mdadm proccess 100% CPU
I have a centos server for some months, everything was working fine. Yesterday I installed docker-io on it to run an image.
From there I see via top that CPU is always close to 100% with a process called mdadm
Inspecting the process I that some directories (on /dev and /sys) are trying to be accessed / changed but are on read-only mode.
What caused this and how should I proceed? If I reboot the server it remains normal for few minutes, but soon the process starts.
stat("/dev/md1", {st_mode=S_IFBLK|0660, st_rdev=makedev(9, 1), ...}) = 0
open("/dev/md1", O_RDONLY) = 3
ioctl(3, GET_ARRAY_INFO, 0x7ffe236faee0) = 0
close(3) = 0
lseek(4, 0, SEEK_SET) = 0
dup(4) = 3
fcntl(3, F_GETFL) = 0x8000 (flags O_RDONLY|O_LARGEFILE)
fstat(3, {st_mode=S_IFREG|0444, st_size=0, ...}) = 0
mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x7f7ce27fc000
lseek(3, 0, SEEK_CUR) = 0
fcntl(3, F_SETFD, FD_CLOEXEC) = 0
read(3, "Personalities : [raid1] \nmd2 : a"..., 1024) = 331
read(3, "", 1024) = 0
read(3, "", 1024) = 0
close(3) = 0
munmap(0x7f7ce27fc000, 4096) = 0
open("/sys/block", O_RDONLY|O_DIRECTORY) = 3
close(3) = 0
open("/sys/block/md1/md/array_state", O_RDWR) = -1 EROFS (Read-only file system)
open("/sys/block/md1/md/level", O_RDWR) = -1 EROFS (Read-only file system)
open("/sys/block", O_RDONLY|O_DIRECTORY) = 3
close(3) = 0
open("/sys/block/md0/md/array_state", O_RDWR) = -1 EROFS (Read-only file system)
open("/sys/block/md0/md/level", O_RDWR) = -1 EROFS (Read-only file system)
open("/sys/block", O_RDONLY|O_DIRECTORY) = 3
close(3) = 0
open("/sys/block/md2/md/array_state", O_RDWR) = -1 EROFS (Read-only file system)
open("/sys/block/md2/md/level", O_RDWR) = -1 EROFS (Read-only file system)
open("/sys/block", O_RDONLY|O_DIRECTORY) = 3
close(3) = 0
open("/sys/block/md1/md/array_state", O_RDWR) = -1 EROFS (Read-only file system)
open("/sys/block/md1/md/level", O_RDWR) = -1 EROFS (Read-only file system)
open("/sys/block", O_RDONLY|O_DIRECTORY) = 3
close(3) = 0