Weasel disk partition plan: Difference between revisions
No edit summary |
No edit summary |
||
Line 42: | Line 42: | ||
| big, network backup | | big, network backup | ||
|} | |} | ||
The main idea is to partition disk failures, so I don't want to bunch all the disks together into a single volume group. When one disk goes bad, don't want to lose access to part of another disk just because it is in the same volume group. |
Revision as of 05:57, 25 June 2008
Currently I have two partitions, root and big. They are both managed by lvm and look like this:
root@weasel:/boot# df -h Filesystem Size Used Avail Use% Mounted on /dev/mapper/Ubuntu-root 224G 107G 106G 51% / /dev/mapper/vg00-big 230G 187G 32G 86% /big
It would be much more reasonable to have a small root partition that is easily backed up and separate partitions for /home, /store, and /big.
/store is a shared network-accessible space. /big is used for automatic backups of root, /home, and /store.
/store is 99G, /home is 3.4G, /var is 1.2G, root is 107G, so if I would split them apart, root could be as small as 4G.
It would be handy if the root partition could be kept small enough to be backed up on a DVD.
So the new scheme would have 4 physical volumes:
Device | Size | Use |
---|---|---|
/dev/hda | 250G | root, store, home, var |
/dev/hdb | 250G | big, network storage |
/dev/hdc | 500G | root backup |
/dev/hdd | 500G | big, network backup |
The main idea is to partition disk failures, so I don't want to bunch all the disks together into a single volume group. When one disk goes bad, don't want to lose access to part of another disk just because it is in the same volume group.