On 11/08/2010 07:33 PM, Rick Moen wrote: > Quoting Alex Mandel (tech_...@wildintellect.com): > >> Good Call, I did look a little at finding a drive case that was both >> eSata and usb. The drive case was the cheapest part by far but esata/usb >> isn't so common. I'm not sure if the board in between would still be an >> issue. If I happen to come upon a good deal on such a case I might try >> it. Anyone have an external eSata they could try to get SMART data on? > > All libata drivers support SMART -- which is what one would expect, > given that libata leverages the kernel's SCSI layers. > https://ata.wiki.kernel.org/index.php/Libata_Feature_Table > (The particular SATA interface, internal vs. eSATA, is not an issue.)
While technically true, often eSATA is combined with a multidisk chassis and has a lame/broken chip that multiplexes a single SATA connection to multiple drives. Said lame/broken chip often hides the SMART data. I find is similarly frustrating when the RAID controller does the same thing. It's really really annoying to have to pull a failed drive to get it's model and serial number so you can RMA it. It's also worth mentioning while SMART is cool, I like the idea, and it sounds really useful. The studies that I've seen show SMART is useless for predicting failures. Sure you can get various interesting metrics but there's little relationship between any of the numbers it gives you and losing your entire disk in the near future. The largest of said studies was the Google paper which covered an impressive number of disks across all major brands. _______________________________________________ vox-tech mailing list vox-tech@lists.lugod.org http://lists.lugod.org/mailman/listinfo/vox-tech