* Remote Alertmanager: Use the same struct for Grafana stat and Mimir full state
* Alertmanager: Add methods to merge nflog and silences
* update grafana/alerting version
* make update-workspace
* update mocks
* remove unnecesary methods from the remote Alertmanager implementation, create separate StateMerger interface
* (WIP) Remote Alertmanager: Optionally merge remote state before starting the internal Alertmanager
* cleanup ngalert.go
* restore defaults.ini
* move state parsing logic to 'remote' package, clean up ngalert.go
* remove GetBase, implement MegeNflog and MergeSilences
* delete fmt.Println
* FetchRemoteState -> GetRemoteState
* UserGrafanaState -> UserState
* remove duplicate clusterpb import
* reorder MimirClient interface
* use general getState() method for Grafana state and Mimir full state
* remove unnecessary state merging methods from the Alertmanager interface
* remove pullState field
* reduce diff
* add info log after merging
* merge silences and nflog entries in the same method
* merge the remote state in the forked AM
* reduce diff
* update remote AM mock
* tests
* make error more specific
* typo
* deduplicate config preparation logic squashes config preparation in 3 places into a single method buildConfiguration
* remove copying config from decrypt because we already use copy
* move logic from decryptConfiguration to buildConfiguration
* move logic from mergeExtraConfigs to buildConfiguration
* load default config with buildConfiguration method and skip if fails
---------
Co-authored-by: Santiago <santiagohernandez.1997@gmail.com>
What is this feature?
This PR further improves concurrent updates to the provenance table (a follow-up for #101688). The fix explicitly checks for a provenance record using SELECT ... FOR UPDATE for the databases that support it before performing either an update or upsert operation. This preemptive locking reduces the possibility of deadlocks in MySQL.
This works only with the alertingProvenanceLockWrites feature flag enabled.
Why do we need this feature?
The current implementation (directly performing upserts without prior locking) may still encounter deadlocks because of the gap and insert-intention locks in some configurations, for example, with repeatable-read transaction isolation level in MySQL+InnoDB.
---------
Co-authored-by: William Wernert <william.wernert@grafana.com>
* Alerting: Contact point export snapshot test
* Fix Jira.fields hcl export type to allow map[string]any
Since gohcl does not support this type, we marshal as a string instead
which will be handled in the TF provider.
* mapToJSONStringCodec encoder needed for
TestContactPointFromContactPointExports
* parse via sse
I need to figure out how to handle the pipeline.execute with our own
client. I think this is important for MT reasons, just like using our
own cache (via legacy) is important.
parsing is done though!
* WIP nonsense
* horrible code but i think it works
* Add support for sql expressions config settings
* Cleanup:
- remove spew from nodes.go
- uncomment out plugin context and use in single tenant flow
- make code more readable and add comments
* Cleanup:
- create separate file for mt ds client builder
- ensure error handling is the same for both expressions and regular queries
- other cleanup
* not working but good thoughts
* WIP, vector not working for non sse
* super hacky but i think vectors work now
* delete delete delete
* Comments for future ref
* break out query handling and start test
* add prom debugger
* clean up: remove comments and commented out bits
* fix query_test
* add prom debugger
* create table-driven tests with testsdata files
* Fix test
* Add test
* go mod??
* idk
* Remove comment
* go enterprise issue maybe
* Fix codeowners
* Delete
* Remove test data
* Clean up
* logger
* Remove go changes hopefully
* idk go man
* sad
* idk i ran go mod tidy and this is what it wants
* Fix readme, with much help from adam
* some linting and testing errors
* lint
* fix lint
* fix lint register.go
* another lint
* address lint in test
* fix dead code and linters for query_test
* Go mod?
* Struggling with go mod
* Fix test
* Fix another test
* Revert headers change
* Its difficult to test this in OSS as it depends on functionality defined in enterprise, let's bring these tests back in some form in enterprise
* Fix codeowners
---------
Co-authored-by: Adam Simpson <adam@adamsimpson.net>
Update grafana/alerting to 8eef376f49f8
Includes support for OAuth2 in most notifiers but does not enable that support
yet in FE. Fixes tests related to the change.
