Tuesday, 2017-07-11

*** gcb has quit IRC00:21
*** sapcc-bot has quit IRC01:37
*** carthaca_ has quit IRC01:37
*** sapcc-bot has joined #openstack-manila01:38
*** carthaca_ has joined #openstack-manila01:38
*** gcb has joined #openstack-manila02:03
openstackgerritRyan Liang proposed openstack/manila master: VNX: bump the version for Pike  https://review.openstack.org/48147302:39
*** dsariel has quit IRC03:09
*** jmlowe has joined #openstack-manila03:58
openstackgerritzhongjun proposed openstack/manila master: Test  https://review.openstack.org/48240204:00
openstackgerritRajat Sharma proposed openstack/manila master: Switch from oslosphinx to openstackdocstheme  https://review.openstack.org/48154304:19
*** arnewiebalck_ has joined #openstack-manila04:33
*** markstur has quit IRC04:55
*** jprovazn has joined #openstack-manila04:57
*** dsariel has joined #openstack-manila05:33
*** jprovazn has quit IRC05:39
*** lpetrut has joined #openstack-manila05:41
*** arnewiebalck_ has quit IRC05:49
*** dsariel has quit IRC05:56
*** pcaruana has joined #openstack-manila06:04
*** dsariel has joined #openstack-manila06:06
*** jprovazn has joined #openstack-manila06:08
*** td has quit IRC06:09
*** zhonghua has quit IRC06:11
*** zhonghua has joined #openstack-manila06:12
*** td has joined #openstack-manila06:18
*** jmlowe has quit IRC06:25
*** dsariel has quit IRC06:30
*** lpetrut has quit IRC06:39
*** arnewiebalck_ has joined #openstack-manila06:45
*** belmoreira has joined #openstack-manila06:46
*** arnewiebalck_ has quit IRC06:46
*** arnewiebalck_ has joined #openstack-manila06:49
*** openstackgerrit has quit IRC07:33
*** a-pugachev has joined #openstack-manila07:55
vponomaryovjprovazn: pong07:55
jprovaznvponomaryov: hi, wanted to check your comment on https://review.openstack.org/#/c/313549/07:56
jprovaznvponomaryov: you want explicit db tests that the table is present on upgrade and table is gone on downgrade?07:57
vponomaryovjprovazn: test of Db schema change07:58
vponomaryovjprovazn: as it is done for other existing DB migrations07:58
jprovaznvponomaryov: in this case db schema change consists of just adding a new db, my understanding of https://github.com/openstack/manila/blob/master/manila/tests/db/migrations/alembic/migrations_data_checks.py is that it's purpose is testing migrations which contain some extra logic (e.g. altering columns). "as it is done for other existing DB migrations" - we don't test upgrade/downgrade if it's just create/drop of a table (w/o any08:01
jprovazn logic)08:01
jprovazns/a new db/a new table/08:01
vponomaryovjprovazn: what do you mean "we don't"? Upgrade part checks that table exists and we can write to table, downgrade part checks that table is absent08:03
vponomaryovjprovazn: we DO test it for existing migrations08:03
vponomaryovjprovazn: example with other proposed feature: https://review.openstack.org/#/c/452158/33/manila/tests/db/migrations/alembic/migrations_data_checks.py08:05
*** raissa has joined #openstack-manila08:12
*** a-pugachev has joined #openstack-manila08:13
jprovaznvponomaryov: ok, fair enough.08:15
*** Manor has joined #openstack-manila08:15
*** openstackgerrit has joined #openstack-manila08:27
*** Manor_ has joined #openstack-manila08:47
*** Manor has quit IRC08:51
*** lpetrut has joined #openstack-manila09:04
openstackgerritzhongjun proposed openstack/manila master: Test  https://review.openstack.org/48240209:11
*** arnewiebalck_ has quit IRC09:15
*** Manor has joined #openstack-manila09:19
*** Manor_ has quit IRC09:22
*** raissa has quit IRC09:25
*** Manor has quit IRC10:00
openstackgerritThomas Bechtold proposed openstack/manila master: Allow docs build without git  https://review.openstack.org/48249410:08
