ECS Disk Snapshots Overview
Created#More Posted time:Jan 28, 2018 17:03 PM
• Snapshots are used to backup and recovery of your ECS cloud data.
• Snapshot rollback enables quick data recovery.
• Snapshots can be manual or automatic
• Snapshots can also be created via Console or APIs.
• Snapshots are taken Live, but disk I/O performance may be affected up to 10%, however OS/services running on ECS are not affected/disrupted.
• Snapshot are taken as First “Full” then “Incremental” afterwards, incremental snapshots are based on changes to disks (system or data), usually incremental snapshot of system disk is finished within 5 minutes, and incremental snapshot of data disk is finished within 10-15 minutes (again it depends on amount of changes!)
• Snapshots don’t delete any data, they just record the changes on disk and capture them
• Automatic snapshot policies can be customized for ECS disks (system and data disks)
• The current snapshot policy execution time defaults to Beijing Time (UTC+8). Please keep this in mind when setting up your snapshot policy.
• List of snapshots taken (Manual or via Automatic policy) and their capacity “actual utilization” and timestamp can be viewed from “Snapshot Chain menu (console)”.
• ECS Snapshot 2.0 data service provides a quota of 64 snapshots for each disk. When the maximum number of snapshots is reached, the oldest snapshot will be deleted to make space for new snapshots.
• If the time to take a snapshot exceeds the time between automatic snapshots, for instance if the snapshot is extremely large, new automatic snapshots will be skipped until the current snapshot completes. For example, suppose a user sets up automatic snapshots at 9:00, 10:00, and 11:00. If the snapshot at 9:00 takes 70 minutes to complete, the snapshot at 10:00 will be skipped, and the next automatic snapshot will take place as scheduled at 11:00.
[abdulhafeez edited the post at Jan 28, 2018 23:24 PM]