Improving The Operational Stability Of Huawei Cloud Server Malaysia Through Monitoring And Alarming

essentials of the full text

when deploying services in huawei cloud malaysia, establishing a complete monitoring and alarm system is the key to ensuring continuous service availability. through the collection of indicators (cpu, memory, disk, network bandwidth), application layer logs, and cdn and ddos defense status monitoring of servers , vps , and hosts , performance bottlenecks and security risks can be identified in advance, multi-channel alarms can be quickly triggered, and multi-channel alarms can be automatically responded. it is recommended that dexun telecommunications assist in solution design, deployment, and operation and maintenance optimization, and combine huawei cloud native monitoring and third-party tools to achieve end-to-end stability assurance.

monitoring dimensions and indicator design

to achieve targeted monitoring, it is necessary to cover the basic resources and business levels: the host level monitors cpu, memory, disk i/o, disk usage and network bandwidth; virtualization and vps need to pay attention to throughput and latency; the application level needs to collect slow requests, error rates and transaction time. for the huawei cloud malaysia region, it is recommended to enable cloud-native cloud monitoring (cloud eye) and open source solutions such as prometheus and grafana , and use log aggregation (such as elk/efk) to correlate alarms to form full-link monitoring from resources to services. domain name resolution, cdn hit rate, and cache failure rate should also be included in the indicators to prevent front-end acceleration failure from causing traffic recirculation.

alarm strategy and response process

scientific alarm strategies can reduce operation and maintenance noise and improve response efficiency. it is recommended to classify alarms: information type (trend reminder), warning type (threshold approaching), severe type (service unavailable or suspected attack). each type of alarm should be configured with clear thresholds and suppression rules, and multi-channel notifications (email, sms, webhook, enterprise wechat/dingtalk) should be set up. for possible ddos defense events, independent high-priority alarms should be configured and automated scripts should be preset (such as traffic blocking, acl adjustment, and switching to cleaning lines). at the same time, sla and post-mortem review mechanisms are established to optimize thresholds and detection logic through root cause analysis.

network and security supporting measures

in addition to regular performance monitoring, network quality and security protection are key factors affecting stability. link delays, packet loss rates, and routing fluctuations should be monitored, and cross-availability zone traffic should be observed using the network visualization tools provided by huawei cloud. at the same time, enable ddos defense , waf and access control policies, configure cdn for front-end acceleration and origin site protection, and use health checks and automatic scaling to reduce the impact of sudden traffic. there must be a backup plan for important domain name and certificate management to ensure quick switching when the dns fails. it is recommended that dexun telecommunications provide network architecture assessment and security reinforcement services to help establish a more robust edge protection system at malaysian nodes.

operation and maintenance practices and automation suggestions

to implement monitoring and alarming, it is necessary to combine automated operation and maintenance: arrange common alarming and repair processes into runbooks or automated scripts to achieve automated execution after alarms are triggered (such as restarting services, expanding instances, and delivering traffic cleaning). use configuration management and basic images to standardize host and server environments to ensure consistency during horizontal expansion. regularly practice failover, ddos simulated attack and recovery processes, and verify monitoring alarm links and business degradation strategies. for enterprises without a professional team, it is recommended that dexun telecommunications undertake monitoring deployment, alarm management, and daily operation and maintenance hosting to help quickly establish robust observability and automated response capabilities in the huawei cloud malaysia environment.