Release Notes and Announcements

Release Notes

Product Selection Guide

GooseFSx

Product Introduction

Quick Start

Purchase Guide

Console Guide

Tool Guide

Practical Tutorial

Service Level Agreement

Glossary

GooseFS

Product Introduction

Billing Overview

Quick Start

Core Features

Console Guide

Developer Guide

Client Tools

Cluster Configuration Practice

Data Security

Service Level Agreement

GooseFS-Lite

GooseFS-Lite Tool

Practical Tutorial

Use GooseFS in Kubernetes to Speed Up Spark Data

Access Bucket Natively with POSIX Semantics Using GooseFS

GooseFS Distributedload Tuning Practice

FAQs

GooseFS Policy

Data Processing And Security Agreement

GooseFS Cluster Monitoring

PDF

Modo Foco

Tamanho da Fonte

Última atualização: 2025-07-17 17:42:52

Introduction
GooseFS provides users with a cluster-level operation metrics monitoring page. On this page, you can view key monitoring data such as cache capacity, cache hit rate, and read and write request count of the GooseFS cluster.
Note:
To log in to the Tencent Cloud GooseFS console through a sub-account and check monitoring data, you need API access permissions for Tencent Cloud Monitor to acquire monitoring data. Go to the Tencent Cloud Cloud Access Management (CAM) page and refer to the reference document to configure the relevant policy.
Cluster-level monitoring metrics collection and reporting have minute-level latency. Usage data involved in monitoring metrics is for real-time monitoring reference only, not as a basis for metered billing.
Operation Steps
1. Log in to the Data Accelerator GooseFS Console.
2. In the left navigation, click GooseFS Cluster List to enter the GooseFS cluster list page.
3. Click the instance ID of the GooseFS cluster you want to monitor, enter the cluster details page, then click the monitoring information entry in the left-side sidebar to open the monitoring information details page.
4. On the monitoring information details page, you can select the designated monitoring period, time granularity, and filter conditions on demand, and set whether to automatically refresh.
5. The monitoring information details page displays all currently existing monitoring metrics by default. You can click the button icon in the top right corner of the monitoring metric card page as needed to perform operations such as configuring an alarm, displaying in full screen, or exporting the monitoring chart for specific monitoring metrics.
﻿
Metrics Introduction
Basic Metric Name
Description in Chinese
Metric Explanation
cluster_bytes_read_local
Local cluster short-circuit read traffic value (MBytes/Min)
When client nodes and Worker nodes are deployed on the same node, if there is data on the Worker node, the client will read data directly from the local node. This metric is used to record the numeric value of local reading traffic.
cluster_bytes_read_remote
Cluster Remote read traffic (MBytes/Min)
When client nodes and Worker nodes are not in the same node, reading cached data from Worker nodes will occur Remote read traffic. This metric is used to record the traffic value of Remote reading.
cluster_bytes_read_ufs_all
Cluster read UFS traffic value (MBytes/Min)
When a file is not in the GooseFS cluster cache, GooseFS retrieves data from the underlying storage engine. This metric records the traffic of GooseFS reading data from the underlying storage engine.
cluster_bytes_read_ufs_throughput
Cluster read UFS throughput value (MBytes/Min)
The throughput of reading data from the underlying storage engine in a GooseFS cluster.
cluster_bytes_write_ufs_throughput
Cluster write UFS throughput value (MBytes/Min)
Throughput of data writing from the GooseFS cluster to the underlying storage engine.
cluster_bytes_written_remote
Cluster Remote write traffic value (MBytes/Min)
Traffic of client writing data to Worker nodes.
cluster_bytes_written_ufs_all
Cluster write UFS traffic value (MBytes/Min)
In write cache mode, clients write data to the underlying storage engine and Worker nodes; the GooseFS cluster writes data to the underlying storage engine.
cluster_cache_hit
Cluster total Cache hit rate (%)
Cache hit rate = (Local short-circuit read traffic + Cluster Remote read traffic) / Total traffic
cluster_capacity_total
Cluster total cache capacity (MBytes)
The total physical capacity of Worker nodes in a GooseFS cluster, such as each Worker node with 4 disks of 3.68TB and 10 Worker nodes in total, so the total capacity is 147.2TB.
cluster_capacity_used
Cluster total existing cache capacity (MBytes)
Total cached file capacity in the GooseFS cluster.
cluster_local_cache_hit
Cluster local Cache hit rate (%)
Local Cache hit rate = locally short-circuited read traffic / total cluster traffic
cluster_remote_cache_hit
Cluster remote Cache hit rate (%)
Remote Cache hit rate = Cluster remote read traffic / Total traffic
cluster_space_used
Cluster space utilization (%)
Cluster space utilization = Existing cluster cache capacity / Total cluster cache capacity
cluster_workers
Total Worker Node Quantity
Number of managed Worker nodes in the cluster.
master_complete_file_ops
Total number of CompleteFile operations in the cluster
The CompleteFile operation is used to close the file handle.
master_create_directory_ops
Total number of requests for the CreateDirectory operation in the cluster
Create directory operation is used to create a directory.
master_delete_path_ops
Total number of Delete operations in the cluster
The Delete operation is used to delete a file.
master_directories_created
Total successful request count for CreateDirectory operation in cluster
Successful request count for the CreateDirectory operation.
master_file_infos_got
Total successful request count for GetFileInfo operation in cluster
GetFileInfo is used to get file metadata.
master_files_freed
Total successful request count for cluster FreeFile operations
FreeFile is used to release cached files.
master_files_pinned
Number of locked cache files in the cluster
Locked cache files will not be eliminated by the GooseFS cluster.
master_mount_ops
Total number of cluster Mount operations
Mount operations are used to create file system mount points on client nodes.
master_total_paths
Total number of files (including directories) in the cluster
The total file count recorded in the cluster metadata includes directories.
master_unmount_ops
Total number of cluster Unmount requests
Unmount is used to unbind the file system mount point on the client node.

