[OpenAFS-devel] du/find hang running in AFS space

[Touretsky, Gregory]

Hi,

   we get a problem with Open AFS 1.2.2 running on RedHat 6.2 machine (Linux
iwsl008 2.2.17-14smp #1 SMP Thu May 17 13:26:42 PDT 2001 i686 unknown).

Both find and du stuck when run in AFS space. Any ideas why this happens?
There is no such problem with IBM AFS. I don't know if this happens in older
versions of OpenAFS.

I can see that the process doesn't return from system calls 106/107:
***************************************************************************
printing stack for pid 1290
init_task.mm Data :
mm.count = 1438894528
mm.context = 1394111936
mm.start_code =  4aae068 mm.end_code = f9ede808
mm.total_vm = 138643920 mm.locked_vm = 1342418703
mm.swap_cnt = -2063285117
task_struct at 0xca120000 (fcs=0x0, esp0=0x22) for pid 1208:
struct task_struct {
      state = 1
      flags = 0x0
      sigpending = 0
      addr_limit = 0xc0000000
      exec_domain = 0xc0270ea0
      need_resched = 0
      counter = 39
      priority = 20
      avg_slice = 84333
      has_cpu = 0
      processor = 1
      pid = 1208
      pgrp = 1208
      leader = 0
      p_opptr = 0xca1c6000
      p_pptr = 0xca1c6000
      pidhash_next = 0x0
      pidhash_pprev = 0xc02d12f0
      tarray_ptr = 0xc026538c
      wait_chldexit = 0xca12008c
      start_time = 82287
      uid = 0
      euid = 0
      suid = 0
      fsuid = 0
      gid = 0
      egid = 0
      sgid = 0
      fsgid = 0
      tss = struct thread_struct {
          esp0 = 0x22
          esp = 0x835082e
          eip = 0x3e653e65
      task_struct print not yet fully implemented
  }
System Call No: 106
***************************************************************************

Looking at the fstrace log:
***************************************************************************
time 803.500554, pid 1290: Access vp 0xd0a0745c mode 0x40 len 0x1000
time 803.500554, pid 1290: Access vp 0xd0a075d0 mode 0x40 len 0x800
time 803.500554, pid 1290: Access vp 0xd0a0b44c mode 0x40 len 0x800
time 803.500554, pid 1290: Access vp 0xd0a209c8 mode 0x40 len 0x1000
time 803.500554, pid 1290: Access vp 0xd0a20cb0 mode 0x40 len 0x1000
time 803.500554, pid 1290: Access vp 0xd0a20f98 mode 0x40 len 0x800
time 803.500554, pid 1290: Access vp 0xd0a20f98 mode 0x40 len 0x800
time 803.500554, pid 1290: RPC GetVolumeByName for 536921657 ( at
0xc9cedec6)
time 803.510554, pid 1290: Analyze RPC op -1 conn 0xce250b00 code 0x0 user
0x41cd90ae
time 803.510554, pid 1290: Analyze RPC op 2 conn 0xcb64bec0 code 0x0 user
0x41cd90ae 
time 803.510554, pid 1290: ProcessFS vp 0xd0a33238 old len 0x1000 new len
0x1000
time 803.510554, pid 1290: Access vp 0xd0a20f98 mode 0x40 len 0x800
time 803.510554, pid 1290: Access vp 0xd0a33238 mode 0x100 len 0x1000
time 803.510554, pid 1290: Open 0xd0a33238 flags 0x10800
time 803.510554, pid 1290: Open 0xd0a33238 flags 0xf423f
***************************************************************************

From the "strace find" output:
***************************************************************************
fchdir(3)                               = 0 <0.000043>
chdir("/afs/iil.intel.com/usr/cad/iaws_2.02/alpha/netlisters") = 0
<0.000370>
stat("4.34.a1", {st_dev=makedev(0, 6), st_ino=391840246,
st_mode=S_IFDIR|0777, st_nlink=2, st_uid=0, st_gid=0, st_blksize=4096,
st_blocks=8, st_size=4096, st_atime=1996/04/18-11:01:55,
st_mtime=1996/04/18-11:01:55, st_ctime=1996/04/18-11:01:55}) = 0 <0.000054>
open("4.34.a1", O_RDONLY|O_NONBLOCK|O_DIRECTORY) = 4 <0.000066>
fstat(4, {st_dev=makedev(0, 6), st_ino=391840246, st_mode=S_IFDIR|0777,
st_nlink=2, st_uid=0, st_gid=0, st_blksize=4096, st_blocks=8, st_size=4096,
st_atime=1996/04/18-11:01:55, st_mtime=1996/04/18-11:01:55,
st_ctime=1996/04/18-11:01:55}) = 0 <0.000031>
fcntl(4, F_SETFD, FD_CLOEXEC)           = 0 <0.000028>
getdents(4, 
***************************************************************************

Gregory Touretsky
IDC Computing / Systems Engineering Group
Unix Server Platforms
gregory.touretsky@intel.com
> (+) 972-4-865-6377, Fax: 04-865-5999
iNET: 465-6377, M/S: IDC-1B