* Alerting: Add endpoint to revert to a previous alertmanager configuration
This endpoint is meant to be used in conjunction with /api/alertmanager/grafana/config/history to
revert to a previously applied alertmanager configuration. This is done by ID instead of raw config
string in order to avoid secure field complications.
* Add fresh context with timeout and same log properties, re-derive logger
* Unify timeout constants
* Move ctx after shortcut that got added through rebasing
* Unify timeouts
* Port opentracing's SpanFromContext and ContextFromSpan to the grafana tracing package
* Support both opentracing and otel variants
* Better document why we're creating a new ctx
* Add new func to FakeSpan which was added after rebase
* Support grafana-specific traceID key in both tracer implementations
This commit adds a number of limits to the Grafana flavor of the
Prometheus Rules API:
1. `limit` limits the maximum number of Rule Groups returned
2. `limit_rules` limits the maximum number of rules per Rule Group
3. `limit_alerts` limits the maximum number of alerts per rule
It sorts Rule Groups and rules within Rule Groups such that data in the
response is stable across requests. It also returns summaries (totals) for
all Rule Groups, individual Rule Groups and rules.
* WIP
* skip invalid historic configurations instead of erroring
* add warning log when bad historic config is found
* remove unused custom marshaller for GettableHistoricUserConfig
* add id to historic user config, move limit check to store, fix typo
* swagger spec
* Alerting: Respect "For" Duration for NoData alerts
This change modifies `resultNoData` to be more inline with the logic of the other state handlers.
The main effects of this are:
1) NoData states with NoDataState config set to Alerting will respect "For" duration.
2) Prevents zero value in StartsAt and EndsAt for alerts that have only even been in normal state. This includes state transitions from NoDataState=OK and ExecErrState=OK.
3) Better state transition logging.
* define 3 feature toggles for rollout phases
* Pass feature toggles along
* Implement first feature toggle
* Try a different strategy with fall-throughs to specific configurations
* Apply toggle overrides once outside of backend composition
* Emit log messages when we coerce backends
* Run code generator for feature toggle files
* Improve wording in flag descs
* Re-run generator
* Use code-generated constants instead of plain strings
* Use converted enum values rather than strings for pre-parsing
* move export rules to definitions package
* move provisioning contact point methods to provisioning package
* move AlertRuleGroupWithFolderTitle to ngalert models and adapter functions to api's compat
* move rule_types files back to where they were before.
* Remove private labels
* No longer index by instance labels
* Labels are now invariant, only build them once
* Remove bucketing since everything is in a single stream
* Refactor statesToStreams to only return a single unified log stream
* Don't query on labels that no longer exist
* Move selector logic to loki layer, genericize client to work in terms of straight logQL
* Add support for line-level label filters in query
* Combine existing selector tests for better parallelism
* Tests for logQL construction
* Underscore instead of dot for unwrapping labels in logql
* Alerting: Add CustomDetails for PagerDuty
* fix default value for 'severity' from 'error' to 'critical'
* minimal docs for notifiers, specifying config for PagerDuty
* replace notifier -> integration
* replace notifier -> integration
* copy AlertQuery from ngmodels to the definition package
* replaces usages of ngmodels.AlertQuery in API models
* create a converter between models of AlertQuery
---------
Co-authored-by: Alex Moreno <alexander.moreno@grafana.com>
* Encode with snappy, always
* JSON encoder type
* Headers
* Copy labels formatter from promtail
* Implement snappy-proto encoding
* Create encoder interface, test both encoders, choose snappy-proto by default
* Make encoder configurable at the LokiCfg level
* Export both encoders
* Touch up comment and tests
* Drop unnecessary conversions after move to plain strings to appease linter
* Rename RecordStatesAsync to Record
* Rename QueryStates to Query
* Implement fanout writes
* Implement primary queries
* Simplify error joining
* Add test for query path
* Add tests for writes and error propagation
* Allow fanout backend to be configured
* Touch up log messages and config validation
* Consistent documentation for all backend structs
* Parse and normalize backend names more consistently against an enum
* Touch-ups to documentation
* Improve clarity around multi-record blocking
* Keep primary and secondaries more distinct
* Rename fanout backend to multiple backend
* Simplify config keys for multi backend mode
* stop using the scheduler's Update and Delete methods all communication must be via the database
* update scheduler's registry to calculate diff before re-setting the cache
* update fetcher to return the diff generated by registry
* update processTick to update rule eval routine if the rule was updated and it is not going to be evaluated at this tick.