openstackgerritMayur Indalkar proposed openstack/manila master: Implementation of Manila driver for Veritas Access  https://review.openstack.org/47219010:10
*** gcb has quit IRC10:13
*** raissa has joined #openstack-manila10:25
openstackgerritAlexander Pugachev proposed openstack/python-manilaclient master: Covers manilaclient/common/apiclient/auth.py:discover_auth_systems().  https://review.openstack.org/48210810:26
openstackgerritAlexander Pugachev proposed openstack/python-manilaclient master: Increases code coverage  https://review.openstack.org/48210810:27
*** gouthamr has quit IRC10:33
*** gouthamr has joined #openstack-manila10:41
*** ganso has joined #openstack-manila10:43
openstackgerritJan Provaznik proposed openstack/manila master: User Messages  https://review.openstack.org/31354910:56
*** jprovazn has quit IRC10:56
*** dsariel has joined #openstack-manila11:03
*** rraja has joined #openstack-manila11:09
*** jprovazn has joined #openstack-manila11:17
*** jprovazn has quit IRC11:18
*** jprovazn has joined #openstack-manila11:18
*** jprovazn has quit IRC11:18
*** jprovazn has joined #openstack-manila11:18
openstackgerritThomas Bechtold proposed openstack/manila master: Allow docs build without git  https://review.openstack.org/48249411:55
*** raissa_ has joined #openstack-manila12:00
*** raissa has quit IRC12:00
*** dustins has joined #openstack-manila12:09
*** MVenesio has joined #openstack-manila12:32
*** openstackgerrit has quit IRC12:47
tbarronvponomaryov: need to get this fix (thank you) backported to stable/ocata and stable/newton: https://review.openstack.org/#/c/482308/12:53
tbarronvponomaryov: please give it your blessing12:53
vponomaryovdone12:55
vponomaryovjprovazn: ping13:06
jprovaznvponomaryov, hi13:08
vponomaryovhello13:08
vponomaryovjprovazn: why in https://review.openstack.org/#/c/313549/56/manila_tempest_tests/tests/api/admin/test_user_messages.py in some tests you use line "messages = body['messages']" in some other don't?13:08
jprovaznvponomaryov, IIRC there is a nasty hack in manila tempest helper which checks whether size of items in reponse is 1 and if so it returns this single item instead of list of items, /me searches for the link13:12
vponomaryovjprovazn: I know about tempest13:13
vponomaryovjprovazn: it means that you API returns more than 1 arg13:13
vponomaryovjprovazn: for the same API13:13
vponomaryovjprovazn: we expect 2 keys only for share_type_list13:13
vponomaryovjprovazn: which has "volume_types" and "share_types"13:14
vponomaryovjprovazn: I guess the problem is in here: https://review.openstack.org/#/c/313549/56/manila/api/views/messages.py13:15
vponomaryovjprovazn: "messages_links" is added13:15
vponomaryovjprovazn: I really doubt it is by design13:16
jprovaznvponomaryov, not sure what you mean?13:18
vponomaryovjprovazn: I mean that your code violates agreed spec for this feature, that is defined here: https://specs.openstack.org/openstack/manila-specs/specs/pike/user-messages.html#rest-api-impact13:18
vponomaryovwhere we have only one top-level key - messages13:18
vponomaryovalso, "detail_id" is absent there -> https://review.openstack.org/#/c/313549/56/manila/db/sqlalchemy/models.py13:19
vponomaryovjprovazn: ^13:19
vponomaryovtbarron: ^13:19
jprovaznvponomaryov, detail_id is on line 120013:20
vponomaryovI am saying it is absent in spec13:20
vponomaryovwe have "detail" in spec13:20
vponomaryovand it is 255 chars long13:20
vponomaryovnot 1013:20
jprovaznvponomaryov, re spec difference - ack - so are you suggesting to remove message_links because it differs from spec?13:21
