Everything posted by bluescale
-
DiskSpeed, hdd/ssd benchmarking (unRAID 6+), version 2.10.10
Raw Read Error Rate for this drive is indeed very high: 7884. Interestingly, the other drive I purchased at the same time also has a high Raw Read Error Rate, although not nearly as high: 163. This drive also has 1 Current Pending Sector, and a Multizone Error Rate of 1 (I'm not sure what this means). That drives performance is perfectly fine (it's disk 3 in the image above), and that pending sector has been there for a long time. Looking at the SMART report, the Device Error Count is 17. Almost all the errors listed happened at the same time, about 11,000 hours ago. I've attached the SMART report if anyone wants to take a look. Would folks recommend moving this drive out of my array as well? It's been stable, and it looks to me as if something happened a year or so ago, but nothing since. Incidentally, I have 2 4TB drives arriving today. I was planning to use one to replace the poor performing drive I've already removed from my array. I was planning to expand my array with the other drive, but if disk 3 needs to come out of my array in a hurry, I can do that. WDC_WD30EFRX-68EUZN0_WD-WCC4N0763742-20190410-1056.txt
-
DiskSpeed, hdd/ssd benchmarking (unRAID 6+), version 2.10.10
Boy oh boy do I wish I'd run across this utility years ago. I'd always read online that unRaid, for all it's positive attributes, was slow. I took that as gospel, without trying to understand what was normal slow vs. abnormal slow. I realize now that parity checks and transfers at 20 MB/s were not normal slow, but rather a function of a wonky drive. Anyway, I have a WD Red with this curve: It took a bunch of tries to even get this drive to complete, as speedgap kept tripping up. I've removed the drive from my array, and my next parity check ran at 108 MB/s. My question is this - is there anything to be done about this drive, or should I just chuck it? I don't know much about drives, so I'm looking for feedback from experts. The drive itself doesn't show any SMART errors. However, I've been trying to get an extended SMART test to complete for the last 24+ hours, and it's been stuck at 90% for over 12 hours. It's 3.5 years old, so RMA is not an option, unfortunately. I'm pretty sure the drive has been like this since the beginning, considering I've always had slow parity checks. If anyone's curious, here's how my whole my array looks. All drives are WD Reds: