Huawei Cloud Computing Operations and Maintenance
Domain scope
This domain tests day-2 operations on Huawei Cloud: monitoring, alarms, logging, auditing, scaling, backup, and routine maintenance.
Core rules
- CES (Cloud Eye) collects metrics and triggers alarms; OS-level metrics such as memory and disk need the Telescope agent installed on ECS.
- SMN delivers alarm notifications by email, SMS, or HTTP.
- LTS centralizes logs; CTS records API audit trails, not application logs.
- AS scales ECS based on policies; it needs a launch template or image, flavor, subnet, and cool-down to avoid thrashing.
- CBR provides backup and restore for ECS/EVS; DRS replicates databases for migration and DR.
- IAM controls who can operate; CTS answers who did what and when.
Worked example
An ECS group must add capacity when average CPU exceeds 80% for 5 minutes. Create a CES alarm on the AS group metric, set period 5 minutes and trigger count 1, bind an SMN topic for notification, and attach a dynamic scaling policy to the AS group with a cool-down of 300 seconds. Reason: the period avoids reacting to a one-minute spike; the cool-down stops repeated scale-out while new instances warm up. Also set a scale-in policy below 30% to control cost.
Traps
CTS is not a log store for application errors; use LTS. A CES alarm without an SMN subscription is silent. AS cannot scale a database unless the service supports it. Backup and snapshot are not the same as cross-region DR. Installing an agent is required for guest-OS metrics that the hypervisor cannot see.