# What this PR does
## Which issue(s) this PR closes
Fixes
https://github.com/grafana/oncall/issues/5244#issuecomment-2493688544
## Checklist
- [x] Unit, integration, and e2e (if applicable) tests updated
- [x] Documentation added (or `pr:no public docs` PR label added if not
required)
- [x] Added the relevant release notes label (see labels prefixed w/
`release:`). These labels dictate how your PR will
show up in the autogenerated release notes.
# What this PR does
Second, and final part of, https://github.com/grafana/oncall/pull/5277
## Checklist
- [x] Unit, integration, and e2e (if applicable) tests updated
- [x] Documentation added (or `pr:no public docs` PR label added if not
required)
- [x] Added the relevant release notes label (see labels prefixed w/
`release:`). These labels dictate how your PR will
show up in the autogenerated release notes.
# What this PR does
- patch recent migration files which drop several deprecated DB columns
which led to a recent (minor/internal) issue
- add a CI job to prevent this from happening in the future and instead,
force folks to drop columns [the _proper_
way](https://github.com/grafana/oncall/tree/dev/dev#removing-a-nullable-field-from-a-model)
(which we have documented internally)
(as documented
[here](https://github.com/grafana/oncall/tree/dev/dev#removing-a-nullable-field-from-a-model),
I have the four additional migration files (which actually do the `DROP
COLUMN`s in the db) saved locally, and will include these in a separate
PR/release)
## Checklist
- [x] Unit, integration, and e2e (if applicable) tests updated
- [x] Documentation added (or `pr:no public docs` PR label added if not
required)
- [x] Added the relevant release notes label (see labels prefixed w/
`release:`). These labels dictate how your PR will
show up in the autogenerated release notes.
# What this PR does
- `AlertGroupPostMortem` has no references in the codebase.. I stumbled
across it while working on https://github.com/grafana/oncall/pull/5224
and decided to just remove it
- Removing old Slack channel related `VARCHAR` columns; these were
refactored to foreign key references to `slack_slackchannel` table in
following PRs:
- https://github.com/grafana/oncall/pull/5224
- https://github.com/grafana/oncall/pull/5199
- https://github.com/grafana/oncall/pull/5191
## Checklist
- [x] Unit, integration, and e2e (if applicable) tests updated
- [x] Documentation added (or `pr:no public docs` PR label added if not
required)
- [x] Added the relevant release notes label (see labels prefixed w/
`release:`). These labels dictate how your PR will
show up in the autogenerated release notes.
# What this PR does
After deploying
[`r439-v1.12.0`](https://github.com/grafana/oncall-private/releases/tag/r439-v1.12.0)
to staging, I noticed that the migrations were taking a long time, and
caused some wonkiness (see
https://raintank-corp.slack.com/archives/C08063QES5N).
```bash
Apply all migrations: [redacted secret grafana-admin-creds:admin-user], alerts, auth, auth_token, base, contenttypes, email, exotel, fcm_django, google, heartbeat, labels, mobile_app, oss_installation, phone_notifications, schedules, sessions, slack, social_django, telegram, twilioapp, user_management, webhooks, zvonok
Running migrations:
source=engine:app google_trace_id=none logger=apps.alerts.migrations.0063_migrate_channelfilter_slack_channel_id Starting migration to populate slack_channel field.
source=engine:app google_trace_id=none logger=apps.alerts.migrations.0063_migrate_channelfilter_slack_channel_id Bulk updated 1 ChannelFilters with their Slack channel.
source=engine:app google_trace_id=none logger=apps.alerts.migrations.0063_migrate_channelfilter_slack_channel_id Finished migration to populate slack_channel field.
Applying alerts.0063_migrate_channelfilter_slack_channel_id... OK
source=engine:app google_trace_id=none logger=apps.alerts.migrations.0064_migrate_resolutionnoteslackmessage_slack_channel_id Starting migration to populate slack_channel field.
source=engine:app google_trace_id=none logger=apps.alerts.migrations.0064_migrate_resolutionnoteslackmessage_slack_channel_id Bulk updated 1 ResolutionNoteSlackMessage records with their Slack channel.
source=engine:app google_trace_id=none logger=apps.alerts.migrations.0064_migrate_resolutionnoteslackmessage_slack_channel_id Finished migration to populate slack_channel field.
Applying alerts.0064_migrate_resolutionnoteslackmessage_slack_channel_id... OK
source=engine:app google_trace_id=none logger=apps.schedules.migrations.0019_auto_20241021_1735 Starting migration to populate slack_channel field.
source=engine:app google_trace_id=none logger=apps.schedules.migrations.0019_auto_20241021_1735 Bulk updated 6 OnCallSchedules with their Slack channel.
source=engine:app google_trace_id=none logger=apps.schedules.migrations.0019_auto_20241021_1735 Finished migration to populate slack_channel field.
Applying schedules.0019_auto_20241021_1735... OK
source=engine:app google_trace_id=none logger=apps.user_management.migrations.0026_auto_20241017_1919 Starting migration to populate default_slack_channel field.
source=engine:app google_trace_id=none logger=apps.user_management.migrations.0026_auto_20241017_1919 Bulk updated 1 organizations with their default Slack channel.
source=engine:app google_trace_id=none logger=apps.user_management.migrations.0026_auto_20241017_1919 Finished migration to populate default_slack_channel field.