Ajuda e Suporte

Esta página foi útil?

Você também pode entrar em contato com a Equipe de vendas ou Enviar um tíquete em caso de ajuda.

comentários

tencent cloud

Data Accelerator Goose FileSystem

GooseFS Cluster Monitoring

Introduction

Operation Steps

Metrics Introduction

Ajuda e Suporte

Basic Metric Name	Description in Chinese	Metric Explanation
cluster_bytes_read_local	Local cluster short-circuit read traffic value (MBytes/Min)	When client nodes and Worker nodes are deployed on the same node, if there is data on the Worker node, the client will read data directly from the local node. This metric is used to record the numeric value of local reading traffic.
cluster_bytes_read_remote	Cluster Remote read traffic (MBytes/Min)	When client nodes and Worker nodes are not in the same node, reading cached data from Worker nodes will occur Remote read traffic. This metric is used to record the traffic value of Remote reading.
cluster_bytes_read_ufs_all	Cluster read UFS traffic value (MBytes/Min)	When a file is not in the GooseFS cluster cache, GooseFS retrieves data from the underlying storage engine. This metric records the traffic of GooseFS reading data from the underlying storage engine.
cluster_bytes_read_ufs_throughput	Cluster read UFS throughput value (MBytes/Min)	The throughput of reading data from the underlying storage engine in a GooseFS cluster.
cluster_bytes_write_ufs_throughput	Cluster write UFS throughput value (MBytes/Min)	Throughput of data writing from the GooseFS cluster to the underlying storage engine.
cluster_bytes_written_remote	Cluster Remote write traffic value (MBytes/Min)	Traffic of client writing data to Worker nodes.
cluster_bytes_written_ufs_all	Cluster write UFS traffic value (MBytes/Min)	In write cache mode, clients write data to the underlying storage engine and Worker nodes; the GooseFS cluster writes data to the underlying storage engine.
cluster_cache_hit	Cluster total Cache hit rate (%)	Cache hit rate = (Local short-circuit read traffic + Cluster Remote read traffic) / Total traffic
cluster_capacity_total	Cluster total cache capacity (MBytes)	The total physical capacity of Worker nodes in a GooseFS cluster, such as each Worker node with 4 disks of 3.68TB and 10 Worker nodes in total, so the total capacity is 147.2TB.
cluster_capacity_used	Cluster total existing cache capacity (MBytes)	Total cached file capacity in the GooseFS cluster.
cluster_local_cache_hit	Cluster local Cache hit rate (%)	Local Cache hit rate = locally short-circuited read traffic / total cluster traffic
cluster_remote_cache_hit	Cluster remote Cache hit rate (%)	Remote Cache hit rate = Cluster remote read traffic / Total traffic
cluster_space_used	Cluster space utilization (%)	Cluster space utilization = Existing cluster cache capacity / Total cluster cache capacity
cluster_workers	Total Worker Node Quantity	Number of managed Worker nodes in the cluster.
master_complete_file_ops	Total number of CompleteFile operations in the cluster	The CompleteFile operation is used to close the file handle.
master_create_directory_ops	Total number of requests for the CreateDirectory operation in the cluster	Create directory operation is used to create a directory.
master_delete_path_ops	Total number of Delete operations in the cluster	The Delete operation is used to delete a file.
master_directories_created	Total successful request count for CreateDirectory operation in cluster	Successful request count for the CreateDirectory operation.
master_file_infos_got	Total successful request count for GetFileInfo operation in cluster	GetFileInfo is used to get file metadata.
master_files_freed	Total successful request count for cluster FreeFile operations	FreeFile is used to release cached files.
master_files_pinned	Number of locked cache files in the cluster	Locked cache files will not be eliminated by the GooseFS cluster.
master_mount_ops	Total number of cluster Mount operations	Mount operations are used to create file system mount points on client nodes.
master_total_paths	Total number of files (including directories) in the cluster	The total file count recorded in the cluster metadata includes directories.
master_unmount_ops	Total number of cluster Unmount requests	Unmount is used to unbind the file system mount point on the client node.