14:02 |
| amoralej | ah, sorry, i missread it | 14:02 |
| amoralej | still only one +2 | 14:03 |
| sean-k-mooney | amoralej: not sure that it the correct way to stablise ci. | 14:20 |
| sean-k-mooney | i left some coments it looks to be at best a workaorund. | 14:20 |
| amoralej | thanks | 14:20 |
| sean-k-mooney | my recomendation is to make the timespan and offset configurable | 14:21 |
| sean-k-mooney | and then update teh gnocii job acordingly | 14:21 |
| sean-k-mooney | a hardcoded 2 minute offset might work but if there is an interaction with the gnoiccii rate and the perfroamcne fo the ci vms we should not hard code it | 14:22 |
| amoralej | ok, making it configurable would make sense | 14:22 |
| opendevreview | David proposed openstack/watcher-tempest-plugin master: Add tests for workload balance with real data https://review.opendev.org/c/openstack/watcher-tempest-plugin/+/950335 | 15:07 |
| opendevreview | David proposed openstack/watcher-tempest-plugin master: Add tests for workload balance with real data https://review.opendev.org/c/openstack/watcher-tempest-plugin/+/950335 | 15:08 |
| opendevreview | Takashi Kajinami proposed openstack/watcher master: devstack: Drop template for mod_wsgi https://review.opendev.org/c/openstack/watcher/+/951662 | 15:14 |
| opendevreview | Joan Gilabert proposed openstack/watcher master: Support zone migration audit without compute_nodes https://review.opendev.org/c/openstack/watcher/+/950665 | 15:19 |
| opendevreview | David proposed openstack/watcher master: Adapt zuul.yaml strategies jobs to include tests with tag 'strategy' outside file test_execute_strategies https://review.opendev.org/c/openstack/watcher/+/951649 | 15:20 |
| opendevreview | Merged openstack/watcher-tempest-plugin master: Change timeframe of fake metrics uploaded to gnocchi. https://review.opendev.org/c/openstack/watcher-tempest-plugin/+/950969 | 15:24 |
| sean-k-mooney | gmaan: if you have time to comment on https://review.opendev.org/c/openstack/watcher-specs/+/949528/comment/354d836d_631424e3/ specificly how we shoudl deal with actions that are skipped if you do an acation show using an api microverion before that state was added | 18:07 |
| sean-k-mooney | the microverion is signalling 2 things in my mind. the presence fo the new state and the aviableity of the new PATCH endpoint for /actions/{id} | 18:08 |
| sean-k-mooney | im wonderign what we shoudl retrun for the state if the action is in skip and we do a get to /actions/{id} for an action in skip with an old microverion | 18:09 |
| sean-k-mooney | we coudl return SKIPPED or we coudl translate it to one of the previously allowed values | 18:09 |
| sean-k-mooney | gmaan: the closet parallel i could draw to nova is the task_state or vm_sate filed in server show | 18:11 |
| sean-k-mooney | i.e. OS-EXT-STS:task_state | 18:11 |
| sean-k-mooney | im not sure we have extened either since we started using micorverions | 18:13 |
| sean-k-mooney | the last time we extened that i think was when we added shelve https://github.com/openstack/nova/commit/c346befc630fd12623d9dc72f9f0c608aaaf2303 | 18:16 |
| sean-k-mooney | if i look at that code i dont think we masked it for older microversoin but that was also newton and i dont know when we intoduced 2.1 off the top of my head | 18:18 |
| sean-k-mooney | hum actully shelve might be older then that we were using micoversion in newton but shelv eexisted before that | 18:20 |
| sean-k-mooney | i guess we have to treat it as unversions given they are just strings | 18:21 |
| opendevreview | Douglas Viroel proposed openstack/watcher master: Add Integrations doc page with support matrix https://review.opendev.org/c/openstack/watcher/+/951699 | 19:14 |
| gmaan | sean-k-mooney: ack, will check | 19:21 |
| sean-k-mooney | gmaan: i think there is no really not good answer os we will jsut have to do the best we can | 19:27 |
| gmaan | sean-k-mooney: replied, I agree with you. here we are just adding new value and does not change field or field's value type so it does not needs to be capped with microversion | 20:41 |
| gmaan | Nova added microversion 2.1 in kilo. For nova shelve case, we added new value in existing field, with same value type. | 20:42 |
| gmaan | if we change field or the type of value it return then we need to take care of change with microversion | 20:43 |
| sean-k-mooney | gmaan: ack i replied in teh spec but thanks for providing your insight | 21:21 |
| sean-k-mooney | that makes things simpler | 21:21 |
| gmaan | np! | 21:21 |
Generated by irclog2html.py 4.0.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!