I get many checksum errors *during* pvmove operations, such as: [13954.139079] bio too big device dm-4 (72 > 8) [13954.139091] bio too big device dm-4 (16 > 8) [13954.139104] bio too big device dm-4 (56 > 8) [14040.223812] bio too big device dm-4 (64 > 8) [14040.223818] btrfs_dev_stat_print_on_error: 9 callbacks suppressed ... [14175.389937] btrfs: bdev /dev/mapper/btrfs2 errs: wr 0, rd 1196, flush 0, corrupt 0, gen 0 [14175.389956] btrfs csum failed ino 508725 extent 3219001344 csum 1636590654 wanted 2132898075 mirror 0 [14206.788000] bio too big device dm-4 (64 > 8) [14206.788011] btrfs: bdev /dev/mapper/btrfs2 errs: wr 0, rd 1197, flush 0, corrupt 0, gen 0 [14206.788035] btrfs csum failed ino 508725 extent 3219001344 csum 3779168556 wanted 2132898075 mirror 0 [14227.853863] bio too big device dm-4 (64 > 8) [14227.853870] btrfs: bdev /dev/mapper/btrfs2 errs: wr 0, rd 1198, flush 0, corrupt 0, gen 0 [14227.853884] btrfs csum failed ino 508725 extent 3219001344 csum 895061498 wanted 2132898075 mirror 0 ... My setup is: sda3 \ sdb3 \__ md0 RAID5 __ LVM2 PV __ Cryptsetup LV __ BTRFS sdc3 / sdd3 / I pv-moved from md0 to md0 (alloc=anywhere to make room for pvresize) pvmove -v --alloc=anywhere /dev/md1:95793-99632 /dev/md1:70107-74032 During pvmove some program crash with SIGBUS. However, after pvmove completes, everything comes back to normal as if nothing has ever happened. I restart crashed program and they work just fine, now errors. A scrub from initramfs completes without any error, I don't think this has a relation to my other issue (Kernel panic on scrub) as my root is directly on /dev/sd[abcd]4 and both do kernel panic when a scrub is started when the system is booted. -- 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
