Reduce severity of some health check issues (#3067)

Author: adameat

ydb/core/health_check/health_check.cpp

@@ -1140,19 +1140,15 @@ class TSelfCheckRequest : public TActorBootstrapped<TSelfCheckRequest> {
     static void Check(TSelfCheckContext& context, const NKikimrWhiteboard::TSystemStateInfo::TPoolStats& poolStats) {
         if (poolStats.name() == "System" || poolStats.name() == "IC" || poolStats.name() == "IO") {
             if (poolStats.usage() >= 0.99) {
-                context.ReportStatus(Ydb::Monitoring::StatusFlag::RED, "Pool usage over 99%", ETags::OverloadState);
+                context.ReportStatus(Ydb::Monitoring::StatusFlag::ORANGE, "Pool usage is over than 99%", ETags::OverloadState);
             } else if (poolStats.usage() >= 0.95) {
-                context.ReportStatus(Ydb::Monitoring::StatusFlag::ORANGE, "Pool usage over 95%", ETags::OverloadState);
-            } else if (poolStats.usage() >= 0.90) {
-                context.ReportStatus(Ydb::Monitoring::StatusFlag::YELLOW, "Pool usage over 90%", ETags::OverloadState);
+                context.ReportStatus(Ydb::Monitoring::StatusFlag::YELLOW, "Pool usage is over than 95%", ETags::OverloadState);
             } else {
                 context.ReportStatus(Ydb::Monitoring::StatusFlag::GREEN);
             }
         } else {
             if (poolStats.usage() >= 0.99) {
-                context.ReportStatus(Ydb::Monitoring::StatusFlag::ORANGE, "Pool usage over 99%", ETags::OverloadState);
-            } else if (poolStats.usage() >= 0.95) {
-                context.ReportStatus(Ydb::Monitoring::StatusFlag::YELLOW, "Pool usage over 95%", ETags::OverloadState);
+                context.ReportStatus(Ydb::Monitoring::StatusFlag::YELLOW, "Pool usage is over than 99%", ETags::OverloadState);
             } else {
                 context.ReportStatus(Ydb::Monitoring::StatusFlag::GREEN);
             }
@@ -1222,7 +1218,7 @@ class TSelfCheckRequest : public TActorBootstrapped<TSelfCheckRequest> {
                             break;
                         case TNodeTabletState::ETabletState::RestartsTooOften:
                             computeTabletStatus.set_state("RESTARTS_TOO_OFTEN");
-                            tabletContext.ReportStatus(Ydb::Monitoring::StatusFlag::RED, "Tablets are restarting too often", ETags::TabletState);
+                            tabletContext.ReportStatus(Ydb::Monitoring::StatusFlag::ORANGE, "Tablets are restarting too often", ETags::TabletState);
                             break;
                         case TNodeTabletState::ETabletState::Dead:
                             computeTabletStatus.set_state("DEAD");
@@ -1261,7 +1257,7 @@ class TSelfCheckRequest : public TActorBootstrapped<TSelfCheckRequest> {
 
         TSelfCheckContext rrContext(&context, "NODE_UPTIME");
         if (databaseState.NodeRestartsPerPeriod[nodeId] >= 30) {
-            rrContext.ReportStatus(Ydb::Monitoring::StatusFlag::RED, "Node is restarting too often", ETags::Uptime);
+            rrContext.ReportStatus(Ydb::Monitoring::StatusFlag::ORANGE, "Node is restarting too often", ETags::Uptime);
         } else if (databaseState.NodeRestartsPerPeriod[nodeId] >= 10) {
             rrContext.ReportStatus(Ydb::Monitoring::StatusFlag::YELLOW, "The number of node restarts has increased", ETags::Uptime);
         } else {