* remove references to the scheduler from api package
* remove unused methods in the scheduler
This commit changes the state package so that errors encountered while
expanding templates for custom labels and annotations are returned
from the function. This is not used at present, but will be used in the
future as we look at how to offer better feedback to users who don't
have access to logs, for example our customers who use Hosted Grafana.
This commit fixes a bug in the $values variable in notification
templates when using Classic Conditions. Since Classic Conditions
are not multi-dimensional, the values of each series that exceeded
the condition should be available as a RefID and offset. For example,
B0, B1, etc. However, this bug meant that instead just a single
condition would be printed as B, not B0.
* Create historian metrics and dependency inject
* Record counter for total number of state transitions logged
* Track write failures
* Track current number of active write goroutines
* Record histogram of how long it takes to write history data
* Don't copy the registerer
* Adjust naming of write failures metric
* Introduce WritesTotal to complement WritesFailedTotal
* Measure TransitionsFailedTotal to complement TransitionsTotal
* Rename all to state_history
* Remove redundant Total suffix
* Increment totals all the time, not just on success
* Drop ActiveWriteGoroutines
* Drop PersistDuration in favor of WriteDuration
* Drop unused gauge
* Make writes and writesFailed per org
* Add metric indicating backend and a spot for future metadata
* Drop _batch_ from names and update help
* Add metric for bytes written
* Better pairing of total + failure metric updates
* Few tweaks to wording and naming
* Record info metric during composition
* Create fakeRequester and simple happy path test using it
* Blocking test for the full historian and test for happy path metrics
* Add tests for failure case metrics
* Smoke test for full annotation persistence
* Create test for metrics on annotation persistence, both happy and failing paths
* Address linter complaints
* More linter complaints
* Remove unnecessary whitespace
* Consistency improvements to help texts
* Update tests to match new descs
* Alerting: Add metrics for active receiver and integrations
Introduces metrics that allows us to track the number of configured receivers and integration in the Alertmanager for all orgs.
As a bonus, I realised that the alert reception metrics where not being exported nor collected. This does that too.
* revert to using folder store from the resolvers
* fixing tests after revert
* api test fixes
---------
Co-authored-by: Kristin Laemmert <mildwonkey@users.noreply.github.com>
* chore(services): replace dependencies on dashboard store with dashboard service
This continues the backend service/store split by replacing dashboard store dependencies with service dependencies. the folder service remains the single exception for now; otherwise we'd have a dependency cycle between the folder and dashboard services. I have some ideas for that, but I'll take care of all the easy parts first.
While doing this, I identified and removed a number of unused arguments from the following functions:
NewFolderNameScopeResolver
NewFolderIDScopeResolver
NewFolderUIDScopeResolver
NewDashboardIDScopeResolver
NewDashboardUIDScopeResolver
resolveDashboardScope
I have a small enterprise PR to support this commit.
* lingering fmt
* Loki backend and client depend on a requester
* Instrument all requests to loki using weaveworks TimedClient
* Construct collector in metrics package
This commit fixes an incorrect comment in the Result struct in eval.go
that I had written some time ago. The comment now documents the
actual behaviour and content of this field.
* Alerting: get alert rules on faults (#61248)
Two functions used to fetch alert rules from DB are updated:
- GetAlertRulesForScheduling
- ListAlertRules
Rows are scanned one by one so good ones are returned.
Common Error is logged with indication how many
rules failed on deserialization.
