Processes hang on vxfs filesystem waiting for i_mutex in Red Hat Enterprise Linux

Issue

• processes calling stat hang on a vxfs filesystem
• smbd processes hang on inode's i_mutex

• The smbd process can be seen trying to acquire a mutex while doing a path lookup:

  kernel: INFO: task smbd:6277 blocked for more than 120 seconds.
  kernel: "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
  kernel: smbd D ffffffff80153806 0 6277 14932 6804 2847 (NOTLB)
  kernel: ffff8101d5dedc78 0000000000000082 0000000000000074 ffff810483c172c0
  kernel: 0000000000000286 000000000000000a ffff8104825980c0 ffff8104afd81860
  kernel: 00002f9e12f3b648 00000000000031a2 ffff8104825982a8 0000000a00000056
  kernel: Call Trace:
  kernel: [<ffffffff80063c4f>] __mutex_lock_slowpath+0x60/0x9b
  kernel: [<ffffffff80063c99>] .text.lock.mutex+0xf/0x14
  kernel: [<ffffffff8000cfb9>] do_lookup+0x90/0x1e6
  kernel: [<ffffffff8000a2cb>] __link_path_walk+0xa2a/0xfb9
  kernel: [<ffffffff8000ea7a>] link_path_walk+0x42/0xb2
  kernel: [<ffffffff8000cda9>] do_path_lookup+0x275/0x2f1
  kernel: [<ffffffff800128ba>] getname+0x15b/0x1c2
  kernel: [<ffffffff800239d2>] __user_walk_fd+0x37/0x4c
  kernel: [<ffffffff800288cc>] vfs_stat_fd+0x1b/0x4a
  kernel: [<ffffffff8006723e>] do_page_fault+0x4fe/0x874
  kernel: [<ffffffff8003a0a3>] fcntl_setlk+0x243/0x273
  kernel: [<ffffffff80023704>] sys_newstat+0x19/0x31
  kernel: [<ffffffff8005dde9>] error_exit+0x0/0x84
  kernel: [<ffffffff8005d116>] system_call+0x7e/0x83