* Alerting: Implement arbitrary data source remote writer.
Recording rules currently remote write to the endpoint speficied in the
configuration file, but we want to change this so that it can be specified
on a per-rule basis. This change implements the updated RemoteWriter
interface, and dispatches writes to the correct data source.
Notes about the implementation:
- There is a defaulting behaviour, because we will need to handle situations
when upgrading from older Grafana versions, where recording rules do not a
data source target specified.
- In order to avoid re-creating the remote write clients for every request,
a cache is used. The entries have a TTL of 30 seconds at which point they
are recreated; this is the delay for picking up new data source settings.
Known limitations:
- There is no per data source configuration for the remote write path suffix.
- Not all data source settings will be used, but basic auth and TLS will be.
- There is no RBAC. This will be a follow up feature.
* Lint
Simplify, by retaining quotes when escaping query
We also don't need to convert to string when passing to fmt.Sprintf
a slice of Bytes is accepted by Sprintf instead.
* Alerting: Refactor NewPrometheusWriter function.
In order to re-use PrometheusWriter, changing the function take a
PrometheusWriterConfig instead of RecordingRulesSettings, and adapt the old
interface onto the new interface.
* Make linter happy
* Adds ability to run integration tests against spanner (by using GRAFANA_TEST_DB=spanner env variable. SPANNER_DB variable then specifies database to use: spannertest, emulator or string like /projects/<project>/instances/<instance>/databases/<db>)
* Adds feature to migration dialects to create database from a snapshot, instead of running individual migrations.
* Adds first version of Spanner snapshot, prepared from "OSS" migrations.
* Uses generated bit-reversed-positive values instead of auto_increment. (As an experiment)
* Clean up authenticator
* Cleanup authorizers and replace org_id and stack_id with namespace authorizer
* Remove dependency on org service
* Extract orgID from /apis/ urls and validate stack id
Extend the recording rule definition to include the target data source, allowing
configuration of where the output of the recording rule is written to. Also
extends the relevant interfaces in preparation for the next set of changes.
* Chore: Bump golang.org/x/net to v0.36.0
* Chore: Use github.com/moby/moby version 27.5.1 instead of 26.0.0
* Chore: Bump github.com/ua-parser/uap-go to v0.0.0-20250213224047-9c035f085b90
* Chore: Bump github.com/grpc-ecosystem/go-grpc-middleware/v2 to fix 32-bit overflow issue
* add feature toggle alertRuleRestore
* Update delete rule to require UserUID, remove all versions and create "delete" version that holds information about who and when deleted the rule
* serviceaccounts/secretscan: test Service more thoroughly
* middleware/cookies: add tests for CookieOptions
* anonymous/anonimpl: cover a couple more methods
* components/imguploader: Implement WebDAV integration tests
* components/apikeygen: also check IsValid method
* bus: cover invalid callback signature cases
* cloudmigration/objectstorage: add basic unit tests
* login/social/connectors: add test case for GitHub OAuth fetch emails+orgs
* expr/classic: cover more evaluator types in tests
* feat(featuremgmt): introduce feature toggle for enabling sri checks
* feat(frontend): use assetSriChecks feature toggle to inject integrity hash into script tags
* chore(webpack): align sri algorithms across dev and prod builds
* docs(featuremgmt): update assetSriChecks to pass CI
* docs(featuremgmt): fix more spelling complaints with assetSriChecks
* Add crossorigin attribute
* chore(webpack): add subresource-integrity plugin
* build(webpack): wrap webpack jsonp loader integrity checks in feature flag checks
* revert(index.html): remove crossorigin attribute if assertSriChecks is disabled
---------
Co-authored-by: Kristian Bremberg <kristian.bremberg@grafana.com>