Resolved: #61248
* updates from review comments
This commit adds filterLabels, filterLabelsRe, removeLabels, and
removeLabelsRe functions to templates for custom labels and annotations.
It allows for use cases such as removing all private labels.
This commit changes the Data struct in template.go to use Labels
instead of map[string]string. It changes how labels are printed
when using {{ .Labels }} from map[foo:bar bar:baz] to
foo=bar, bar=baz.
This commit changes how labels are printed in templates for custom
annotations and labels from map[foo:bar bar:baz] to foo=bar, bar=baz.
Labels are comma separated, and sorted in increasing order.
This commit moves templating from the state package to a sub-package
called template. This sub-package will be the logical package for
future ease-of-use improvements to templating custom annotations
and labels.
* Use existing row struct instead of [2]string, add deserialization helper
* Replace Stream struct with stream struct which is exactly the same
* Drop unused status field
* Don't export queryRes and queryData
* Tests for custom marshalling
* Rename row fields to T and V for consistency with prometheus samples
* Rename row to sample
The `rule_groups_rules` metric is currently defined and computed by `State`.
It makes more sense for this metric to be computed off of the configured rule
set, not based on the rule evaluation state. There could be an edge condition
where a rule does not have a state yet, and so is uncounted.
Additionally, we would like this metric (and others), to have a `rule_group`
label, and this is much easier to achieve if the metric is produced from the
`Scheduler` package.
* Remove Result field from AddDataSourceCommand
* Remove DatasourcesPermissionFilterQuery Result
* Remove GetDataSourceQuery Result
* Remove GetDataSourcesByTypeQuery Result
* Remove GetDataSourcesQuery Result
* Remove GetDefaultDataSourceQuery Result
* Remove UpdateDataSourceCommand Result
* change error to warning, apply correct format
* Update pkg/services/ngalert/store/alertmanager.go
Co-authored-by: George Robinson <george.robinson@grafana.com>
* different wording and data
---------
Co-authored-by: George Robinson <george.robinson@grafana.com>
* Alerting: implement loki query for alert state history
* extract selector building
* add unit tests for selector creation
* backup
* give selectors their own type
* build dataframe
* add some tests
* small changes after manual testing
* use struct client
* golint
* more golint
* Make RuleUID optional for Loki implementation
* Drop initial assumption that we only have one series
* Pare down to three columns, fix timestamp overflows, improve failure cases in loki responses
* Embed structred log lines in the dataframe as objects rather than json strings
* Include state history label filter
* Remove dead code
---------
Co-authored-by: Alex Weaver <weaver.alex.d@gmail.com>
* Alerting: Fix template validation in provisioning api
Fix issue where provisioning API accepts a malformed template having extra
text outside of definition block and template name matching definition name.
* Mark AM configuration as applied
* add missing checks, make linter happy
* fix deadlock, mark as valid on save and on load
* mark configurations only if needed
* check error after applyConfig()
* code review comments
* code review changes
* more code review changes
* clean HistoricConfigFromAlertConfig function
* Define endpoint and generate
* Wire up and register endpoint
* Cleanup, define authorization
* Forgot the leading slash
* Wire up query and SignedInUser
* Wire up timerange query params
* Add todo for label queries
* Drop comment
* Update path to rules subtree
* Extract label merge, add test file
* Extract error/NoData to first class fields, remove a layer from values
* Include dashUID and panelID as line-level fields
* Drop unnecessary object receiver
* Add tests for stream building
* Drop NoData field from log lines
* Set YAML as default value for exporting alert rules
* use YAML format for rule list export
Co-authored-by: Sonia Aguilar <33540275+soniaAguilarPeiron@users.noreply.github.com>
* lint
* Add new format query param to swagger+docs
* Fix broken test
---------
Co-authored-by: Gilles De Mey <gilles.de.mey@gmail.com>
Co-authored-by: Matt Jacobson <matthew.jacobson@grafana.com>
This change exposes more metrics from the embedded Alertmanager, which are
valuable for troubleshooting Alertmanager operation particularly in HA setups.
