Hello Guys, I got one report in which buffered write IO hangs in balance_dirty_pages, after one nvme block device is unplugged physically, then umount can't succeed. Turns out it is one long-term issue, and it can be triggered at least since v5.14 until the latest v6.3. And the issue can be reproduced reliably in KVM guest: 1) run the following script inside guest: mkfs.ext4 -F /dev/nvme0n1 mount /dev/nvme0n1 /mnt dd if=/dev/zero of=/mnt/z.img& sleep 10 echo 1 > /sys/block/nvme0n1/device/device/remove 2) dd hang is observed and /dev/nvme0n1 is gone actually [root@ktest-09 ~]# ps -ax | grep dd 1348 pts/0 D 0:33 dd if=/dev/zero of=/mnt/z.img 1365 pts/0 S+ 0:00 grep --color=auto dd [root@ktest-09 ~]# cat /proc/1348/stack [<0>] balance_dirty_pages+0x649/0x2500 [<0>] balance_dirty_pages_ratelimited_flags+0x4c6/0x5d0 [<0>] generic_perform_write+0x310/0x4c0 [<0>] ext4_buffered_write_iter+0x130/0x2c0 [ext4] [<0>] new_sync_write+0x28e/0x4a0 [<0>] vfs_write+0x62a/0x920 [<0>] ksys_write+0xf9/0x1d0 [<0>] do_syscall_64+0x59/0x90 [<0>] entry_SYSCALL_64_after_hwframe+0x63/0xcd [root@ktest-09 ~]# lsblk | grep nvme [root@ktest-09 ~]# BTW, my VM sets 2G ram, and the nvme disk size is 40GB. So far only observed on ext4 FS, not see it on XFS. I guess it isn't related with disk type, and not tried such test on other type of disks yet, but will do. Seems like dirty pages aren't cleaned after ext4 bio is failed in this situation? Thanks, Ming
