I keep coming across this tired old ghost story, and so even though there is another, better, more in-depth look at the problem out there, I just wanted to reiterate for anyone who happens to Google it:

“There’s nothing special about ZFS that requires/encourages the use of ECC RAM more so than any other filesystem.”

SO STOP TALKING ABOUT IT.  The idea never made any sense at all, and in fact, (Open)ZFS will do a far better job protecting your data from RAM-based corruption than any other file system out there.

You do not need ECC RAM for your ZFS system any more than you need it for any other system.  Sure, it’s nice to have, but I’ve been using non-ECC RAM my whole life and I’ve never had a significant data corruption problem caused by faulty RAM.  Faulty RAM usually kills your system pretty quick; it hardly offers a chance to do a lot of destructive data work.  And with ZFS, the extra integrity checks will only go further to practically ensure your faulty RAM can’t harm your data.

And of course, you can very simply protect yourself with a solid backup solution.  OpenZFS makes even awesome, automated, off-site backup solutions incredibly easy when paired with OpenSSH.  You can scan your data sets for differences since their last snapshots, snapshot those data sets which differ, and then send the incremental change data over SSH to a remote system which will store it in a separate zpool, disconnected from your production system.

If you don’t want to be that fancy, you can just get a nice big external disk and make it a single vdev zpool.  You can even use OpenZFS’ built in support for various compression algorithms or deduplication to make the absolute most out of a smaller disk.  Sure, it won’t be able to self-heal, but because every block is stored with a hash, you’ll be able to scrub the disk and be alerted if any bad blocks are found.  You can then simply copy from your production system to recover the bad blocks.

Seriously, OpenZFS is amazing, and a lack of ECC RAM is absolutely no reason not to use it.  If anything, as I (and others) write above, it will help mitigate damage from bad RAM, not exacerbate it.  A small system hosting an off-site backup which is automatically updated nightly is not only possible with OpenZFS, but far less expensive and far more important and protective for your data than ECC RAM.

Prioritize and be reasonable, everybody.

This entry was posted in Information Technology and tagged , . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s