* replace PostableUserConfig with GrafanaAlertmanagerConfig to decouple from internal Grafana models
* update alertmanager + tests
* calculate hash of the GrafanaAlertmanagerConfig
* introduce ExtraConfigsCrypto
* delete getLastConfiguration and move all code into method Get
* update legacy store to encrypt ExtraConfigs same way we do in MultiOrgAlertmanager
* update legacy store init
* delete PersistConfig
* remove unnecessary unmarshalling
* add better error
* add tests for legacy store
* add tests that cover extra config validation
* add integration test for conflicting routes
---------
Signed-off-by: Yuri Tseretyan <yuriy.tseretyan@grafana.com>
* Alerting: Fix notification policy conflicts originating from provenance mismatch
Sometimes the provenance field on the notification policy route can mismatch with the actual provenance in the provenance store.
This leads to 409s when trying to update the notification policy via the k8s apis.
We avoid this by ignoring the provenance field when calculating the fingerprint.
Co-authored-by: Matt Jacobson <matthew.jacobson@grafana.com>
* fix tests which validate stability
---------
Co-authored-by: Matt Jacobson <matthew.jacobson@grafana.com>
* automatically rename integration tests to follow the common convention
* name tests differently
* alter column type to bigint
* update another column to bigint
* add another alter
* fix subquery for mysql
Several niche bugs have surfaced as a result of the decrypt code Grafana uses in receivers API being different than what is used to decrypt secrets before sending to remote AM. Example:
- Dingding notifier not abiding by new Patching added to local AM, thus causing missing url errors.
* noop refactor to simplify decryptConfiguration
* Move compat function package
* Use new receiver models to encrypt/decrypt in remote AM
What is this feature?
Implements the POST endpoint for deleting imported Mimir Alertmanager configurations:
POST /api/convert/api/v1/alerts
The API endpoint creates the extra Alertmanager configuration with the provided identifier and matchers.
**What is this feature?**
This PR implements a new Prometheus historian backend that allows Grafana alerting to write alert state history as Prometheus-compatible `ALERTS` metrics to remote Prometheus-compatible data sources.
The metric includes a few additional labels:
* `grafana_alertstate`: Grafana's full alert state, more granular than Prometheus.
* `grafana_rule_uid`: Grafana's alert rule UID.
Grafana states are included in the `grafana_alertstate` label also mapped to Prometheus-compatible `alertstate` values:
| Grafana alert state | `alertstate` | `grafana_alertstate` |
|---------------------|-----------------------|-----------------------|
| `Alerting` | `firing` | `alerting` |
| `Recovering` | `firing` | `recovering` |
| `Pending` | `pending` | `pending` |
| `Error` | `firing` | `error` |
| `NoData` | `firing` | `nodata` |
| `Normal` | _(no metric emitted)_ | _(no metric emitted)_ |
This adds the ability to filter rules with the prometheus compatible api using:
1. `receiver_name` to filter by contact point name
2. `health` to filter by the health status of the rule (one of `ok`, `error`, `nodata`, or `unknown`)
This also ensures that groups with no rules (due to filters) are not returned.
* declare dingding url as secret
patch raw settings before parsing because DingDing's config parser does not know about secrets
* fix integration test
---------
Co-authored-by: Yuri Tseretyan <yuriy.tseretyan@grafana.com>
* Add to available channels
* Export
* Fix bug in deeply nested secrets
BE: Slice re-use bug when traversing deeply.
FE: Only at most one level of nesting was being taken into account
when determining secureFields keys. This change adds a new field on
NotificationChannelOption: secureFieldKey. This is populated on API GET via
transform. This change gives us the option to hardcode secureFieldKey in the
backend and no longer calculate the key via settings topology.
* Update grafana/alerting to 3e20fda3b872
* Prettier
* Linting
* Fix IntegrationConfig test to catch secure field mismatch
Alexander Akhmetov
Santiago
Yuri TseretyanandSantiago
Moustafa Baiou
Matthew Jacobson
Sarah ZingerandAdam Simpson
Vadim Stepanov
Ryan McKinley
Fayzal Ghantiwala
Serge Zaitsev
Stephanie Hingtgen
Kevin MinehartandYuri Tseretyan
Pepe Cano