-----Original Message----- From: Ric Wheeler [mailto:ricwheeler@xxxxxxxxx] Sent: Thursday, May 22, 2008 9:43 AM To: David Lethe Cc: Keith Roberts; linux-raid@xxxxxxxxxxxxxxx Subject: Re: Two Drive Failure on RAID-5 David Lethe wrote: > the disk manufacturers stopped making them last year, and stopped R&D on them way before that. > > -----Original Message----- > > From: "Keith Roberts" <keith@xxxxxxxxxxxx> > Subj: RE: Re: Re: Two Drive Failure on RAID-5 > Date: Tue May 20, 2008 5:20 pm > Size: 1K > To: "linux-raid@xxxxxxxxxxxxxxx" <linux-raid@xxxxxxxxxxxxxxx> > > On Tue, 20 May 2008, David Lethe wrote: > > >> To: Cry <cry_regarder@xxxxxxxxx>, linux-raid@xxxxxxxxxxxxxxx >> From: David Lethe <david@xxxxxxxxxxxx> >> Subject: RE: Re: Re: Two Drive Failure on RAID-5 >> >> Here is a good analogy that puts this in perspective. I haven't seen >> anybody equate the two yet, so get the name right if you quote this ;) >> >> Disk drives are like light bulbs. You can buy the server class (similar >> to CFLs), or desktop (incandescent). If you don't mind the dark, >> replace them as they fail, and buy spares as they go on sale. >> Conversely, if you have to maintain a vaulted ceiling chandelier, and >> are afraid of heights, then spending twice as much for never having to >> deal with *THAT* again will seem like a bargain. >> >> - David Lethe >> > > So are there such things as server class EIDE drives? Or are > they all SCSI or SATA? > > Keith > Different vendors have different strategies around how to market their various drives - you have archival (think big, slow S-ATA for things like a Tivo), S-ATA drives that are consumer grade or slightly higher class drive like SAS (serial attached SCSI) and then the highest quality drives (Fibre channel). You can get good results from all classes of drives, but you need to make sure that you periodically check them pro-actively for errors and try to repair them in place if possible. Also, make sure you get an updated kernel so we don't kick out drives that would be otherwise perfectly reasonable ;-) ric ============ I must add a caveat to what Ric said. Disks are designed to withstand specific duty cycles as well as tolerance for errors and intelligence when it comes to automated error recovery. His statement is only valid up to the point your usage doesn't exceed the workload that a specific disk was designed to tolerate. Furthermore, the superior server class drives have significantly better error recovery & management. As such, they are less prone to having errors for any given load. Read the specs & programming manuals for specifics. So, for any given load, you can purchase disk drives that will actually generate FEWER errors .. which is a heck of a lot better than having to pro-actively test & repair and deal with drives that kernels kick out. That's why the disk manufacturers have 146 GB disks that retail for $150 to $1000 dollars. The difference is not as many have claimed, just the physical interface. David -- To unsubscribe from this list: send the line "unsubscribe linux-raid" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
