RHEL7 において、別のプログラムは NFS から応答を得られている一方で、nfs_wait_bit_uninterruptible で NFS に書き込んでいるプログラムがハングアップする

Issue

• defunct プロセスが nfs_wait_bit_interruptible で停止しています。
• 他のプロセスは問題なく NFS にアクセスできるにも関わらず、NFS ファイルに書き込むプロセスが停止します。
• NFS でバックアップを取得していると、RHEL 7.2 システムがハングアップします。

Apr 25 22:44:30 localhost kernel:INFO: task kernel:22442 blocked for more than 120 seconds.
Apr 25 22:44:30 localhost kernel:"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
Apr 25 22:44:30 localhost kernel: kernel          D ffff8823c49d7b30     0 22442   5299 0x00000080
Apr 25 22:44:30 localhost kernel: ffff8823c49d79d0 0000000000000086 ffff880a04789700 ffff8823c49d7fd8
Apr 25 22:44:30 localhost kernel: ffff8823c49d7fd8 ffff8823c49d7fd8 ffff880a04789700 ffff88283fc14780
Apr 25 22:44:30 localhost kernel:0000000000000000 7fffffffffffffff ffffffffa0598bc0 ffff8823c49d7b30
Apr 25 22:44:30 localhost kernel:Call Trace:
Apr 25 22:44:30 localhost kernel:[<ffffffffa0598bc0>] ? nfs_pgio_header_free+0x20/0x20 [nfs]
Apr 25 22:44:30 localhost kernel:[<ffffffff8163a879>] schedule+0x29/0x70
Apr 25 22:44:30 localhost kernel:[<ffffffff81638569>] schedule_timeout+0x209/0x2d0
Apr 25 22:44:30 localhost kernel:[<ffffffff8101cd15>] ? native_sched_clock+0x35/0x80
Apr 25 22:44:30 localhost kernel:[<ffffffff8101c829>] ? read_tsc+0x9/0x10
Apr 25 22:44:30 localhost kernel:[<ffffffffa0598bc0>] ? nfs_pgio_header_free+0x20/0x20 [nfs]
Apr 25 22:44:30 localhost kernel:[<ffffffff81639eae>] io_schedule_timeout+0xae/0x130
Apr 25 22:44:30 localhost kernel:[<ffffffff81639f48>] io_schedule+0x18/0x20
Apr 25 22:44:30 localhost kernel:[<ffffffffa0598bce>] nfs_wait_bit_uninterruptible+0xe/0x20 [nfs]
Apr 25 22:44:30 localhost kernel:[<ffffffff816386f0>] __wait_on_bit+0x60/0x90
Apr 25 22:44:30 localhost kernel:[<ffffffffa0598bc0>] ? nfs_pgio_header_free+0x20/0x20 [nfs]
Apr 25 22:44:30 localhost kernel:[<ffffffff816387a7>] out_of_line_wait_on_bit+0x87/0xb0
Apr 25 22:44:30 localhost kernel:[<ffffffff810a6b60>] ? wake_atomic_t_function+0x40/0x40
Apr 25 22:44:30 localhost kernel:[<ffffffffa059a593>] nfs_wait_on_request+0x33/0x40 [nfs]
Apr 25 22:44:30 localhost kernel:[<ffffffffa059f161>] nfs_updatepage+0x121/0x8a0 [nfs]
Apr 25 22:44:30 localhost kernel:[<ffffffffa058efa1>] nfs_write_end+0x121/0x350 [nfs]
Apr 25 22:44:30 localhost kernel:[<ffffffff81169d74>] generic_file_buffered_write+0x184/0x290
Apr 25 22:44:30 localhost kernel:[<ffffffff8116b235>] __generic_file_aio_write+0x1d5/0x3e0
Apr 25 22:44:30 localhost kernel:[<ffffffff8116b49d>] generic_file_aio_write+0x5d/0xc0
Apr 25 22:44:30 localhost kernel:[<ffffffffa058e03b>] nfs_file_write+0xbb/0x1d0 [nfs]
Apr 25 22:44:30 localhost kernel:[<ffffffff811ddcbd>] do_sync_write+0x8d/0xd0
Apr 25 22:44:30 localhost kernel:[<ffffffff811de4dd>] vfs_write+0xbd/0x1e0
Apr 25 22:44:30 localhost kernel:[<ffffffff811def7f>] SyS_write+0x7f/0xe0
Apr 25 22:44:30 localhost kernel:[<ffffffff816458c9>] system_call_fastpath+0x16/0x1