I don’t have much luck with hard disks these days. One of the new hard disks in my server started reporting a lot of write errors after only a few days of service. I swapped its place with the other new hard disk to make sure it’s not a problem with either cabling or the motherboard but the problem remained. 🙁
Luckily I ordered the two new hard disks via Amazon who offer an excellent return service. They already sent me a replacement hard disk before I even shipped the broken one. Let’s hope this one works better.
Due to massive problems I had to replace two of the hard disks in my server in a hurry recently. As I had positive experience with Western Digital RAID Edition drives in the past I bought two RE4‑GP WD2002FYPS drives. Silke and I installed them earlier last week and our server is in a satisfactory state again.
A few afterthoughts on the whole affair:
- It is very disappointing that one of my server grade Seagate Barracuda ES.2 hard disks died within little over two years.
- The good news is that Seagate offer a five year warrantee on this model. I am therefore entitled to a replacement which I have requested via their website.
- Before you return a faulty hard disk to Seagate they however ask you to run SeaTools to get an error code. The quality of this tool is really impressive. You can download a bootable ISO image with a DOS version that doesn’t seem to work. I tried it to boot on both my HP Proliant ML110 G4 and on an old Pentium 4 PC but it hung after loading the main program. As a result I had to use the Windows version. This version of SeaTools has a charming Windows for Workgroups style user interface. But to be fair it got the job done in the end and confirmed that one my Barracuda hard disks was indeed broken.
- Seagate’s warranty serviced turned out to be very good. I shipped the broken hard disk, received the receipt acknowledgement on Tuesday and the new hard disk on Friday. The replacement hard disk is only a year old and looks brand new.
- Considering that only one of the disks was really broken I’m still a bit puzzled why the second hard disks initially showed problems as well. In a system using SATA with its point-to-point connections you would hope that disks don’t interfere with each other. But somehow they seem to do nevertheless which is not what you would hope for.
Anyway, I’m decided to give the two Seagate hard disks a try and installed them as the second RAID 1 in my server.
My home server is using RAID 1 to keep all the files stored on it safe. The idea is that if one of the hard disks fails I can just replace it without losing any data. Well, that is the theory.
Last night both of the hard disks in the primary RAID 1 volume of my server started to report bad blocks. As a result one of the hard disks got removed from the volume. Two attempts to add it back have failed because parts of the other hard disk cannot be read as well. My attempt to create a backup of the file system containing the bad block revealed more bad blocks and locked up the remaining disk. I will have to restore the data from last Saturday’s full backup. 🙁
I just hope the partially working hard disk will keep running until the two replacements arrive.