grafana/pkg/services/ngalert/api/lotex_ruler.go

232 lines
5.8 KiB
Go
Raw Normal View History

package api
import (
"bytes"
"fmt"
"io"
"net/http"
"net/url"
"gopkg.in/yaml.v3"
"github.com/grafana/grafana/pkg/api/response"
"github.com/grafana/grafana/pkg/infra/log"
contextmodel "github.com/grafana/grafana/pkg/services/contexthandler/model"
apimodels "github.com/grafana/grafana/pkg/services/ngalert/api/tooling/definitions"
"github.com/grafana/grafana/pkg/web"
)
Alerting: Add support to distinguish Prometheus datasource subtypes (Mimir, Cortex and Vanilla Prometheus) (#46771) * Add basic UI for custom ruler URL * Add build info fetching for alerting data sources * Add keeping data sources build info in the store * Use data source build info to construct data source urls * Remove unused code * Add custom ruler support in prometheus api calls * Migrate actions * Use thunk condition to prevent multiple data source buildinfo fetches * Unify prom and ruler rules loading * Upgrade RuleEditor tests * Upgrade RuleList tests * Upgrade PanelAlertTab tests * Upgrade actions tests * Build info refactoring * Get rid of lotex ruler support action * Add prom ruler availability checking when the buildinfo is not available * Add rulerUrlBuilder tests * Improve prometheus data source validation, small build info refactoring * Change prefix based on Prometheus subtype * Use the correct path * Revert config routing * Add deprecation notice for /api/prom prefix * Add tests to the datasource subtype * Remove custom ruler support * Remove deprecation notice * Prevent fetching ruler rules when ruler api is not available * Add build info tests * Unify naming of ruler methods * Fix test * Change buildinfo data source validation * Use strings for subtype params and unveil mimir * organise imports * frontend changes and wordsmithing * fix test suite * add a nicer verbose message for prometheus datasources * detect Mimir datasource * fix test * fix buildinfo test for Mimir * shrink vectors * add some code documentation * DRY prepareRulesFilterQueryParams * clarify that Prometheus does not support managing rules * Improve buildinfo error handling Co-authored-by: gotjosh <josue.abreu@gmail.com> Co-authored-by: gillesdemey <gilles.de.mey@gmail.com>
2022-04-04 12:30:17 -05:00
const (
Prometheus = "prometheus"
Cortex = "cortex"
Mimir = "mimir"
)
const (
PrometheusDatasourceType = "prometheus"
LokiDatasourceType = "loki"
mimirPrefix = "/config/v1/rules"
prometheusPrefix = "/rules"
lokiPrefix = "/api/prom/rules"
subtypeQuery = "subtype"
)
var dsTypeToRulerPrefix = map[string]string{
Alerting: Add support to distinguish Prometheus datasource subtypes (Mimir, Cortex and Vanilla Prometheus) (#46771) * Add basic UI for custom ruler URL * Add build info fetching for alerting data sources * Add keeping data sources build info in the store * Use data source build info to construct data source urls * Remove unused code * Add custom ruler support in prometheus api calls * Migrate actions * Use thunk condition to prevent multiple data source buildinfo fetches * Unify prom and ruler rules loading * Upgrade RuleEditor tests * Upgrade RuleList tests * Upgrade PanelAlertTab tests * Upgrade actions tests * Build info refactoring * Get rid of lotex ruler support action * Add prom ruler availability checking when the buildinfo is not available * Add rulerUrlBuilder tests * Improve prometheus data source validation, small build info refactoring * Change prefix based on Prometheus subtype * Use the correct path * Revert config routing * Add deprecation notice for /api/prom prefix * Add tests to the datasource subtype * Remove custom ruler support * Remove deprecation notice * Prevent fetching ruler rules when ruler api is not available * Add build info tests * Unify naming of ruler methods * Fix test * Change buildinfo data source validation * Use strings for subtype params and unveil mimir * organise imports * frontend changes and wordsmithing * fix test suite * add a nicer verbose message for prometheus datasources * detect Mimir datasource * fix test * fix buildinfo test for Mimir * shrink vectors * add some code documentation * DRY prepareRulesFilterQueryParams * clarify that Prometheus does not support managing rules * Improve buildinfo error handling Co-authored-by: gotjosh <josue.abreu@gmail.com> Co-authored-by: gillesdemey <gilles.de.mey@gmail.com>
2022-04-04 12:30:17 -05:00
PrometheusDatasourceType: prometheusPrefix,
LokiDatasourceType: lokiPrefix,
}
var subtypeToPrefix = map[string]string{
Prometheus: prometheusPrefix,
Cortex: prometheusPrefix,
Mimir: mimirPrefix,
}
// The requester is primarily used for testing purposes, allowing us to inject a different implementation of withReq.
type requester interface {
withReq(ctx *contextmodel.ReqContext, method string, u *url.URL, body io.Reader, extractor func(*response.NormalResponse) (any, error), headers map[string]string) response.Response
}
type LotexRuler struct {
log log.Logger
*AlertingProxy
requester requester
}
func NewLotexRuler(proxy *AlertingProxy, log log.Logger) *LotexRuler {
return &LotexRuler{
log: log,
AlertingProxy: proxy,
requester: proxy,
}
}
func (r *LotexRuler) RouteDeleteNamespaceRulesConfig(ctx *contextmodel.ReqContext, namespace string) response.Response {
legacyRulerPrefix, err := r.validateAndGetPrefix(ctx)
if err != nil {
return ErrResp(500, err, "")
}
return r.requester.withReq(
ctx,
http.MethodDelete,
withPath(
*ctx.Req.URL,
fmt.Sprintf("%s/%s", legacyRulerPrefix, url.PathEscape(namespace)),
),
nil,
messageExtractor,
nil,
)
}
func (r *LotexRuler) RouteDeleteRuleGroupConfig(ctx *contextmodel.ReqContext, namespace string, group string) response.Response {
legacyRulerPrefix, err := r.validateAndGetPrefix(ctx)
if err != nil {
return ErrResp(500, err, "")
}
return r.requester.withReq(
ctx,
http.MethodDelete,
withPath(
*ctx.Req.URL,
fmt.Sprintf(
"%s/%s/%s",
legacyRulerPrefix,
url.PathEscape(namespace),
url.PathEscape(group),
),
),
nil,
messageExtractor,
nil,
)
}
func (r *LotexRuler) RouteGetNamespaceRulesConfig(ctx *contextmodel.ReqContext, namespace string) response.Response {
legacyRulerPrefix, err := r.validateAndGetPrefix(ctx)
if err != nil {
return ErrResp(500, err, "")
}
return r.requester.withReq(
ctx,
http.MethodGet,
withPath(
*ctx.Req.URL,
fmt.Sprintf(
"%s/%s",
legacyRulerPrefix,
url.PathEscape(namespace),
),
),
nil,
yamlExtractor(apimodels.NamespaceConfigResponse{}),
nil,
)
}
func (r *LotexRuler) RouteGetRulegGroupConfig(ctx *contextmodel.ReqContext, namespace string, group string) response.Response {
legacyRulerPrefix, err := r.validateAndGetPrefix(ctx)
if err != nil {
return ErrResp(500, err, "")
}
return r.requester.withReq(
ctx,
http.MethodGet,
withPath(
*ctx.Req.URL,
fmt.Sprintf(
"%s/%s/%s",
legacyRulerPrefix,
url.PathEscape(namespace),
url.PathEscape(group),
),
),
nil,
yamlExtractor(&apimodels.GettableRuleGroupConfig{}),
nil,
)
}
func (r *LotexRuler) RouteGetRulesConfig(ctx *contextmodel.ReqContext) response.Response {
legacyRulerPrefix, err := r.validateAndGetPrefix(ctx)
if err != nil {
return ErrResp(500, err, "")
}
return r.requester.withReq(
ctx,
http.MethodGet,
withPath(
*ctx.Req.URL,
legacyRulerPrefix,
),
nil,
yamlExtractor(apimodels.NamespaceConfigResponse{}),
nil,
)
}
func (r *LotexRuler) RoutePostNameRulesConfig(ctx *contextmodel.ReqContext, conf apimodels.PostableRuleGroupConfig, ns string) response.Response {
legacyRulerPrefix, err := r.validateAndGetPrefix(ctx)
if err != nil {
return ErrResp(500, err, "")
}
yml, err := yaml.Marshal(conf)
if err != nil {
return ErrResp(500, err, "Failed marshal rule group")
}
u := withPath(*ctx.Req.URL, fmt.Sprintf("%s/%s", legacyRulerPrefix, ns))
return r.requester.withReq(ctx, http.MethodPost, u, bytes.NewBuffer(yml), jsonExtractor(nil), nil)
}
func (r *LotexRuler) validateAndGetPrefix(ctx *contextmodel.ReqContext) (string, error) {
datasourceUID := web.Params(ctx.Req)[":DatasourceUID"]
if datasourceUID == "" {
return "", fmt.Errorf("datasource UID is invalid")
}
Caching: Refactor enterprise query caching middleware to a wire service (#65616) * define initial service and add to wire * update caching service interface * add skipQueryCache header handler and update metrics query function to use it * add caching service as a dependency to query service * working caching impl * propagate cache status to frontend in response * beginning of improvements suggested by Lean - separate caching logic from query logic. * more changes to simplify query function * Decided to revert renaming of function * Remove error status from cache request * add extra documentation * Move query caching duration metric to query package * add a little bit of documentation * wip: convert resource caching * Change return type of query service QueryData to a QueryDataResponse with Headers * update codeowners * change X-Cache value to const * use resource caching in endpoint handlers * write resource headers to response even if it's not a cache hit * fix panic caused by lack of nil check * update unit test * remove NONE header - shouldn't show up in OSS * Convert everything to use the plugin middleware * revert a few more things * clean up unused vars * start reverting resource caching, start to implement in plugin middleware * revert more, fix typo * Update caching interfaces - resource caching now has a separate cache method * continue wiring up new resource caching conventions - still in progress * add more safety to implementation * remove some unused objects * remove some code that I left in by accident * add some comments, fix codeowners, fix duplicate registration * fix source of panic in resource middleware * Update client decorator test to provide an empty response object * create tests for caching middleware * fix unit test * Update pkg/services/caching/service.go Co-authored-by: Arati R. <33031346+suntala@users.noreply.github.com> * improve error message in error log * quick docs update * Remove use of mockery. Update return signature to return an explicit hit/miss bool * create unit test for empty request context * rename caching metrics to make it clear they pertain to caching * Update pkg/services/pluginsintegration/clientmiddleware/caching_middleware.go Co-authored-by: Marcus Efraimsson <marcus.efraimsson@gmail.com> * Add clarifying comments to cache skip middleware func * Add comment pointing to the resource cache update call * fix unit tests (missing dependency) * try to fix mystery syntax error * fix a panic * Caching: Introduce feature toggle to caching service refactor (#66323) * introduce new feature toggle * hide calls to new service behind a feature flag * remove licensing flag from toggle (misunderstood what it was for) * fix unit tests * rerun toggle gen --------- Co-authored-by: Arati R. <33031346+suntala@users.noreply.github.com> Co-authored-by: Marcus Efraimsson <marcus.efraimsson@gmail.com>
2023-04-12 11:30:33 -05:00
ds, err := r.DataProxy.DataSourceCache.GetDatasourceByUID(ctx.Req.Context(), datasourceUID, ctx.SignedInUser, ctx.SkipDSCache)
if err != nil {
return "", err
}
// Validate URL
if ds.URL == "" {
return "", fmt.Errorf("URL for this data source is empty")
}
prefix, ok := dsTypeToRulerPrefix[ds.Type]
if !ok {
return "", fmt.Errorf("unexpected datasource type. expecting loki or prometheus")
}
Alerting: Add support to distinguish Prometheus datasource subtypes (Mimir, Cortex and Vanilla Prometheus) (#46771) * Add basic UI for custom ruler URL * Add build info fetching for alerting data sources * Add keeping data sources build info in the store * Use data source build info to construct data source urls * Remove unused code * Add custom ruler support in prometheus api calls * Migrate actions * Use thunk condition to prevent multiple data source buildinfo fetches * Unify prom and ruler rules loading * Upgrade RuleEditor tests * Upgrade RuleList tests * Upgrade PanelAlertTab tests * Upgrade actions tests * Build info refactoring * Get rid of lotex ruler support action * Add prom ruler availability checking when the buildinfo is not available * Add rulerUrlBuilder tests * Improve prometheus data source validation, small build info refactoring * Change prefix based on Prometheus subtype * Use the correct path * Revert config routing * Add deprecation notice for /api/prom prefix * Add tests to the datasource subtype * Remove custom ruler support * Remove deprecation notice * Prevent fetching ruler rules when ruler api is not available * Add build info tests * Unify naming of ruler methods * Fix test * Change buildinfo data source validation * Use strings for subtype params and unveil mimir * organise imports * frontend changes and wordsmithing * fix test suite * add a nicer verbose message for prometheus datasources * detect Mimir datasource * fix test * fix buildinfo test for Mimir * shrink vectors * add some code documentation * DRY prepareRulesFilterQueryParams * clarify that Prometheus does not support managing rules * Improve buildinfo error handling Co-authored-by: gotjosh <josue.abreu@gmail.com> Co-authored-by: gillesdemey <gilles.de.mey@gmail.com>
2022-04-04 12:30:17 -05:00
// If the datasource is Loki, there's nothing else for us to do - it doesn't have subtypes.
if ds.Type == LokiDatasourceType {
return prefix, nil
}
// A Prometheus datasource, can have many subtypes: Cortex, Mimir and vanilla Prometheus.
