Monday, 2026-06-01

*** darmach10026 is now known as darmach100200:25
opendevreviewGrzegorz Grasza proposed openstack/keystone stable/2025.2: Move keystone-tempest-federation to noble  https://review.opendev.org/c/openstack/keystone/+/99089509:36
xek^ this should unblock the gate09:37
opendevreviewArtem Goncharov proposed openstack/keystone stable/2025.2: Add audience mapper to devstack Keycloak client  https://review.opendev.org/c/openstack/keystone/+/99061809:57
fricklerxek: thanks, I think this ^^ will to be squashed to make all jobs pass at the same time, but we can wait for CI results first09:58
frickler*will need to09:58
opendevreviewMerged openstack/keystone master: Fix user impersonation through application credentials (CVE-2026-42998)  https://review.opendev.org/c/openstack/keystone/+/99048611:08
opendevreviewMerged openstack/keystone master: Forbid trust operations using application credentials (CVE-2026-43000)  https://review.opendev.org/c/openstack/keystone/+/99048711:08
opendevreviewMerged openstack/keystone master: Preserve expires_at when rescoping federated tokens (CVE-2026-44394)  https://review.opendev.org/c/openstack/keystone/+/99048811:08
opendevreviewMerged openstack/keystone master: Prevent RBAC policy bypass via JSON body and query filters (CVE-2026-42999)  https://review.opendev.org/c/openstack/keystone/+/99048911:08
opendevreviewMerged openstack/keystone stable/2026.1: Enforce delegation project boundary for delegated tokens  https://review.opendev.org/c/openstack/keystone/+/99049011:08
fricklerxek: gtema: in addition to the above two fixes for 2025.2, it looks like another one is needed for keystone-protection-functional failing in keystone_tempest_plugin.tests.rbac.v3.test_limit.DomainAdminTests.test_identity_list_limits https://zuul.opendev.org/t/openstack/build/4668c5606b1a43c8b77aff60659ff55b11:11
opendevreviewGrzegorz Grasza proposed openstack/keystone stable/2025.2: Make keystone-protection-functional non-voting on stable/2025.2  https://review.opendev.org/c/openstack/keystone/+/99090811:51
opendevreviewGrzegorz Grasza proposed openstack/keystone stable/2025.2: Make keystone-protection-functional non-voting on stable/2025.2  https://review.opendev.org/c/openstack/keystone/+/99090811:54
opendevreviewGrzegorz Grasza proposed openstack/keystone stable/2025.2: Update hard-coded policy for GET /v3/limits  https://review.opendev.org/c/openstack/keystone/+/99091412:17
opendevreviewGrzegorz Grasza proposed openstack/keystone stable/2025.2: Update hard-coded policy for GET /v3/limits  https://review.opendev.org/c/openstack/keystone/+/99091412:21
opendevreviewArtem Goncharov proposed openstack/keystone stable/2025.2: Add audience mapper to devstack Keycloak client  https://review.opendev.org/c/openstack/keystone/+/99061812:22
opendevreviewMerged openstack/keystone stable/2026.1: Fix user impersonation through application credentials (CVE-2026-42998)  https://review.opendev.org/c/openstack/keystone/+/99049112:23
mgariepytobias-urdin: do you mind if i update your patch for last weeks cves to depends on https://review.opendev.org/c/openstack/keystone/+/990631 ? 12:30
mgariepyon the 2024.1 branch that is.12:31
opendevreviewMerged openstack/keystone stable/2026.1: Forbid trust operations using application credentials (CVE-2026-43000)  https://review.opendev.org/c/openstack/keystone/+/99049212:37
tobias-urdinmgariepy: no, please do, the patches is part of large chain since they contain the older keystone cve fixes as well – please rebase the entire chain so we can get all of them merged if they look good12:47
mgariepyi wonder i adding the depends on the first patch that fails will fix the whole chain.12:50
fricklermgariepy: if all changes are in the same repo and branch, doing a git rebase would be seen as the better solution mostly, makes the stack more obvious. depends-on is better used only for cross-repo or cross-branch deps12:53
fricklerand you'll want to trigger rechecks for all changes anyway, which this will implicitly do12:53
mgariepywell i added the depend and rebases all the changes :) 12:54
fricklerah, I missed the updates because gerritbot doesn't mention them here. the stack actually starts at https://review.opendev.org/c/openstack/keystone/+/987061/1 and that will also need the gate fix. but maybe let's see CI results first now12:57
mgariepyho, i didnt saw the first 2 were passing because they were almost a month old. :/12:59
opendevreviewMerged openstack/keystone stable/2026.1: Preserve expires_at when rescoping federated tokens (CVE-2026-44394)  https://review.opendev.org/c/openstack/keystone/+/99049313:00
opendevreviewMerged openstack/keystone stable/2026.1: Prevent RBAC policy bypass via JSON body and query filters (CVE-2026-42999)  https://review.opendev.org/c/openstack/keystone/+/99049413:00
mgariepyfrickler: do you think it will pass anyway ? or should i undo and rebase the base?13:10
mgariepythe real question is more, will all of them be check together on merge or they will be tested independently ?13:15
bbobrovI somehow missed the discussion around https://review.opendev.org/q/I8f59c59b4edd233e173274f1979e9a3ff0f3cfa5. Why was it decided to fix it this way? I am still not switched to the "new defaults", and maintain old-style policy, where "project admin" is confined to their own project, and "cloud admin" is something else, and now my project admins become superadmins for limits?13:19
mgariepypep8.. 80 > 79. /o\13:34
fricklerbbobrov: I don't know how the original patch came about, but I agree this is questionable for a stable branch. cc gtema xek 14:54
mgariepydo you have a guide to format the pep8 failing tests? some fails are only the comma or double-quote at the end of the line that exceed the 79 char limit.15:14
mgariepy 15:14
bbobrovmgariepy: have you run tox -e pep8 locally?15:19
mgariepyhttps://review.opendev.org/c/openstack/keystone/+/990607?tab=change-view-tab-header-zuul-results-summary15:20
mgariepyjust reading the logs from pep8 test15:20
bbobrovyes, but have you ran it locally?15:20
mgariepynop.15:20
bbobrovon your computer15:20
bbobrovtry it!15:20
bbobrovit will autoformat everything for you15:20
mgariepyi'll check it a bit later not currently setup for this it's not something i ususally do ;)15:25
bbobrovhmmm maybe it will not work on 2024.1 though.15:28
mgariepynot sure why the backport of the patch is not passing.. but pass on other branches.15:29
fricklermaybe the switch to using pre-commit for pep8 did change this. stephenfin might know?15:37
*** Unknown123 is now known as Mike--17:43
mgariepymaster has also 80 char on the patch for theses lines and it passes. did the check moved away from 79char limit ?17:45
*** Unknown123 is now known as Mike--17:52
opendevreviewAde Lee proposed openstack/keystone-specs master: spec: support caller-specified IDs for projects, users, and domains  https://review.opendev.org/c/openstack/keystone-specs/+/99099120:45
opendevreviewAde Lee proposed openstack/keystone-specs master: spec: support caller-specified IDs for projects, users, and domains  https://review.opendev.org/c/openstack/keystone-specs/+/98344020:51

Generated by irclog2html.py 4.1.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!