Merge pull request #1746 from rancher/bugfixes

Bugfixes

internal/cortex/config/validation/limits.proto

@@ -5,6 +5,12 @@ import "github.com/rancher/opni/internal/codegen/cli/cli.proto";
 import "google/protobuf/duration.proto";
 option go_package = "github.com/rancher/opni/internal/cortex/config/validation";
 option (cli.generator) = { generate: true, generate_deepcopy: true };
+message DisabledRuleGroup {
+  // namespace in which the rule group belongs
+  optional string namespace = 1 [(cli.flag) = { skip: true }];
+  // name of the rule group
+  optional string name = 2 [(cli.flag) = { skip: true }];
+}
 message Limits {
   // Per-user ingestion rate limit in samples per second.
   optional double ingestion_rate = 1 [(cli.flag) = { default: "25000" }];
@@ -116,7 +122,7 @@ message Limits {
   optional bool alertmanager_receivers_firewall_block_private_addresses = 54 [(cli.flag) = { default: "false" }];
   // Per-user rate limit for sending notifications from Alertmanager in notifications/sec. 0 = rate limit disabled. Negative value = no notifications are allowed.
   optional double alertmanager_notification_rate_limit = 55 [(cli.flag) = { default: "0" }];
-  // Per-integration notification rate limits. Value is a map, where each key is integration name and value is a rate-limit (float). On command line, this map is given in JSON format. Rate limit has the same meaning as -alertmanager.notification-rate-limit, but only applies for specific integration. Allowed integration names: webhook, email, pagerduty, opsgenie, wechat, slack, victorops, pushover, sns, telegram, discord, webex.
+  // Per-integration notification rate limits. Value is a map, where each key is integration name and value is a rate-limit (float). On command line, this map is given in JSON format. Rate limit has the same meaning as -alertmanager.notification-rate-limit, but only applies for specific integration. Allowed integration names: webhook, email, pagerduty, opsgenie, wechat, slack, victorops, pushover, sns, telegram, discord, webex, msteams.
   map<string, double> alertmanager_notification_rate_limit_per_integration = 56 [(cli.flag) = { default: "{}" }];
   // Maximum size of configuration file for Alertmanager that tenant can upload via Alertmanager API. 0 = no limit.
   optional int32 alertmanager_max_config_size_bytes = 57 [(cli.flag) = { default: "0" }];
@@ -130,6 +136,8 @@ message Limits {
   optional int32 alertmanager_max_alerts_count = 61 [(cli.flag) = { default: "0" }];
   // Maximum total size of alerts that a single user can have, alert size is the sum of the bytes of its labels, annotations and generatorURL. Inserting more alerts will fail with a log message and metric increment. 0 = no limit.
   optional int32 alertmanager_max_alerts_size_bytes = 62 [(cli.flag) = { default: "0" }];
+  // list of rule groups to disable
+  repeated DisabledRuleGroup disabled_rule_groups = 63 [(cli.flag) = { skip: true }];
 }
 message RelabelConfig {
   repeated string source_labels = 1;

pkg/clients/stats_unix.go

@@ -23,17 +23,9 @@ type ConnStats struct {
 	Raw       unix.TCPInfo
 }
 
-// Computes the tx/rx throughput by comparing against a previous snapshot.
-func (c *ConnStats) CalcThroughput(since ConnStats) (txRate uint64, rxRate uint64) {
-	dt := c.Timestamp.Sub(since.Timestamp).Seconds()
-	if dt <= 0 {
-		return 0, 0
-	}
-	dsent := (c.Raw.Bytes_sent - since.Raw.Bytes_sent)
-	drecvd := (c.Raw.Bytes_received - since.Raw.Bytes_received)
-	txRate = uint64(float64(dsent) / dt)
-	rxRate = uint64(float64(drecvd) / dt)
-	return
+// Returns the socket throughput in bytes per second
+func (c *ConnStats) DeliveryRate() uint64 {
+	return c.Raw.Delivery_rate
 }
 