vponomaryovjprovazn: first of all, the reason for it is not clear13:21
vponomaryovjprovazn: second, yes, spec is different13:21
vponomaryovjprovazn: it is completely unclear what is "detail_id"13:21
vponomaryovthat longs 10 chars13:21
jprovaznvponomaryov, so there is a parallel user messages patch for cinder tommylikehu is working on - IIRC detail -> detail_id came from my intention to keep both versions more aligned13:22
vponomaryovjprovazn: "id" fields refer to some table13:23
vponomaryovyou don't have new table for details13:23
vponomaryovwhich means, it is etiher uncompleted commit or it has wrong name and size of column for detail13:24
jprovaznvponomaryov, in this case detail_id is ID of a detail which is mapped to detail message13:24
vponomaryovmapped to which table?13:24
vponomaryovyou added only one -> https://review.openstack.org/#/c/313549/56/manila/db/sqlalchemy/models.py13:24
jprovaznhttps://review.openstack.org/#/c/313549/56/manila/message/message_field.py13:26
jprovaznmapping is done in message_field, I don't mean reference on db level13:27
*** openstackgerrit has joined #openstack-manila13:27
openstackgerritMerged openstack/python-manilaclient master: Updated from global requirements  https://review.openstack.org/47808013:27
*** gcb has joined #openstack-manila13:32
vponomaryovjprovazn: ok, let it be so, there are other concerns, left comments13:33
vponomaryov jprovazn: in commit13:33
jprovaznvponomaryov, ok, thanks, I'll look in a moment13:34
jprovaznvponomaryov, re: "contradicts spec" comments - are these comments just statement of the fact or are do you expect update of the patch? IMO these deviations are minor, not a dramatic violation of the spec. Spec is a proposal at the time of writing, IMO doing minor changes during development is fine.13:44
vponomaryovjprovazn: comments for action_id and detail_id are for other reviewers, it will be enough just to update notes before these lines where is it taken from13:48
*** eharney has joined #openstack-manila13:48
jprovaznvponomaryov, ok, I suppose you are ok with keeping message_links too?13:52
vponomaryovjprovazn: either keep this key always or remove it13:52
vponomaryovjprovazn: when we don't have messages key "messages" always exists  , just with empty value13:53
vponomaryovempty list*13:53
jprovaznvponomaryov, I think it makes sense to keep it there unconditionally, on the other side if I grep other APis it seems it's usually added conditionally13:56
jprovazn(which is the reason I added condition too)13:57
jprovaznvponomaryov, looking closer, it seems that consensus across manila and cinder is to return xxx_links conditionally (only if the query used 'limit' param) - I would lean to be consistent with the rest14:06
bswartztbarron: the newton backport is having issues14:26
*** xyang1 has joined #openstack-manila14:27
vponomaryovjprovazn: if so, then need to make tempest safe14:33
zhongjuntbarron: xyang1 : Addressed your comments and responses some comments. Could you take a look at again. Thanks https://review.openstack.org/#/c/462468/14:41
xyang1zhongjun: sure14:44
zhongjunxyang1: :)14:45
*** openstackgerrit has quit IRC14:48
*** openstackgerrit has joined #openstack-manila14:49
openstackgerritTom Barron proposed openstack/manila-ui master: DO NOT MERGE  https://review.openstack.org/48259114:49
tbarronbswartz: thanks, looks like I need a separate fix in tests in stable/newton14:50
tbarronzhongjun: will try to look this afternoon; also, on the client side patch please make sure we discuss the proposed change from the spec from '~' to '_' in the weekly meeting.14:51
