database change LGTM
@bmarjanovic could you please take it over? Thanks
description_versions table, setting issue_id to the value of the corresponding epic's issue_id, and setting epic_id to null.Not provided
https://console.postgres.ai/gitlab/gitlab-production-main/sessions/49881/commands/148334
Evaluate this MR against the MR acceptance checklist. It helps you analyze changes to reduce risks in quality, performance, reliability, security, and maintainability.
Related to #454439
Vitali Tatarintev (49a575dd) at 17 Mar 08:53
Vitali Tatarintev (503c5d69) at 17 Mar 08:53
Merge branch '592212-purge-version-from-onboarding-status' into 'ma...
... and 1 more commit
Adds a batched background migration to remove the version key from the onboarding_status JSONB column on user_details. This property was introduced for a registration experiment but is no longer used.
Uses the same batched background migration approach and batch sizes as QueueRemoveExperimentsFromUserDetailsOnboardingStatus, which performed the identical operation for the experiments key on the same table.
A follow-up MR will remove the version property from the JSON schema once this migration has been confirmed to have purged all records.
Follows the removal process documented in the wiki.
Relates to #592212
Please evaluate this MR against the MR acceptance checklist.
This is a batched background migration. Each sub-batch executes:
UPDATE user_details
SET onboarding_status = onboarding_status - 'version'
WHERE (onboarding_status ? 'version')
Given 0 rows in production, this executes instantly with no risk of timeout or lock contention.
Absolutely
Causing a 500 error incident for certain customers attempting to view their projects.
master resolution DRI.~"regression:*" label.~"regression:*" label.Thanks @Bixilon
I'm afraid we cannot apply this change, because it was fixed intentionally. I left a comment in the related issue #588801 (comment 3165280848)
I will suggest we continue discussion there. I will close the MR for now, but we can reopen it later if there is a reason for that.
graphql: allow unauthenticated users to query users
This is needed for viewing user profile pages. This should not impose any new risk, as all details are possible to query with html scraping. GraphQL only makes this machine readable.
Maybe removing the whole authorize! parth is better, not sure.
Fixes #588801
GraphQL request:
query {
user(username: "bixilon") {
username
}
}
Evaluate this MR against the MR acceptance checklist. It helps you analyze changes to reduce risks in quality, performance, reliability, security, and maintainability.
It looks like there is a reason why authentication exist Unauthenticated user can enumerate instance use... (#354209 - closed)
That's why it was fixed here Require authentication in User(s)Resolver (!88020 - merged)
I think @ifarkas can provide more context on that.
LGTM
Adds a batched background migration to remove the version key from the onboarding_status JSONB column on user_details. This property was introduced for a registration experiment but is no longer used.
Uses the same batched background migration approach and batch sizes as QueueRemoveExperimentsFromUserDetailsOnboardingStatus, which performed the identical operation for the experiments key on the same table.
A follow-up MR will remove the version property from the JSON schema once this migration has been confirmed to have purged all records.
Follows the removal process documented in the wiki.
Relates to #592212
Please evaluate this MR against the MR acceptance checklist.
This is a batched background migration. Each sub-batch executes:
UPDATE user_details
SET onboarding_status = onboarding_status - 'version'
WHERE (onboarding_status ? 'version')
Given 0 rows in production, this executes instantly with no risk of timeout or lock contention.
Vitali Tatarintev (797cb07b) at 16 Mar 10:03
Merge branch '589138-fix-Lint/EmptyBlock-kwlje' into 'master'
... and 1 more commit
Vitali Tatarintev (e970f763) at 16 Mar 10:02
Fixes non-auto-correctable Lint/EmptyBlock offenses
| Before | After |
|---|---|
Evaluate this MR against the MR acceptance checklist. It helps you analyze changes to reduce risks in quality, performance, reliability, security, and maintainability.
Related to #589138