Applying user_management.0026_auto_20241017_1919... OK
```
**NOTE**: wrt these migrations already being run for certain OSS stacks;
it shouldn't have much of an impact on OSS deployments, as it's really
only an issue for _very large_ versions of these tables (particularly
the `ResolutionNoteSlackMessage` table, which by its nature, has a
tendency to generate a lot of data).
## Checklist
- [ ] Unit, integration, and e2e (if applicable) tests updated
- [ ] Documentation added (or `pr:no public docs` PR label added if not
required)
- [ ] Added the relevant release notes label (see labels prefixed w/
`release:`). These labels dictate how your PR will
show up in the autogenerated release notes.
# What this PR does
- add missing db migration files generated via `python manage.py
makemigrations`
- fail the `lint-migrations-backend-mysql-rabbitmq` GitHub Actions CI
job if there are missing Django database migration files
# What this PR does
Adds mobile app push notifications for shift swap requests.
## Which issue(s) this PR fixes
https://github.com/grafana/oncall/issues/2630
## Checklist
- [x] Unit, integration, and e2e (if applicable) tests updated
- [x] Documentation added (or `pr:no public docs` PR label added if not
required)
- [x] `CHANGELOG.md` updated (or `pr:no changelog` PR label added if not
required)
# What this PR does
**Shift Swap Requests**
https://www.loom.com/share/860c3337b338412cbd2ac4024260f3e8?sid=3d91b558-b4de-4351-8b45-8a99b7302346
**Other**
- Drastically improve the typing in the `slack` Django app, and several
other models/functions that were consumed by logic within the `slack`
Django app (ex. setting `RelatedManager` type hints on various models)
https://www.loom.com/share/da6b9984519c48d59a45d3c93c08d7dc
## Checklist
- [x] Unit, integration, and e2e (if applicable) tests updated
- [x] Documentation added (or `pr:no public docs` PR label added if not
required)
- [x] `CHANGELOG.md` updated (or `pr:no changelog` PR label added if not
required)
# What this PR does
This PR should allow us to start working on _most_ of the remaining
tasks for this feature set.
- Adds a basic `ShiftSwapRequest` model + CRUD endpoints.
- Adds a `POST /api/internal/v1/shift_swaps/<id>/take` endpoint which
allows a benefactor to take a request (only when certain conditions
about the ssr are met)
Closes#2587
## Checklist
- [x] Unit, integration, and e2e (if applicable) tests updated
- [x] Documentation added (or `pr:no public docs` PR label added if not
required) will be done in #2589
- [x] `CHANGELOG.md` updated (or `pr:no changelog` PR label added if not
required) (will update once we ship the finalized feature set)
# What this PR does
Sometimes plugin sync fails with the following exception:
```
Cannot delete or update a parent row: a foreign key constraint fails (`schedules_oncallschedule`, CONSTRAINT `alerts_oncallschedul_team_id_4e633f4b_fk_user_mana` FOREIGN KEY (`team_id`) REFERENCES `user_management_team` (`id`))'
```
How to reproduce:
1. Create a new Grafana team
2. Create two schedules with different types (e.g. ICal and Web) and
assign both schedules to the new team
3. Delete the team in Grafana
4. Trigger plugin sync, the sync will fail with the exception above
This happens because the `OnCallSchedule` Django model is a polymorphic
model and there's a [known
bug](https://github.com/django-polymorphic/django-polymorphic/issues/229)
in `django-polymorphic` with deleting related objects when using
`SET_NULL` and `CASCADE`. This PR adds non-polymorphic versions of
`SET_NULL` and `CASCADE` to use in schedule FKs as per this
[comment](https://github.com/django-polymorphic/django-polymorphic/issues/229#issuecomment-398434412).
This also applies to two other schedule FKs: `organization` and
`user_group`, which are not working properly as well.
## Checklist
- [x] Unit, integration, and e2e (if applicable) tests updated
- [x] Documentation added (or `pr:no public docs` PR label added if not
required)
- [x] `CHANGELOG.md` updated (or `pr:no changelog` PR label added if not
required)
Related to #828
- Enable web UI for API/Terraform schedules to add overrides
- Refactor backend to add a flag toggling between web-based and
iCal-based overrides (these options are mutually exclusive)
Also updated read-only tooltips (related to #1483)
# What this PR does
This PR adds
[django-migration-linter](https://github.com/3YOURMIND/django-migration-linter)
to keep database migrations
backwards compatible
- we can automatically run migrations and they are zero-downtime, e.g.
old code can work with the migrated database
- we can run and rollback migrations without worrying about data safety
- OnCall is deployed to the multiple environments core team is not able
to control
See [django-migration-linter
checklist](https://github.com/3YOURMIND/django-migration-linter/blob/main/docs/incompatibilities.md)
for the common mistakes and best practices
## Which issue(s) this PR fixes
## Checklist
- [ ] Tests updated
- [ ] Documentation added
- [ ] `CHANGELOG.md` updated
---------
Co-authored-by: Joey Orlando <joey.orlando@grafana.com>
* Centralize timezone validation into one spot + add serializer validation
for schedules and oncall shifts (both public and internal API)
* add engine-manage make command