*** jmlowe has joined #openstack-manila14:56
*** markstur has joined #openstack-manila15:06
zhongjuntbarron: ok, we should discuss it15:09
*** catintheroof has joined #openstack-manila15:28
*** YuYangWang has quit IRC15:30
*** belmoreira has quit IRC15:38
*** lpetrut has quit IRC15:56
*** jmlowe has quit IRC15:57
*** eharney has quit IRC16:00
*** jmlowe has joined #openstack-manila16:00
*** markstur has quit IRC16:01
*** markstur has joined #openstack-manila16:02
*** jmlowe has quit IRC16:22
*** dsariel has quit IRC16:49
*** a-pugachev has quit IRC17:12
*** lpetrut has joined #openstack-manila17:19
*** YuYangWang has joined #openstack-manila17:30
tbarronvponomaryov: ganso bswartz https://review.openstack.org/#/c/482572/17:40
*** eharney has joined #openstack-manila17:40
tbarron^^^ stable/newton version of that manila-ui fix needed a couple tweaks17:40
*** raissa_ has quit IRC17:49
*** lpetrut has quit IRC17:51
*** eharney has quit IRC18:05
*** lpetrut has joined #openstack-manila18:13
*** eharney has joined #openstack-manila18:18
*** rraja has quit IRC18:18
*** lpetrut has quit IRC18:47
openstackgerritDavid Hill proposed openstack/manila master: Define meta_value to 'none' if it's not defined  https://review.openstack.org/48267219:04
*** arnewiebalck_ has joined #openstack-manila19:06
openstackgerritBen Swartzlander proposed openstack/python-manilaclient master: Enable IPv6 in manila(client)  https://review.openstack.org/32893219:52
openstackgerritBen Swartzlander proposed openstack/python-manilaclient master: Enable IPv6 in manila(client)  https://review.openstack.org/32893219:58
bswartztbarron: ping20:01
tbarronbswartz: pong20:01
bswartztbarron: I noticed something weird today20:01
bswartzI can do IPv6 mounts on xenial...20:02
tbarronnice!20:02
bswartzI'm not sure what changed20:02
tbarronwell, me either but that's a good thing.20:03
bswartztbarron: I wonder if the debian or ubuntu maintainers backported the IPv6 bits to the older package20:04
tbarronipv6 marches forward, or lurches20:04
bswartzI don't have time to dig through the changelogs but it seems that unless I'm really confused we might be able to test ipv6 in our existing jobs20:04
tbarronbswartz: so you are starting to test zhongjun's patches20:04
bswartzwell you probably saw the client one needed some updating ^^^20:05
bswartzbut with those updates I think it's working now20:05
bswartzjust need to add unit tests to my patches and get the whole thing merged20:06
bswartzalso I have some more functional testing to do20:06
bswartzbut I'm feeling good about his20:07
bswartzabout this20:07
tbarronbswartz: sorry, got disconnected, but yeah, just roll with it and let's get this stuff tested.  the centos lvm job is there if we need it.20:11
*** MVenesio has quit IRC20:25
*** arnewiebalck_ has quit IRC20:44
openstackgerritDavid Hill proposed openstack/manila master: Define meta_value to 'none' if it's not defined  https://review.openstack.org/48267220:46
*** carthaca_ has quit IRC21:04
*** sapcc-bot has quit IRC21:04
*** sapcc-bot has joined #openstack-manila21:04
*** carthaca_ has joined #openstack-manila21:04
*** catintheroof has quit IRC21:04
*** jmlowe has joined #openstack-manila21:05
*** dustins has quit IRC21:22
*** pcaruana has quit IRC21:23
openstackgerritDavid Hill proposed openstack/manila master: Define meta_value to 'none' if it's not defined  https://review.openstack.org/48267221:24
*** jprovazn has quit IRC21:43
*** eharney has quit IRC21:47
*** jmlowe has quit IRC21:57
*** jmlowe has joined #openstack-manila22:01
*** jmlowe has quit IRC22:09
*** ganso has quit IRC23:21

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!