Description of problem: [RHEL-7.5] losetup can not detach all loop devices Version-Release number of selected component (if applicable): util-linux-2.23.2-52.el7.x86_64 How reproducible: often Steps to Reproduce: 1. losetup /dev/loopx -f /tmp/tfile_loopx 2. losetup -d 3. losetup -a Actual results: # losetup --all /dev/loop0: [64768]:1001699 (/tmp/tfile_loop0) /dev/loop2: [64768]:1001699 (/tmp/tfile_loop0) /dev/loop8: [64768]:1001681 (/tmp/tfile_loop1) # losetup -D # losetup --all /dev/loop0: [64768]:1001699 (/tmp/tfile_loop0) /dev/loop2: [64768]:1001699 (/tmp/tfile_loop0) # strace losetup -d /dev/loop2 execve("/usr/sbin/losetup", ["losetup", "-d", "/dev/loop2"], [/* 39 vars */]) = 0 brk(NULL) = 0x1f88000 mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x7f72200dc000 access("/etc/ld.so.preload", R_OK) = -1 ENOENT (No such file or directory) open("/etc/ld.so.cache", O_RDONLY|O_CLOEXEC) = 3 fstat(3, {st_mode=S_IFREG|0644, st_size=44171, ...}) = 0 mmap(NULL, 44171, PROT_READ, MAP_PRIVATE, 3, 0) = 0x7f72200d1000 close(3) = 0 open("/lib64/libc.so.6", O_RDONLY|O_CLOEXEC) = 3 read(3, "\177ELF\2\1\1\3\0\0\0\0\0\0\0\0\3\0>\0\1\0\0\0\340$\2\0\0\0\0\0"..., 832) = 832 fstat(3, {st_mode=S_IFREG|0755, st_size=2151832, ...}) = 0 mmap(NULL, 3981792, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_DENYWRITE, 3, 0) = 0x7f721faef000 mprotect(0x7f721fcb2000, 2093056, PROT_NONE) = 0 mmap(0x7f721feb1000, 24576, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x1c2000) = 0x7f721feb1000 mmap(0x7f721feb7000, 16864, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0x7f721feb7000 close(3) = 0 mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x7f72200d0000 mmap(NULL, 8192, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x7f72200ce000 arch_prctl(ARCH_SET_FS, 0x7f72200ce740) = 0 mprotect(0x7f721feb1000, 16384, PROT_READ) = 0 mprotect(0x611000, 4096, PROT_READ) = 0 mprotect(0x7f72200dd000, 4096, PROT_READ) = 0 munmap(0x7f72200d1000, 44171) = 0 brk(NULL) = 0x1f88000 brk(0x1fa9000) = 0x1fa9000 brk(NULL) = 0x1fa9000 open("/usr/lib/locale/locale-archive", O_RDONLY|O_CLOEXEC) = 3 fstat(3, {st_mode=S_IFREG|0644, st_size=106070960, ...}) = 0 mmap(NULL, 106070960, PROT_READ, MAP_PRIVATE, 3, 0) = 0x7f72195c6000 close(3) = 0 stat("/sys/block", {st_mode=S_IFDIR|0755, st_size=0, ...}) = 0 uname({sysname="Linux", nodename="storageqe-52.rhts.eng.pek2.redhat.com", ...}) = 0 stat("/dev/loop-control", {st_mode=S_IFCHR|0660, st_rdev=makedev(10, 237), ...}) = 0 open("/dev/loop2", O_RDONLY|O_CLOEXEC) = 3 ioctl(3, LOOP_CLR_FD) = 0 close(3) = 0 close(1) = 0 close(2) = 0 exit_group(0) = ? +++ exited with 0 +++ # losetup --all /dev/loop0: [64768]:1001699 (/tmp/tfile_loop0) /dev/loop2: [64768]:1001699 (/tmp/tfile_loop0) Expected results: Additional info:
This is nothing unusual, if there is some reader of the loop device than kernel does not destroy the device, but mark by LO_FLAGS_AUTOCLEAR. The device will be detached later when reference counting will be zero. As you can see from the strace than ioctl(3, LOOP_CLR_FD) = 0 kernel returns success, in this case losetup cannot do anything more. man losetup (upstream): -d, --detach loopdev... Detach the file or device associated with the specified loop device(s). Note that since Linux v3.7 kernel uses "lazy device destruction". The detach operation does not return EBUSY error anymore if device is actively used by system, but it is marked by autoclear flag and destroyed later. Maybe we can add this note to RHEL7.6 too.
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHBA-2018:3178