Lately, I have been getting Buffer I/O errors during boot time. Seemingly, it appears randomly and I could not be able to reproduce it. However, it happens in every 5-8 boot on average. Here is the relevant part from dmesg:
[ 24.630158] blk_update_request: I/O error, dev loop5, sector 0
[ 24.630199] blk_update_request: I/O error, dev loop5, sector 0
[ 24.630229] Buffer I/O error on dev loop5, logical block 0, async page read
[ 24.630272] blk_update_request: I/O error, dev loop5, sector 0
[ 24.630301] Buffer I/O error on dev loop5, logical block 0, async page read
[ 24.630340] blk_update_request: I/O error, dev loop5, sector 0
[ 24.630369] Buffer I/O error on dev loop5, logical block 0, async page read
[ 24.630411] blk_update_request: I/O error, dev loop5, sector 0
[ 24.630454] Buffer I/O error on dev loop5, logical block 0, async page read
[ 24.630494] blk_update_request: I/O error, dev loop5, sector 0
[ 24.630522] Buffer I/O error on dev loop5, logical block 0, async page read
[ 24.630566] blk_update_request: I/O error, dev loop5, sector 0
[ 24.630593] Buffer I/O error on dev loop5, logical block 0, async page read
[ 24.630634] blk_update_request: I/O error, dev loop5, sector 0
[ 24.630661] Buffer I/O error on dev loop5, logical block 0, async page read
[ 24.630698] blk_update_request: I/O error, dev loop5, sector 0
[ 24.630726] Buffer I/O error on dev loop5, logical block 0, async page read
[ 24.630763] blk_update_request: I/O error, dev loop5, sector 0
[ 24.630791] Buffer I/O error on dev loop5, logical block 0, async page read
[ 24.630832] Buffer I/O error on dev loop5, logical block 0, async page read
This loop device is generated by snap, here is the the lsblk output:
$ lsblk
NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT
loop0 7:0 0 17.3M 1 loop /snap/node/1767
loop1 7:1 0 91M 1 loop /snap/core/6350
loop2 7:2 0 91.1M 1 loop /snap/core/6259
loop3 7:3 0 91M 1 loop /snap/core/6405
loop4 7:4 0 17.3M 1 loop /snap/node/1729
loop5 7:5 0 17.3M 1 loop /snap/node/1793
sda 8:0 0 238.5G 0 disk
├─sda1 8:1 0 524M 0 part /boot/efi
├─sda2 8:2 0 954M 0 part /boot
└─sda3 8:3 0 237G 0 part
└─sda3_crypt 254:0 0 237G 0 crypt
├─xxx--vg-root--lv 254:1 0 229.1G 0 lvm /
└─xxx--vg-swap--lv 254:2 0 7.9G 0 lvm [SWAP]
The OS is Debian 9 with kernel version 4.9.0-8-amd64 and snapd version 2.37.2.
Edit: Per @frostschutz's comment, below is df -h output. So what I understand is that the total size of the loop devices created by snap is equal to the used size. And there is much empty space in /dev/mapper/xxx--vg-root--lv (where the / is), so I don't think it is the problem.
$ df -h
Filesystem Size Used Avail Use% Mounted on
udev 3.9G 0 3.9G 0% /dev
tmpfs 788M 9.5M 778M 2% /run
/dev/mapper/xxx--vg-root--lv 225G 123G 91G 58% /
tmpfs 3.9G 23M 3.9G 1% /dev/shm
tmpfs 5.0M 4.0K 5.0M 1% /run/lock
tmpfs 3.9G 0 3.9G 0% /sys/fs/cgroup
/dev/loop0 18M 18M 0 100% /snap/node/1767
/dev/loop1 91M 91M 0 100% /snap/core/6350
/dev/loop2 92M 92M 0 100% /snap/core/6259
/dev/loop3 91M 91M 0 100% /snap/core/6405
/dev/loop4 18M 18M 0 100% /snap/node/1729
/dev/sda2 939M 70M 823M 8% /boot
/dev/sda1 523M 132K 523M 1% /boot/efi
tmpfs 788M 20K 788M 1% /run/user/118
tmpfs 788M 40K 788M 1% /run/user/1000
/dev/loop5 18M 18M 0 100% /snap/node/1793
df -hfor available free space.