news
BSD and Linux Kernel: ZFS, ZPool, and fsync()
-
Dan Langille ☛ zfs: setting compression and adding new vdevs
If you read my recent posts, I’m replacing an existing zpool with new devices. At first, I went to copy the old zpool to a new zpool. I then decided instead of copying, to replace. I’m also going to see about compression. I’m sure it won’t take effect, because replace is a block-by-block copy, or so I think. We’ll confirm.
-
Dan Langille ☛ Moving a zpool to new devices – after the syncoid copy – oh wait, zfs replace
The new devices are staying in this host. It is better for me to add the new devices to the zpool and remove the old devices. That is so much easier. The existing zpool can stay in use and there is no interruption in service.
NOTE: after publishing this entry, I realized the command is zpool replace, not zfs replace, as the title suggests.
-
Dan Langille ☛ Creating the new zpools
I have 4 new storage devices to create 2 new zpools, each a two-vdev mirror. Let’s go.
-
University of Toronto ☛ A (filesystem) journal can be a serialization point for durable writes
Suppose that you have a filesystem that uses some form of a journal to provide durability (as many do these days) and you have a bunch of people (or processes) writing and updating things all over the filesystem that they want to be durable, so these processes are all fsync()'ing their work on a regular basis (or the equivalent system call or synchronous write operation). In a number of filesystem designs, this creates a serialization point on the filesystem's journal.