// Based on these subtypes, we want to use a different proxying path.
subtype := ctx.Query(subtypeQuery)
subTypePrefix, ok := subtypeToPrefix[subtype]
if !ok {
r.log.Debug(
2022-11-02 17:36:14 -05:00
"Unable to determine prometheus datasource subtype, using default prefix",
"datasource", ds.UID, "datasourceType", ds.Type, "subtype", subtype, "prefix", prefix)
Alerting: Add support to distinguish Prometheus datasource subtypes (Mimir, Cortex and Vanilla Prometheus) (#46771) * Add basic UI for custom ruler URL * Add build info fetching for alerting data sources * Add keeping data sources build info in the store * Use data source build info to construct data source urls * Remove unused code * Add custom ruler support in prometheus api calls * Migrate actions * Use thunk condition to prevent multiple data source buildinfo fetches * Unify prom and ruler rules loading * Upgrade RuleEditor tests * Upgrade RuleList tests * Upgrade PanelAlertTab tests * Upgrade actions tests * Build info refactoring * Get rid of lotex ruler support action * Add prom ruler availability checking when the buildinfo is not available * Add rulerUrlBuilder tests * Improve prometheus data source validation, small build info refactoring * Change prefix based on Prometheus subtype * Use the correct path * Revert config routing * Add deprecation notice for /api/prom prefix * Add tests to the datasource subtype * Remove custom ruler support * Remove deprecation notice * Prevent fetching ruler rules when ruler api is not available * Add build info tests * Unify naming of ruler methods * Fix test * Change buildinfo data source validation * Use strings for subtype params and unveil mimir * organise imports * frontend changes and wordsmithing * fix test suite * add a nicer verbose message for prometheus datasources * detect Mimir datasource * fix test * fix buildinfo test for Mimir * shrink vectors * add some code documentation * DRY prepareRulesFilterQueryParams * clarify that Prometheus does not support managing rules * Improve buildinfo error handling Co-authored-by: gotjosh <josue.abreu@gmail.com> Co-authored-by: gillesdemey <gilles.de.mey@gmail.com>
2022-04-04 12:30:17 -05:00
return prefix, nil
}
2022-11-02 17:36:14 -05:00
r.log.Debug("Determined prometheus datasource subtype",
"datasource", ds.UID, "datasourceType", ds.Type, "subtype", subtype)
Alerting: Add support to distinguish Prometheus datasource subtypes (Mimir, Cortex and Vanilla Prometheus) (#46771) * Add basic UI for custom ruler URL * Add build info fetching for alerting data sources * Add keeping data sources build info in the store * Use data source build info to construct data source urls * Remove unused code * Add custom ruler support in prometheus api calls * Migrate actions * Use thunk condition to prevent multiple data source buildinfo fetches * Unify prom and ruler rules loading * Upgrade RuleEditor tests * Upgrade RuleList tests * Upgrade PanelAlertTab tests * Upgrade actions tests * Build info refactoring * Get rid of lotex ruler support action * Add prom ruler availability checking when the buildinfo is not available * Add rulerUrlBuilder tests * Improve prometheus data source validation, small build info refactoring * Change prefix based on Prometheus subtype * Use the correct path * Revert config routing * Add deprecation notice for /api/prom prefix * Add tests to the datasource subtype * Remove custom ruler support * Remove deprecation notice * Prevent fetching ruler rules when ruler api is not available * Add build info tests * Unify naming of ruler methods * Fix test * Change buildinfo data source validation * Use strings for subtype params and unveil mimir * organise imports * frontend changes and wordsmithing * fix test suite * add a nicer verbose message for prometheus datasources * detect Mimir datasource * fix test * fix buildinfo test for Mimir * shrink vectors * add some code documentation * DRY prepareRulesFilterQueryParams * clarify that Prometheus does not support managing rules * Improve buildinfo error handling Co-authored-by: gotjosh <josue.abreu@gmail.com> Co-authored-by: gillesdemey <gilles.de.mey@gmail.com>
2022-04-04 12:30:17 -05:00
return subTypePrefix, nil
}
func withPath(u url.URL, newPath string) *url.URL {
u.Path, _ = url.PathUnescape(newPath)
u.RawPath = newPath
return &u
}