Yes, but usually a system has several layer of error-detecting/recovering stuff in different granularity. Disk CRC works on Sector level, Ceph CRC mostly work on object level, and we also have replication/erasure coding in system level. The CRC in ceph mainly handle the case, imaging you have an object there, but part of the object has been mistakenly rewritten by some other process, in this case, since disk works well, disk CRC cannot provide any help. From: ceph-users [mailto:ceph-users-bounces@xxxxxxxxxxxxxx] On Behalf Of ??? Sent: Thursday, September 4, 2014 4:31 PM To: ceph-users at ceph.com Subject: ceph data consistency hi, everyone: when I read the filestore.cc, I find the ceph use crc the check the data. Why should check the data? In my knowledge, the disk has error-correcting code<http://en.wikipedia.org/wiki/Error-correcting_code> (ECC) for each sector. Looking at wiki: http://en.wikipedia.org/wiki/Disk_sector, "In disk drives, each physical sector is made up of three basic parts, the sectorheader<http://en.wikipedia.org/wiki/Header_(computing)>, the data area and the error-correcting code<http://en.wikipedia.org/wiki/Error-correcting_code> (ECC)". So if the data is not correct. the disk can recovery it or return i/o error. Does anyone can explain it? Thanks -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.ceph.com/pipermail/ceph-users-ceph.com/attachments/20140910/03b4bbfb/attachment.htm>
 |