Re: Compute loop + no processing with spamass-milter

[Terry Kennedy]

> This is wort of weird.  There's really nothing in mlfi_eoh.  All it
> does it write a blank line so spamc knows where the headers stop and
> the body starts.  Does BSD/OS have strace/truss/ktrace?  Try tracing
> the milter while it's hung like that and see what syscalls it's
> calling.  A gdb stack trace might be interesting too.

  As I said, it can die in various places. More below.

> Does it process any emails at all, or is there a special set of killer
> messages?  If it doesn't do anything, there might be an interaction
> with your poll emulation library, especially since the milter is
> threaded.  I can convert the poll to select easy enough and send you a
> patch for testing.  Poll is just a nicer interface to use, which is why
> I started with it.

  No email messages get through at all, until the milter times out and
sendmail delivers the unfiltered message. Even a simple "hello world"-
type message wedges it.

  I had the same behavior before you switched to poll from select, and the
libpoll emulator is in use at a variety of sites, so I don't think that is
the problem. But I'm willing to try anything you have...

  In 0.1.2, I have also seen a continuous bunch of:

Aug 18 18:08:44 test2 spamass-milter[2003]: mlfi_eoh: enter
Aug 18 18:08:44 test2 spamass-milter[2003]: mlfi_body: enter
Aug 18 18:08:46 test2 spamass-milter[2003]: poll returned 0
Aug 18 18:08:46 test2 spamass-milter[2003]: polling
Aug 18 18:08:48 test2 spamass-milter[2003]: poll returned 0
Aug 18 18:08:48 test2 spamass-milter[2003]: polling
Aug 18 18:08:49 test2 spamass-milter[2003]: poll returned 0
Aug 18 18:08:49 test2 spamass-milter[2003]: polling
Aug 18 18:08:51 test2 spamass-milter[2003]: poll returned 0
Aug 18 18:08:51 test2 spamass-milter[2003]: polling
Aug 18 18:08:53 test2 spamass-milter[2003]: poll returned 0
[polling repeats forever until killed]

along with the previous behavior. It appears that the code is losing contact
with something and wedging.

  I have a ktrace session with both the parent and child. The kdump is time-
stamped to make it easier to see what happened when, but it is too big to
attach here (most of the space is taken up with its attempt to close fd's 5
through 0x1a0b!!!) so you can view it at: 
http://www.tmk.com/transient/kdump.out

  If you need more info, let me know.

        Thanks,
        Terry Kennedy             http://www.tmk.com
        address@hidden             New York, NY USA