6

I recently installed OpenIndiana because of an interest in ZFS. I like the feature-list of ZFS, but I'm having a hard time wrapping my head around how to configure it. All of the documentation I have come across so far seems to presuppose an understanding of concepts like pools and file systems. Where can I find introductory-level information that defines these concepts in terms of ZFS?

ewwhite
  • 201,205

3 Answers3

5

Two things...

The most important document for you to read is the Oracle Solaris ZFS Administration Guide. It provides a comprehensive description of ZFS features and terminology. It also gives excellent examples.

Beyond that, you may want to look at NexentaStor as an option if you're primarily interested in the storage aspects of ZFS. It's a storage-oriented appliance solution centered around ZFS. Other than that, use the OpenIndiana mailing lists.

ewwhite
  • 201,205
1

One more thing you have to be aware of is that ZFS exists in various versions. Due to Oracle's shiny new closed source policy, some features that exist in Solaris 11 did not make it into OpenIndiana. For example, ZFS in Solaris 11 has encryption, whereas OpenIndiana does not implement this (yet). See the zfs upgrade -v and zpool upgrade -v commands to see the ZFS version history and the version you're on. When playing with the zfs and zpool commands on whatever OS and version, it always makes sense to refer to the manpage or zfs help command of the system you're currently on (and not to some possibly outdated guide).

When it comes to the general concepts, though, both "forks" of ZFS are still very similar.

By the way: Here's a more recent version of the Oracle Solaris ZFS Administration Guide (targeting Solaris 11). The one posted by ewwhite seems to be a previous revision.

If deduplication is one of the features you're interested in (as most ZFS beginners are), I'd advise you to read this excellent blog post on ZFS dedup first, before making any premature plans. I've had to learn the hard way that dedup might be great in theory, but doesn't work out for me in practice. Once you've had dedup enabled for a large dataset for a while, it can become very hard to go back to non-dedup land.

nlx-ck
  • 415