johnsanc Posted June 10, 2020 Share Posted June 10, 2020 I recently upgraded my parity drives to 12TB. For now, my largest data drives are 10TB. I noticed that during a parity check, the check progresses past the 10TB mark and processes the full 12TB even through there is no data to check against. Why does this happen? Wouldn't it be more efficient to stop after the last bit of data from the data drives? No support needed, just a general question I was pondering. Quote Link to comment
JonathanM Posted June 10, 2020 Share Posted June 10, 2020 9 minutes ago, johnsanc said: Wouldn't it be more efficient to stop after the last bit of data from the data drives? Depends on your perspective. If, when adding a drive, you had to evaluate if that drive was larger than the current largest data drive, and apply a different set of criteria for adding it, then that logic would be more complex and possibly a point for failure. As it is now, theoretically if the drive you are adding is all binary zeroes, there is already valid parity all the way to the extent of any drive you add. 13 minutes ago, johnsanc said: processes the full 12TB even through there is no data to check against. It's checking that the parity drive has all zeroes past that point. It would suck to add a 12TB data drive only to start getting errors on the parity drive when you were trying to add the new data drive. The secondary function of parity checks is to verify the health of the whole surface of all drives. Quote Link to comment
johnsanc Posted June 10, 2020 Author Share Posted June 10, 2020 Quote It's checking that the parity drive has all zeroes past that point. It would suck to add a 12TB data drive only to start getting errors on the parity drive when you were trying to add the new data drive. Now that makes sense. Thanks! Quote Link to comment
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.