Re: pam problems

bsudbury_at_email.com
Date: 05/18/05


Date: 18 May 2005 00:58:03 -0700

Hi There,

I have run strace on the perl script while the problem is not happening
and again when it is happening. The first sign of problems occurs when
the following is output in the trace log:

9016 open("/var/db/__db.005", O_RDWR|O_CREAT|O_LARGEFILE, 0660) = 3
9016 fcntl64(3, F_SETFD, FD_CLOEXEC) = 0
9016 mmap2(NULL, 8192, PROT_READ|PROT_WRITE, MAP_SHARED, 3, 0) =
0x40317000
9016 close(3) = 0
9016 getpid() = 9016
9016 getpid() = 9016
9016 getpid() = 9016
9016 select(0, NULL, NULL, NULL, {0, 1000}) = 0 (Timeout)
9016 select(0, NULL, NULL, NULL, {0, 2000}) = 0 (Timeout)
9016 select(0, NULL, NULL, NULL, {0, 4000}) = 0 (Timeout)
9016 select(0, NULL, NULL, NULL, {0, 8000}) = 0 (Timeout)
9016 select(0, NULL, NULL, NULL, {0, 16000}) = 0 (Timeout)
9016 select(0, NULL, NULL, NULL, {0, 32000}) = 0 (Timeout)
9016 select(0, NULL, NULL, NULL, {0, 64000}) = 0 (Timeout)
9016 select(0, NULL, NULL, NULL, {0, 128000}) = 0 (Timeout)
9016 select(0, NULL, NULL, NULL, {0, 256000}) = 0 (Timeout)
9016 select(0, NULL, NULL, NULL, {0, 512000}) = 0 (Timeout)
9016 select(0, NULL, NULL, NULL, {1, 0}) = 0 (Timeout)
9016 select(0, NULL, NULL, NULL, {1, 0}) = 0 (Timeout)
9016 select(0, NULL, NULL, NULL, {1, 0}) = 0 (Timeout)
9016 select(0, NULL, NULL, NULL, {1, 0}) = 0 (Timeout)

In a normal situation, this would output:

28335 open("/var/db/__db.005", O_RDWR|O_CREAT|O_LARGEFILE, 0660) = 3
28335 fcntl64(3, F_SETFD, FD_CLOEXEC) = 0
28335 mmap2(NULL, 8192, PROT_READ|PROT_WRITE, MAP_SHARED, 3, 0) =
0x40317000
28335 close(3) = 0
28335 getpid() = 28335
28335 getpid() = 28335
28335 getpid() = 28335
28335 getpid() = 28335
28335 getpid() = 28335
28335 getpid() = 28335
28335 getpid() = 28335
28335 getpid() = 28335
28335 getpid() = 28335

After a while it seemed to move on, but then ends up in the same
problem when trying to access /var/db/__db.005. It more or less seems
that it gets stuck in a loop trying to access a number of different
files, but when it comes back to __db.005 it times out.

Could it be that somehow this file is getting locked or has been
corrupted? Does anyone have any ideas on what to do next?

Regards,

Ben.