That's what CRC does...why do you think it won't? You could possibly argue that if you have a bad controler that still functions 98% AND a bad hard drive that still gives the illusion of working that this could happen...otherwise this makes no sense. And under your way, with one drive in the box and one drive via usb/firewire...why wouldn't bad data be written then?
On Thu, Sep 4, 2008 at 6:47 PM, Tom Piwowar <[EMAIL PROTECTED]> wrote: > > Actually I don't think this is right. Hardware RAID does crc on the > data, > > so corruption is almost impossible, the RAID would drop the bad drive off > if > > it started returning errors. Of course this is moo as joey says if he > is > > doing soft RAID. So the question is...hard or soft? > > You keep worying about the wrong type of errors. RAID is from an age > where drive MTBFs were 2000 hours. Today MTBFs are at 300,000 hours. > Today the kinds of errors we see the most are "soft" errors. RAID error > checking will not see them, it will pass them on unfixed, but they will > destroy your data just the same. > > > ************************************************************************* > ** List info, subscription management, list rules, archives, privacy ** > ** policy, calmness, a member map, and more at http://www.cguys.org/ ** > ************************************************************************* > ************************************************************************* ** List info, subscription management, list rules, archives, privacy ** ** policy, calmness, a member map, and more at http://www.cguys.org/ ** *************************************************************************
