84669 人学习
152542 人学习
20005 人学习
5487 人学习
7821 人学习
359900 人学习
3350 人学习
180660 人学习
48569 人学习
18603 人学习
40936 人学习
1549 人学习
1183 人学习
32909 人学习
学习是最好的投资！
under LVM2, once the file in the snapshot has been modified, an entry to that effect is made in the exception table, and when the (modified, in-snapshot) file is accessed, the blocks read are the modified snapshot blocks, not those from the original volume ("block ... never gets copied from the original volume")LVM-HOWTO感谢MadHatter
under LVM2, once the file in the snapshot has been modified, an entry to that effect is made in the exception table, and when the (modified, in-snapshot) file is accessed, the blocks read are the modified snapshot blocks, not those from the original volume ("block ... never gets copied from the original volume")
以下回答为本人猜测 注意看图,我觉得最关键的注意点就是快照区与原本的LV共用A~I,也就是说A~I只有一份,而非快照区一份+原LV一份=2份互相独立的区块这种情况. 正常情况下,快照区是用来记录原本LV区域中改动内容的作用存在的,这点看图就能明白. 但有一种特殊情况是鸟哥487页说的用快照区当作测试区,这种情况下用户主动掌控的快照区只有最左面3个空白块,也就是说用户往快照区写的东西只会限定在左侧3块空白处,鸟哥没说要是写入的改动内容大小超过3个空白区块的容量会怎样,我个人估计要是内容超过这个量就会失效
快照区与原本的LV共用A~I
A~I只有一份
快照区一份+原LV一份=2份互相独立的区块
好吧以上内容很难说清楚,鸟哥自己也没讲清,不过我觉得我的理解应该没有漏洞,我都是依据A~I的PE为共用区域这句话推断出来的
A~I的PE为共用区域
under LVM2, once the file in the snapshot has been modified, an entry to that effect is made in the exception table, and when the (modified, in-snapshot) file is accessed, the blocks read are the modified snapshot blocks, not those from the original volume ("block ... never gets copied from the original volume")LVM-HOWTO
感谢MadHatter
以下回答为本人猜测
注意看图,我觉得最关键的注意点就是
快照区与原本的LV共用A~I,也就是说A~I只有一份,而非快照区一份+原LV一份=2份互相独立的区块这种情况.正常情况下,快照区是用来记录原本LV区域中改动内容的作用存在的,这点看图就能明白.
但有一种特殊情况是鸟哥487页说的用快照区当作测试区,这种情况下用户主动掌控的快照区只有最左面3个空白块,也就是说用户往快照区写的东西只会限定在左侧3块空白处,鸟哥没说要是写入的改动内容大小超过3个空白区块的容量会怎样,我个人估计要是内容超过这个量就会失效
好吧以上内容很难说清楚,鸟哥自己也没讲清,不过我觉得我的理解应该没有漏洞,我都是依据
A~I的PE为共用区域这句话推断出来的