Re: SMART statistics for 40, 000 disk drives, from Backblaze
Tim Small <[email protected]>
| Newsgroups | gmane.linux.utilities.smartmontools |
|---|---|
| Message-ID | <[email protected]> |
On 30/11/14 20:09, Bruce Allen wrote: > the mds (software raid) driver. Nine years ago, this was near the top of Neil's "to-do list": > > http://neil.brown.name/blog/20050727141521 > > (see item "Don't kick drives on read errors"). My understanding is that was in fact implemented and is now part of the standard mds driver. But I'm not sure, perhaps someone else here can confirm or deny that. Yes, Linux md has done this for many years now. You can also trigger this prospectively using the md driver's array checking functionality. See the 'md' man page, under the "Scrubbing and Mismatches" section. If you want to verify this behaviour, you can do-so with the use of "hdparm --make-bad-sector". The corner-case with this is that if you get a bad block in metadata (e.g. md metadata, or drive partition table) you still lose, but as this consumes such a small proportion of a modern drives's capacity, the chances of this happening (on a drive which is still otherwise reasonably serviceable) are remote. Tim. -- South East Open Source Solutions Limited Registered in England and Wales with company number 06134732. Registered Office: 2 Powell Gardens, Redhill, Surrey, RH1 1TQ VAT number: 900 6633 53 http://seoss.co.uk/ +44-(0)1273-808309 ------------------------------------------------------------------------------ Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server from Actuate! Instantly Supercharge Your Business Reports and Dashboards with Interactivity, Sharing, Native Excel Exports, App Integration & more Get technology previously reserved for billion-dollar corporations, FREE http://pubads.g.doubleclick.net/gampad/clk?id=157005751&iu=/4140/ostg.clktrk