Josh Zerlan writes:
> At 22:52 1/24/2002, you wrote:
>> Josh Zerlan writes:
>>> select(1, [0], NULL, NULL, {600, 0}) = 1 (in [0], left {566, 480000})
>>> read(0, "RCPT TO:<[EMAIL PROTECTED]>\r\n", 5120) = 30
>>> write(5, "[EMAIL PROTECTED]\t\t\n", 21) = -1 EPIPE (Broken pipe)
>>> --- SIGPIPE (Broken pipe) ---
>>> write(1, "432 Service temporarily unavaila"..., 38) = 38
>>> rt_sigprocmask(SIG_BLOCK, [CHLD], [], 8) = 0
>>> rt_sigaction(SIGCHLD, NULL, {SIG_DFL}, 8) = 0
>>> rt_sigprocmask(SIG_SETMASK, [], NULL, 8) = 0
>>> nanosleep({8, 0}, {8, 0}) = 0
>>> time([1011920440]) = 1011920440
>>> time([1011920440]) = 1011920440
>>> select(1, [0], NULL, NULL, {600, 0}
>>> -=-=-=-=-=-=-=-=-=-=-=-=-=-
>>> As you can see here, it immediately dies with a broken pipe... I really
>>> have no idea what it's trying to do here. Is it trying to call a
>>> function that doesn't work/exist? What's going on?
>>
>> It's the pipe to the submit child process, which terminated for some
>> reason. You need to use the -f option to strace to also trace the child
>> processes, and use that to determine how submit terminated.
>
> That strace _IS_ with the -f option, that's why I'm confused.
No it's not. If you did use the -f option, each line of the trace would be
prefixed with the process id.
--
Sam
_______________________________________________
courier-users mailing list
[EMAIL PROTECTED]
Unsubscribe: https://lists.sourceforge.net/lists/listinfo/courier-users
