[ale] debug

[Jim Lynch]

Chris Fowler wrote:
> On Fri, 2010-02-26 at 09:43 -0500, Jim Lynch wrote:
>
>   
>
> strace -f program
>
> The let it run for a while.
>   
More info.

I tried the strace, after a while it died again.  This time it mentioned 
FUTEX a lot.  When I went to try it again, it hung almost immediately on 
a FUTEX call.  I googled FUTEX and found some notes that suggest using 
Linux threads instead of whatever that FUTEX is related to, so I put:
export LD_ASSUME_KERNEL=2.4.1

I then ran it again, but this time it died some time later with I think 
a signal 33.  What ever that is.

The first dozen or so lines are normal.  The first abnormal line appears 
to be the rt_sigsuspend ... line.  Kill -l doesn't list a signal 33 on 
Centos.

[pid 23607] gettimeofday({1267248781, 613478}, NULL) = 0
[pid 23607] stat64("/etc/localtime", {st_mode=S_IFREG|0644, 
st_size=2194, ...}) = 0
[pid 23607] stat64("/etc/localtime", {st_mode=S_IFREG|0644, 
st_size=2194, ...}) = 0
[pid 23607] write(3, "2010-02-27 08:33:01.613 DEBUG 65"..., 90) = 90
[pid 23607] _llseek(3, 0, [11857155], SEEK_CUR) = 0
[pid 23607] time(NULL)                  = 1267248781
[pid 23607] gettimeofday({1267248781, 614440}, NULL) = 0
[pid 23607] stat64("/etc/localtime", {st_mode=S_IFREG|0644, 
st_size=2194, ...}) = 0
[pid 23607] stat64("/etc/localtime", {st_mode=S_IFREG|0644, 
st_size=2194, ...}) = 0
[pid 23607] write(3, "2010-02-27 08:33:01.614 DEBUG 65"..., 66) = 66
[pid 23607] _llseek(3, 0, [11857221], SEEK_CUR) = 0
[pid 23607] nanosleep({5, 0},  <unfinished ...>
[pid 23600] <... poll resumed> [{fd=6, events=POLLIN}], 1, 2000) = 0
[pid 23600] getppid()                   = 23592
[pid 23600] poll( <unfinished ...>
[pid 23609] <... rt_sigsuspend resumed> ) = ? ERESTARTNOHAND (To be 
restarted)
[pid 23608] <... nanosleep resumed> 0)  = ? ERESTART_RESTARTBLOCK (To be 
restarted)
[pid 23607] <... nanosleep resumed> 0)  = ? ERESTART_RESTARTBLOCK (To be 
restarted)
[pid 23606] <... rt_sigsuspend resumed> ) = ? ERESTARTNOHAND (To be 
restarted)
[pid 23605] <... nanosleep resumed> 0)  = ? ERESTART_RESTARTBLOCK (To be 
restarted)
[pid 23601] <... rt_sigsuspend resumed> ) = ? ERESTARTNOHAND (To be 
restarted)
[pid 23592] <... rt_sigsuspend resumed> ) = ? ERESTARTNOHAND (To be 
restarted)
[pid 23608] --- SIGTERM (Terminated) @ 0 (0) ---
Process 23608 detached
[pid 23606] --- SIGTERM (Terminated) @ 0 (0) ---
Process 23606 detached
[pid 23605] --- SIGTERM (Terminated) @ 0 (0) ---
Process 23605 detached
[pid 23600] <... poll resumed> [{fd=6, events=POLLIN}], 1, 2000) = -1 
EINTR (Interrupted system call)
[pid 23600] --- SIGRT_1 (Unknown signal 33) @ 0 (0) ---
[pid 23600] sigreturn()                 = ? (mask now ~[TRAP KILL STOP])
[pid 23600] --- SIGRT_1 (Unknown signal 33) @ 0 (0) ---
[pid 23600] sigreturn()                 = ? (mask now ~[TRAP KILL STOP])
[pid 23600] --- SIGRT_1 (Unknown signal 33) @ 0 (0) ---
[pid 23600] sigreturn()                 = ? (mask now ~[TRAP KILL STOP])
[pid 23600] getppid()                   = 23592
[pid 23600] waitpid(-1, [{WIFSIGNALED(s) && WTERMSIG(s) == SIGTERM}], 
WNOHANG|__WCLONE) = 23605
[pid 23600] kill(23609, SIGTERM)        = 0
[pid 23600] kill(23608, SIGTERM)        = 0
[pid 23600] kill(23607, SIGTERM <unfinished ...>
[pid 23609] --- SIGTERM (Terminated) @ 0 (0) ---
Process 23609 detached
[pid 23607] --- SIGTERM (Terminated) @ 0 (0) ---
Process 23607 detached
[pid 23601] --- SIGTERM (Terminated) @ 0 (0) ---
Process 23601 detached
[pid 23600] <... kill resumed> )        = 0
[pid 23600] kill(23606, SIGTERM)        = 0
[pid 23600] kill(23601, SIGTERM)        = 0
[pid 23600] kill(23592, SIGTERM)        = 0
[pid 23600] exit_group(0)               = ?
Process 23600 detached
--- SIGTERM (Terminated) @ 0 (0) ---
+++ killed by SIGTERM +++