Backport #69871 to 24.6: add readonly_duration to the system.replicas table

src/Storages/MergeTree/ReplicatedMergeTreeRestartingThread.cpp

@@ -1,3 +1,4 @@
+#include <atomic>
 #include <IO/Operators.h>
 #include <Storages/StorageReplicatedMergeTree.h>
 #include <Storages/MergeTree/ReplicatedMergeTreeRestartingThread.h>
@@ -361,6 +362,13 @@ void ReplicatedMergeTreeRestartingThread::setReadonly(bool on_shutdown)
     bool old_val = false;
     bool became_readonly = storage.is_readonly.compare_exchange_strong(old_val, true);
 
+    if (became_readonly)
+    {
+        const UInt32 now = static_cast<UInt32>(
+            std::chrono::system_clock::to_time_t(std::chrono::system_clock::now()));
+        storage.readonly_start_time.store(now, std::memory_order_relaxed);
+    }
+
     /// Do not increment the metric if replica became readonly due to shutdown.
     if (became_readonly && on_shutdown)
         return;
@@ -393,6 +401,8 @@ void ReplicatedMergeTreeRestartingThread::setNotReadonly()
         CurrentMetrics::sub(CurrentMetrics::ReadonlyReplica);
         chassert(CurrentMetrics::get(CurrentMetrics::ReadonlyReplica) >= 0);
     }
+
+    storage.readonly_start_time.store(0, std::memory_order_relaxed);
 }
 
 }

src/Storages/MergeTree/ReplicatedTableStatus.h

@@ -27,6 +27,7 @@ struct ReplicatedTableStatus
     UInt32 total_replicas;
     UInt32 active_replicas;
     UInt64 lost_part_count;
+    UInt32 readonly_start_time;
     String last_queue_update_exception;
     /// If the error has happened fetching the info from ZooKeeper, this field will be set.
     String zookeeper_exception;

src/Storages/StorageReplicatedMergeTree.cpp

@@ -1,5 +1,6 @@
 #include <Core/Defines.h>
 
+#include <atomic>
 #include <ranges>
 #include <chrono>
 
@@ -6992,6 +6993,7 @@ void StorageReplicatedMergeTree::getStatus(ReplicatedTableStatus & res, bool wit
     res.active_replicas = 0;
     res.lost_part_count = 0;
     res.last_queue_update_exception = getLastQueueUpdateException();
+    res.readonly_start_time = readonly_start_time.load(std::memory_order_relaxed);
 
     if (with_zk_fields && !res.is_session_expired)
     {

src/Storages/StorageReplicatedMergeTree.h

@@ -411,6 +411,8 @@ private:
     /// If true, the table is offline and can not be written to it.
     /// This flag is managed by RestartingThread.
     std::atomic_bool is_readonly {true};
+    std::atomic_uint32_t readonly_start_time{0};
+
     /// If nullopt - ZooKeeper is not available, so we don't know if there is table metadata.
     /// If false - ZooKeeper is available, but there is no table metadata. It's safe to drop table in this case.
     std::optional<bool> has_metadata_in_zookeeper;

