zfs

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
zfs [2025-06-28 13:20] weichbrzfs [2025-11-27 14:06] (current) weichbr
Line 3: Line 3:
 === Create a pool === === Create a pool ===
  
-   * Get disk ids from ''/dev/disk/by-id''+   * Get disk ids from ''/dev/disk/by-id'', do not use ''/dev/sdX''
    * ''zpool create <name> <type> <disks>''    * ''zpool create <name> <type> <disks>''
-   * ''zpool create ssds mirror nvme-Samsung_SSD_980_TB_ABCD nvme-Samsung_SSD_980_TB_EFGH''+   Example: ''zpool create ssds mirror nvme-Samsung_SSD_980_TB_ABCD nvme-Samsung_SSD_980_TB_EFGH''
    * Don't set ''ashift'', let zfs figure it out. You can check with ''zdb -C <pool>'' what value it uses.    * Don't set ''ashift'', let zfs figure it out. You can check with ''zdb -C <pool>'' what value it uses.
-   * Only set it, if it's wrong. ''ashift=12'' for 4K physical sectors+      * Only set it, if it's wrong. ''ashift=12'' for 4K physical sectors
    * Set ''autotrim=on'' for SSDs    * Set ''autotrim=on'' for SSDs
 +   * When creating a pool ''-o'' is for pool options. To pass options to the root dataset, use ''-O''.
 +
 +A good set of defaults for a new pool:
 +''-o autotrim=on -O acltype=posixacl -O xattr=sa -O normalization=formD -O compression=zstd -O atime=on -O relatime=on''
 +
  
 === Create a dataset === === Create a dataset ===
Line 17: Line 22:
    * Enable comporession: ''-o compression=zstd''    * Enable comporession: ''-o compression=zstd''
    * ''recordsize'' of 128K is default. When storing large files, e.g., media, set this higher, e.g., 1M    * ''recordsize'' of 128K is default. When storing large files, e.g., media, set this higher, e.g., 1M
-    
  • zfs.1751109653.txt.gz
  • Last modified: 2025-06-28 13:20
  • by weichbr