Thursday, 2025-12-04

*** mhen_ is now known as mhen02:17
opendevreviewMoutaz Chaara proposed openstack/keystone master: Fix role assignment cache for federated users  https://review.opendev.org/c/openstack/keystone/+/96704813:34
stanislav-zI was trying to cherry-pick https://review.opendev.org/c/openstack/keystone/+/939778 + https://review.opendev.org/c/openstack/keystone/+/959279, but it doesn't work as [the implementation of `keystone/common/... (full message at <https://matrix.org/oftc/media/v1/media/download/AYHKN58v1m1BOC3ljtfZEXs1ZN3BN9-82uO-ihMpxOah6ZzAG16QDoTPiD7YsALywll9MbsUQkya8KPJJhyJgglCebMa34owAG1hdHJpeC5vcmcvQUhQcVpwTVJKWEtiUEhwT3pveHBESGRI>)13:37
opendevreviewMoutaz Chaara proposed openstack/keystone master: Fix role assignment cache for federated users  https://review.opendev.org/c/openstack/keystone/+/96704813:42
stanislav-z<stanislav-z> "I was trying to cherry-pick..." <- I don't find it reasonable to implemented it without `passlib`. Whereas with `passlib` it's absolutely trivial - but guess it's not what you want for the upstream.14:23
gtema<stanislav-z> "I don't find it reasonable to..." <- no, this is now absolutely unacceptable to return passlib14:59
tkajinamstanislav-z, passlib depends on crypt module which is not available in Python 3.13 .15:21
tkajinamstanislav-z, see https://foss.heptapod.net/python-libs/passlib/-/issues/148 and https://foss.heptapod.net/python-libs/passlib/-/issues/18715:21
tkajinamre-introducing a dead lib can never be reasonable :-P15:21
tkajinamgtema d34dh0r53, I'm just wondering what's your thoughts about https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/thread/YYIKWMILASUVZH2GO3N5ZGYXK5N7RSAT/#YYIKWMILASUVZH2GO3N5ZGYXK5N7RSAT ?15:23
tkajinam(I don't remember if I threw this question in this channel before I sent it...15:23
gtemaI am not aware of anybody using it. I would personally prefer us dropping them - the whole idea with such auth is not good for me15:24
tkajinamyeah15:26
tkajinamamorin, I wonder if you got chance to check which s3token middleware you use ?15:26
tkajinamIt's not urgent though we may want to deprecate these during this cycle so that we can drop these in next15:26
amorindamn yes I did!15:27
amorinwe are all good, sorry for not answering before15:27
amorinthe one we use is only on swift part15:27
amorinwhich is, IIRC, vendorized / embed15:28
tkajinamamorin, no problem. and that's exactly I expected.15:28
d34dh0r53I'm not aware of anyone using it, I'm in favor of deprecating it given the limited response to your email17:01
opendevreviewStephen Finucane proposed openstack/oslo.limit master: Run mypy from tox  https://review.opendev.org/c/openstack/oslo.limit/+/96905718:08
opendevreviewMerged openstack/oslo.limit master: Delay string interpolations at logging calls  https://review.opendev.org/c/openstack/oslo.limit/+/96944219:45
opendevreviewMerged openstack/oslo.policy master: Delay string interpolations at logging calls  https://review.opendev.org/c/openstack/oslo.policy/+/96944819:58
opendevreviewArnaud Morin proposed openstack/keystone master: Add a new index on revocation_event table  https://review.opendev.org/c/openstack/keystone/+/92973622:20

Generated by irclog2html.py 4.0.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!