 func (c *ConnStats) RTT() time.Duration {

pkg/update/syncer.go

@@ -42,7 +42,6 @@ func (conf SyncConfig) DoSync(ctx context.Context) error {
 	statsDone := make(chan struct{})
 	go func() {
 		defer close(statsDone)
-		var prevStats *clients.ConnStats
 		startTime := time.Now()
 		var printStats func(string)
 		if conf.StatsClient == nil {
@@ -60,24 +59,20 @@ func (conf SyncConfig) DoSync(ctx context.Context) error {
 					lg.With(zap.Error(err)).Warn("failed to query connection stats")
 					return
 				}
-				if prevStats == nil {
-					prevStats = &stats
-					return
+				rate, err := util.Humanize(stats.DeliveryRate())
+				if err != nil {
+					rate = "0"
 				}
-				_, rx := stats.CalcThroughput(*prevStats)
-				rxStr, _ := util.Humanize(rx)
 				recvdStr := stats.HumanizedBytesReceived()
 				elapsedTime := time.Since(startTime)
 				mins := elapsedTime / time.Minute
 				elapsedTime -= mins * time.Minute
 				secs := elapsedTime / time.Second
 				elapsedTime -= secs * time.Second
 				millis := elapsedTime / time.Millisecond
-				lg.Debugf("%s%s | %sB/s | %02d:%02d.%03d", msg, recvdStr, rxStr, mins, secs, millis)
-				prevStats = &stats
+				lg.Debugf("%s%sB | %sB/s | %02d:%02d.%03d", msg, recvdStr, rate, mins, secs, millis)
 			}
 		}
-		printStats("")
 		for {
 			select {
 			case <-ticker.C:

plugins/metrics/apis/cortexops/extensions.go

@@ -59,7 +59,27 @@ func init() {
 
 				for _, preset := range presets.Items {
 					if preset.GetId().GetId() == fromPreset {
-						_, err := client.SetConfiguration(cmd.Context(), preset.GetSpec())
+						spec := preset.GetSpec()
+						if cmd.Flags().Lookup("interactive").Value.String() == "true" {
+							drr, err := client.DryRun(cmd.Context(), &DryRunRequest{
+								Target: driverutil.Target_ActiveConfiguration,
+								Action: driverutil.Action_Set,
+								Spec:   spec,
+							})
+							if err != nil {
+								return err
+							}
+							modified := drr.GetModified()
+							comments := []string{}
+							for _, err := range drr.GetValidationErrors() {
+								comments = append(comments, err.GetMessage())
+							}
+							if modified, err = cliutil.EditInteractive(modified, comments...); err != nil {
+								return err
+							}
+							spec = modified
+						}
+						_, err := client.SetConfiguration(cmd.Context(), spec)
 						return err
 					}
 				}

plugins/metrics/pkg/gateway/drivers/opni_manager/opni_manager.go

@@ -77,11 +77,12 @@ func (methods) FillConfigFromObject(obj *opnicorev1beta1.MonitoringCluster, conf
 }
 
 // FillObjectFromConfig implements crds.ValueStoreMethods.
-func (methods) FillObjectFromConfig(obj *opnicorev1beta1.MonitoringCluster, conf *cortexops.CapabilityBackendConfigSpec) {
+func (m methods) FillObjectFromConfig(obj *opnicorev1beta1.MonitoringCluster, conf *cortexops.CapabilityBackendConfigSpec) {
 	obj.Spec.Cortex.Enabled = conf.Enabled
 	obj.Spec.Cortex.CortexConfig = conf.CortexConfig
 	obj.Spec.Cortex.CortexWorkloads = conf.CortexWorkloads
 	obj.Spec.Grafana.GrafanaConfig = conf.Grafana
+	obj.Spec.Gateway.Name = m.controllerRef.GetName()
 }
 
 func NewOpniManagerClusterDriver(options OpniManagerClusterDriverOptions) (*OpniManager, error) {