On Wed, Sep 10, 2014 at 10:31 AM, Emmanuel Florac <eflorac@xxxxxxxxxxxxxx> wrote: > Le Tue, 09 Sep 2014 20:43:08 -0500 > Leslie Rhorer <lrhorer@xxxxxxxxxxxx> écrivait: > >> None of the failed drives were WD green. All three and the >> previous four were Seagate. I realize that is not a large >> statistical sample. >> > > If you're interested in large statistical samples, on a grand total of > 4000 1 TB Seagate Barracuda ES2, I had to replace 2100 of them over the > course of 3 years. I still have a couple of hundred of these > unfortunate pieces of crap in service, and they still represent the > vast majority of unexpected RAID malfunctions, urgent replacements, > late night calls and other "interesting side activities". > > I wouldn't buy anything labeled Seagate nowadays. Their drives have > been the baddest train wreck since the dreaded 9 GB Micropolis back in > 1994 (or was it 1995?). I buy about 100 drives a year, but I don't work them very hard. Just lots of data to store and I need to keep my data sets segregated for legal reasons. I don't use raid, just lots of individual disks and most data maintained redundantly. About 4 years ago (or maybe 5), Seagate had a catastrophic drive situation. I can remember buying a batch of 10 drives and having 8 of them fail in the first 2 months. The bad part was they mostly survived a 10 hour burn-in, so they tended to fail with real data on them. I had one case (at a minimum) that summer where I put the data on 3 different Seagate drives and all 3 failed. Fortunately, I was able to swap the disk controller card from one of the working drives with one of the dead drives and recover the data. Regardless, ignoring the summer of discontent, I find Seagate to be my preferred drives. fyi: In June I bought 30 or so WD elements drives to try them out. These are not the green drives, just bare bones WD drives. None of them were DOA, but 3 failed within 4 weeks, so a 10% failure rate in the first month. Only one of them had unique data on it, so I had to recreate that data. Fortunately the source of the data was still available. All of those drives have been pulled out of routine service. Greg _______________________________________________ xfs mailing list xfs@xxxxxxxxxxx http://oss.sgi.com/mailman/listinfo/xfs
