mirror of
https://github.com/ClickHouse/ClickHouse.git
synced 2024-12-11 00:42:29 +00:00
Store exception generated when we tried to update the queue last time
The use case is to alert when queue contains broken entries. Especially important when ClickHouse breaks backwards compatibility between versions and log entries written by newer versions aren't parseable by old versions. ``` Code: 27, e.displayText() = DB::Exception: Cannot parse input: expected 'quorum: ' before: 'merge_type: 2\n' ```
This commit is contained in:
parent
e83ca20ac3
commit
8b07a7f180
@ -153,11 +153,20 @@ bool ReplicatedMergeTreeRestartingThread::tryStartup()
|
||||
|
||||
storage.cloneReplicaIfNeeded(zookeeper);
|
||||
|
||||
try
|
||||
{
|
||||
storage.queue.load(zookeeper);
|
||||
|
||||
/// pullLogsToQueue() after we mark replica 'is_active' (and after we repair if it was lost);
|
||||
/// because cleanup_thread doesn't delete log_pointer of active replicas.
|
||||
storage.queue.pullLogsToQueue(zookeeper);
|
||||
}
|
||||
catch (...)
|
||||
{
|
||||
storage.last_queue_update_exception.set(std::make_unique<String>(getCurrentExceptionMessage(false)));
|
||||
throw;
|
||||
}
|
||||
|
||||
storage.queue.removeCurrentPartsFromMutations();
|
||||
storage.last_queue_update_finish_time.store(time(nullptr));
|
||||
|
||||
|
@ -3087,21 +3087,24 @@ void StorageReplicatedMergeTree::queueUpdatingTask()
|
||||
last_queue_update_finish_time.store(time(nullptr));
|
||||
queue_update_in_progress = false;
|
||||
}
|
||||
catch (const Coordination::Exception & e)
|
||||
catch (...)
|
||||
{
|
||||
last_queue_update_exception.set(std::make_unique<String>(getCurrentExceptionMessage(false)));
|
||||
tryLogCurrentException(log, __PRETTY_FUNCTION__);
|
||||
|
||||
try
|
||||
{
|
||||
throw;
|
||||
}
|
||||
catch (const Coordination::Exception & e)
|
||||
{
|
||||
if (e.code == Coordination::Error::ZSESSIONEXPIRED)
|
||||
{
|
||||
restarting_thread.wakeup();
|
||||
return;
|
||||
}
|
||||
|
||||
queue_updating_task->scheduleAfter(QUEUE_UPDATE_ERROR_SLEEP_MS);
|
||||
}
|
||||
catch (...)
|
||||
{
|
||||
tryLogCurrentException(log, __PRETTY_FUNCTION__);
|
||||
|
||||
queue_updating_task->scheduleAfter(QUEUE_UPDATE_ERROR_SLEEP_MS);
|
||||
}
|
||||
}
|
||||
@ -5562,6 +5565,10 @@ void StorageReplicatedMergeTree::getStatus(Status & res, bool with_zk_fields)
|
||||
res.total_replicas = 0;
|
||||
res.active_replicas = 0;
|
||||
|
||||
MultiVersion<String>::Version queue_exception = last_queue_update_exception.get();
|
||||
if (queue_exception)
|
||||
res.last_queue_update_exception = *queue_exception;
|
||||
|
||||
if (with_zk_fields && !res.is_session_expired)
|
||||
{
|
||||
try
|
||||
|
@ -174,6 +174,7 @@ public:
|
||||
UInt64 absolute_delay;
|
||||
UInt8 total_replicas;
|
||||
UInt8 active_replicas;
|
||||
String last_queue_update_exception;
|
||||
/// If the error has happened fetching the info from ZooKeeper, this field will be set.
|
||||
String zookeeper_exception;
|
||||
};
|
||||
@ -329,6 +330,7 @@ private:
|
||||
ReplicatedMergeTreeQueue queue;
|
||||
std::atomic<time_t> last_queue_update_start_time{0};
|
||||
std::atomic<time_t> last_queue_update_finish_time{0};
|
||||
MultiVersion<String> last_queue_update_exception;
|
||||
|
||||
DataPartsExchange::Fetcher fetcher;
|
||||
|
||||
|
@ -50,6 +50,7 @@ StorageSystemReplicas::StorageSystemReplicas(const StorageID & table_id_)
|
||||
{ "absolute_delay", std::make_shared<DataTypeUInt64>() },
|
||||
{ "total_replicas", std::make_shared<DataTypeUInt8>() },
|
||||
{ "active_replicas", std::make_shared<DataTypeUInt8>() },
|
||||
{ "last_queue_update_exception", std::make_shared<DataTypeString>() },
|
||||
{ "zookeeper_exception", std::make_shared<DataTypeString>() },
|
||||
}));
|
||||
setInMemoryMetadata(storage_metadata);
|
||||
@ -183,6 +184,7 @@ Pipe StorageSystemReplicas::read(
|
||||
res_columns[col_num++]->insert(status.absolute_delay);
|
||||
res_columns[col_num++]->insert(status.total_replicas);
|
||||
res_columns[col_num++]->insert(status.active_replicas);
|
||||
res_columns[col_num++]->insert(status.last_queue_update_exception);
|
||||
res_columns[col_num++]->insert(status.zookeeper_exception);
|
||||
}
|
||||
|
||||
|
Loading…
Reference in New Issue
Block a user