src/Storages/System/StorageSystemReplicas.cpp

@@ -4,6 +4,7 @@
 #include <Columns/ColumnString.h>
 #include <DataTypes/DataTypeString.h>
 #include <DataTypes/DataTypesNumber.h>
+#include <DataTypes/DataTypeNullable.h>
 #include <DataTypes/DataTypeDateTime.h>
 #include <DataTypes/DataTypeMap.h>
 #include <Storages/System/StorageSystemReplicas.h>
@@ -200,8 +201,8 @@ StorageSystemReplicas::StorageSystemReplicas(const StorageID & table_id_)
     : IStorage(table_id_)
     , impl(std::make_unique<StorageSystemReplicasImpl>(128))
 {
-    StorageInMemoryMetadata storage_metadata;
-    storage_metadata.setColumns(ColumnsDescription({
+
+    ColumnsDescription description = {
         { "database",                             std::make_shared<DataTypeString>(),   "Database name."},
         { "table",                                std::make_shared<DataTypeString>(),   "Table name."},
         { "engine",                               std::make_shared<DataTypeString>(),   "Table engine name."},
@@ -212,6 +213,7 @@ StorageSystemReplicas::StorageSystemReplicas(const StorageID & table_id_)
         { "can_become_leader",                    std::make_shared<DataTypeUInt8>(),    "Whether the replica can be a leader."},
         { "is_readonly",                          std::make_shared<DataTypeUInt8>(),    "Whether the replica is in read-only mode. This mode is turned on if the config does not have sections with ClickHouse Keeper, "
                                                                                           "if an unknown error occurred when reinitializing sessions in ClickHouse Keeper, and during session reinitialization in ClickHouse Keeper."},
+        { "readonly_start_time", std::make_shared<DataTypeNullable>(std::make_shared<DataTypeDateTime>()), "The timestamp when the replica transitioned into readonly mode. Null if the replica is not in readonly mode." },
         { "is_session_expired",                   std::make_shared<DataTypeUInt8>(),    "Whether the session with ClickHouse Keeper has expired. Basically the same as `is_readonly`."},
         { "future_parts",                         std::make_shared<DataTypeUInt32>(),   "The number of data parts that will appear as the result of INSERTs or merges that haven't been done yet."},
         { "parts_to_check",                       std::make_shared<DataTypeUInt32>(),   "The number of data parts in the queue for verification. A part is put in the verification queue if there is suspicion that it might be damaged."},
@@ -243,7 +245,14 @@ StorageSystemReplicas::StorageSystemReplicas(const StorageID & table_id_)
         { "last_queue_update_exception",          std::make_shared<DataTypeString>(),   "When the queue contains broken entries. Especially important when ClickHouse breaks backward compatibility between versions and log entries written by newer versions aren't parseable by old versions."},
         { "zookeeper_exception",                  std::make_shared<DataTypeString>(),   "The last exception message, got if the error happened when fetching the info from ClickHouse Keeper."},
         { "replica_is_active",                    std::make_shared<DataTypeMap>(std::make_shared<DataTypeString>(), std::make_shared<DataTypeUInt8>()), "Map between replica name and is replica active."}
-    }));
+    };
+
+    description.setAliases({
+        {"readonly_duration", std::make_shared<DataTypeNullable>(std::make_shared<DataTypeDateTime>()), "if(isNull(readonly_start_time), NULL, now() - readonly_start_time)"},
+    });
+
+    StorageInMemoryMetadata storage_metadata;
+    storage_metadata.setColumns(description);
     setInMemoryMetadata(storage_metadata);
 }
 
@@ -515,6 +524,10 @@ Chunk SystemReplicasSource::generate()
         res_columns[col_num++]->insert(status.is_leader);
         res_columns[col_num++]->insert(status.can_become_leader);
         res_columns[col_num++]->insert(status.is_readonly);
+        if (status.readonly_start_time != 0)
+            res_columns[col_num++]->insert(status.readonly_start_time);
+        else
+            res_columns[col_num++]->insertDefault();
         res_columns[col_num++]->insert(status.is_session_expired);
         res_columns[col_num++]->insert(status.queue.future_parts);
         res_columns[col_num++]->insert(status.parts_to_check);

tests/queries/0_stateless/02117_show_create_table_system.reference

@@ -860,6 +860,7 @@ CREATE TABLE system.replicas
     `is_leader` UInt8,
     `can_become_leader` UInt8,
     `is_readonly` UInt8,
+    `readonly_start_time` Nullable(DateTime),
     `is_session_expired` UInt8,
     `future_parts` UInt32,
     `parts_to_check` UInt32,
@@ -888,7 +889,8 @@ CREATE TABLE system.replicas
     `lost_part_count` UInt64,
     `last_queue_update_exception` String,
     `zookeeper_exception` String,
-    `replica_is_active` Map(String, UInt8)
+    `replica_is_active` Map(String, UInt8),
+    `readonly_duration` Nullable(DateTime) ALIAS if(readonly_start_time IS NULL, NULL, now() - readonly_start_time)
 )
 ENGINE = SystemReplicas
 COMMENT 'Contains information and status of all table replicas on current server. Each replica is represented by a single row.'