```
grafana_alerting_notifications_total
grafana_alerting_notifications_failed_total
grafana_alerting_notification_requests_total
grafana_alerting_notification_requests_failed_total
grafana_alerting_notification_latency_seconds
grafana_alerting_nflog_gc_duration_seconds
grafana_alerting_nflog_snapshot_duration_seconds
grafana_alerting_nflog_snapshot_size_bytes
grafana_alerting_nflog_queries_total
grafana_alerting_nflog_query_errors_total
grafana_alerting_nflog_query_duration_seconds
grafana_alerting_nflog_gossip_messages_propagated_total
grafana_alerting_dispatcher_aggregation_groups
grafana_alerting_dispatcher_alert_processing_duration_seconds
```
Note that `alertmanager_dispatcher_aggregation_group_limit_reached_total` is
explicitly not exposed, as the group limit metrics are not enabled.
* Add is_paused attr to the POST alert rule group endpoint
* Add is_paused to alerting API POST alert rule group
* Fixed tests
* Add is_paused to alerting gettable endpoints
* Fix integration tests
* Alerting: allow to pause existing rules (#62401)
* Display Pause Rule switch in Editing Rule form
* add isPaused property to form interface and dto
* map isPaused prop with is_paused value from DTO
Also update test snapshots
* Append '(Paused)' text on alert list state column when appropriate
* Change Switch styles according to discussion with UX
Also adding a tooltip with info what this means
* Adjust styles
* Fix alignment and isPaused type definition
Co-authored-by: gillesdemey <gilles.de.mey@gmail.com>
* Fix test
* Fix test
* Fix RuleList test
---------
Co-authored-by: gillesdemey <gilles.de.mey@gmail.com>
* wip
* Fix tests and add comments to clarify AlertRuleWithOptionals
* Fix one more test
* Fix tests
* Fix typo in comment
* Fix alert rule(s) cannot be paused via API
* Add integration tests for alerting api pausing flow
* Remove duplicated integration test
---------
Co-authored-by: Virginia Cepeda <virginia.cepeda@grafana.com>
Co-authored-by: gillesdemey <gilles.de.mey@gmail.com>
Co-authored-by: George Robinson <george.robinson@grafana.com>
* Fix json serialization of state values
* Simplify two of the tests
* Fix linter complaint
* Don't return error if we fail to look up dashboard, just log it and move on
* Address linter complaint
* Use suggested value for uid
* update the snapshot
* use __expr__
* replace all -100 with __expr__
* update snapshot
* more changes
* revert redundant change
* Use expr.DatasourceUID where it's possible
* generate files
* Allow pausing alerts from provisioning
* Update swagger
* Add IsPaused to provision export endpoints
* Add pause field in sample.yml
* Add exception for reset state in first loop iteration of scheduler if rule is paused
* Update provision definition and swagger docs
* Fix provisioning export tests
* Suggestion: Simplify if condition
* Add more context to a comment
* add nested folder scope inheritance to managed permission services
* add a more specific erorr
* remove circular dependencies
* use errutil for returning erorr
* fix tests
* fix tests
* define a new error in ac package
This adds provisioning endpoints for downloading alert rules and alert rule groups in a
format that is compatible with file provisioning. Each endpoint supports both json and
yaml response types via Accept header as well as a query parameter
download=true/false that will set Content-Disposition to recommend initiating a download
or inline display.
This also makes some package changes to keep structs with potential to drift closer
together. Eventually, other alerting file structs should also move into this new file
package, but the rest require some refactoring that is out of scope for this PR.
* update Delete and Reset methods to return state transitions
this will be used by notifier code to decide whether alert needs to be sent or not.
* update scheduler to provide reason to delete states and use transitions
* update FromAlertsStateToStoppedAlert to accept StateTransition and filter by old state
* fixup
* fix tests
