On 2019/5/21 8:38 下午, Nix wrote: > On 13 May 2019, Coly Li said: >> Since there is no simple way to detect whether the backing divice is a >> md raid device > > Is this really true? I find this very surprising. Lots of mkfs programs > can detect exactly this (to determine the RAID stripe width), and it's > obvious in sysfs as well... > >> this patch simply ignores I/O failures for read-ahead >> bios on backing device, to avoid bogus backing device failure on a >> degrading md raid array. > > ... but this feels like a good idea in any case. Readahead is a > performance optimization: if it fails, it is not a user-visible sign of > device failure in the same way that explicit I/O or metadata I/O is. > It seems three people are agreed on same idea, cool :-) -- Coly Li
