DataCore 備份設定 - CDP 備份



CDP 連續資料保護:可持續追蹤紀錄 vdisk 的資料異動,此異動會記錄於 History log 裡,可將 vdisk 回復到以秒為單位的時間點狀態。


How CDP Works

可回復 (Rollback) 時間點 (在 History log 裡保留的異動資料) 會依據 licensed Retention Time 和 History Log 檔大小,不管哪個先到達。寫入的 IO 數量會決定 log 多快會達大所設最大 size。

例如:Retention Time 為 48 小時,但如果 log 檔空間不足,無法容納 48 小時的異動量時,新的異動紀錄會去覆蓋最早的異動紀錄,這樣可能就只能回復到最近幾個小時內的資料。


Retention Time:根據 License,可回復的時間區間,2 小時 ~14 天內。

History Log 檔:預設最大可用是 vdisk 的 5% (預設至少顯示 8GB),手動設定最大可到 1TB。


Important Notes:

       1. CDP must be enabled on the virtual disk in order to create rollbacks from the virtual disk.

       2. After creating rollbacks, stop I/O from the host to the virtual disk immediately. Keep the I/O stopped until the rollback is deleted or split to ensure the restore point does not reach the end of the history log.  

注意在建立 rollback 後,如要避免此 rollback 還原點被覆蓋,請停掉前端 IO 寫入到此 vdisk。或是盡快 split。

      3. If CDP is disabled on a virtual disk, all rollbacks created from the virtual disk will be deleted and space allocated to the history log will be reclaimed. (Allocated disk space is not reclaimed until CDP is disabled.)

如果 CDP 關閉,rollback 就會被清除,然後 log 空間回收。因為會先將 log 的 data 先寫回 vdisk,最後才會刪除、回收 log 空間,所以要注意當下 pool 可用空間,至少要有當下 log 大小空間。

        4. Replacing the storage source of a CDP-enabled virtual disk will disable the CDP feature if the side which has the history log is replaced. After the virtual disk has recovered and data is up-to-date, the feature can be enabled again. Replacing the side which does not also have the history log will not affect the feature for the virtual disk.

Replace 注意:

       5. Rollbacks are created based on the system clock of the DataCore Server that is the source of the rollback, but rollbacks are displayed in localized time in the management console. In other words, two management consoles that are open from two different time zones will show different hours. The absolute time will be the same but because of localization the times in the interfaces will be different.

       6. The SANsymphony-V Management Console works with times taken from the system clock of the DataCore Server that is the source of the rollback. If the system time of the computer running the management console is not synchronized with the system time of the DataCore Server, administrators will notice a difference. For example, when creating a rollback, the oldest rollback time may not be the same as the time displayed in the management console; it will be later or earlier based on the difference in the system clocks of the two servers. The best practice is to synchronize the times of the computers in the SAN before enabling CDP on virtual disks. Note that daylight savings time is taken into consideration and does not affect the rollback.

        7. Changing the system clock on a server with history logs will affect the creation of rollbacks and other CDP operations until the time when the system clock has changed has passed from the history log, which could be up to 48 hours. (This warning only applies to manual changes to the system clock and is not effected by daylight savings time.)



Best Practices

1. CDP requires adequate resources (memory, CPU, disk pool storage performance and disk capacity) and should not be enabled on DataCore Servers with limited resources.

CDP 會需要 memory、CPU、pool storage 效能和空間。

2. Use dedicated pools for CDP-enabled virtual disks and the history logs for those virtual disks. Disk pools used should have sufficient free space at all times. System Health thresholds and email notification via tasks should be configured for notification when disk pool free space reaches the attention threshold to ensure sufficient free space.

3. Enabling CDP for a virtual disk increases the amount of write I/O to that virtual disk. This may increase I/O latency to the disk pools used by the virtual disk and the history log and decrease host I/O performance to virtual disks using these disk pools. CDP should be used with caution to protect mission critical data only.

CDP 啟用後,會增加寫入 IO 的 workload,所以會對效能有影響。建議 CDP 只用在mission critical data 保護。

4. The default history log size (5% of the virtual disk size with a minimum size of 8 GB) may not be adequate for all virtual disks. The history log size should be set according to I/O load and retention time requirements. Once set, the retention period can be monitored and the history log size can be increased if necessary. The current actual retention period for the history log is provided in the Virtual Disk Details>Info Tab (see Retention period).

history log 大小根據 I/O load and retention time 需求。可透過監控去做調整。

5. When copying large amounts of data at one time to newly created virtual disks, enable the feature after copying the data to avoid a significant I/O load.

6. After an event that requires restoration of data, I/O to the affected virtual disk should be immediately suspended and then rollbacks should be created. Follow this practice to keep older data changes from being destaged, which in turn will keep the rollback from expiring or I/O to the virtual disk from failing. Keep I/O suspended until virtual disk recovery is complete.

 (如要 restore,來源 vdisk 先中斷 IO 卸載,再建立 rollback,也可避免要 rollback 的時間點 log 被覆蓋)

7. Rollbacks should only be created for the purpose of finding a consistent condition prior to a disruptive event and restoring the virtual disk data using the best rollback.

8. Delete rollbacks if no longer needed. (rollback 不要用就刪掉)