I understood my mistake on using consumer drives and this is why I bought the RED versions some days ago. I would have done this earlier if I had the money. So to sum up. I have upgraded my btrfs-progs and I have mounted the filesystem with # mount -o degraded /dev/sdi1 /mnt/mountpoint I want to minimize my risk. Now I should do # btrfs device delete /dev/sdc1 ? or # btrfs check --repair --init-csum-tree ? constantine On Sun, Feb 8, 2015 at 11:34 PM, Chris Murphy <lists@xxxxxxxxxxxxxxxxx> wrote: > On Sun, Feb 8, 2015 at 4:09 PM, constantine <costas.magnuse@xxxxxxxxx> wrote: >> By the way, /dev/sdc just completed the extended offline test without >> any error... I feel so confused,.... > > First, we know from a number of studies, including the famous (and now > kinda old) Google study that a huge percent of drive failures come > with no SMART errors. > > Second, SMART is only saying its internal test is good. The errors are > related to data transfer, so that implicates the enclosure (bridge > chipset or electronics), the cable, or the controller interface. > Actually it could also be a flaky controller or RAM on the drive > itself too which I don't think get checked with SMART tests. > > -- > Chris Murphy -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
