Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[benn600]

Another major event happened with my long-discussed 7TB server.  Yesterday (July 22) I noticed a drive fail in the morning.  Remember I'm almost 2,000 miles away from home right now.  Funny how just days before I was thinking how fortunate I've been to not have any problems while away.  So I'm sitting at work normally when I get a second email...uhhoo.  A second drive failed!  6 hours after the first!  The same day!  Now it's serious.

Long story short, I called home and had the family work together to buy two new drives, carefully determine which drives need replaced, pull, and replace the drives.  My 15 year old sister performed the hard drive swap.  Worse yet: my sister was mowing so the drives had to be purchased and then she had to be picked up.  The server is at our old house so everyone had to then drive to the old house.

All in all the replacement went flawless.  It immediately started rebuilding drive #15 and then this morning I see it's already past 20% on the #10th drive.  So it's approaching full recovery.

I was absolutely expecting a third to fail when I noticed two fail so closely together.

[benn600]

Now it's getting ridiculous.  Minutes ago another drive failed (got an email).  Thankfully I ordered more drives (better price) a few days ago when the other two failed.  Unfortunately they won't be here for a while.  I thought I ordered 2 but somehow placed two orders (4 drives total).  At this failure rate I better order a dozen.  Strange though because I hadn't had a single problem for several months.  I suppose I'm making a mistake in that I would often pull and swap "failed" drives back in.  In fact, I'll probably end up doing that this time.  It has always rebuilt fine.  That will at least delay the problem until the new drives arrive.

Now I just hope a second drive doesn't fail or it will be serious again.  If I wouldn't have replaced the other two 2 days ago, theoretically the array would be lost right now.  The up front cost for this server was high but I could live with that.  I'm not liking this continuous ordering of drives.  Any thoughts on what might be causing this trouble?  I honestly feel like the algorithm that determines when a drive fails is a little questionable and fails drives that I can use standalone fine.