[Queue-developers] other problems with GNU Queue 1.30.1

[Caroline B. Y. <cb...@te...>]

I've setup two RH6.2 machines and running queue seems to work fine locally
on each machine, but when I attempt to send a job from one machine to the
other, it hangs. This is my 3rd attempt at configuring gnu queue, and I've
combed through the doco without an idea for what's gone wrong.

From the machine named "cluster3", this command works:
  queue -i -w -p -- emacs -nw

And this command doesn't work (it hangs):
  queue -i -w -p -h cluster4 -- emacs -nw

Where "cluster4" is a machine cluster3 can reach using other protocols
(ping, ssh, nfs, telnet, rlogin, etc).  

an strace of the hanging command, dies at:
...
getrlimit(0x6 /* RLIMIT_??? */, {rlim_cur=2*1024, rlim_max=2*1024}) = 0
getrlimit(RLIMIT_NOFILE, {rlim_cur=1024, rlim_max=1024}) = 0
write(4, ":*.png=01;35:*.tif=01;35:\0_=/usr"..., 117) = 117
close(4)                                = 0
munmap(0x40015000, 4096)                = 0
listen(3, 10)                           = 0
accept(3, 

And it hangs indefinitely.

I am guessing that queue_manager is not functioning or not functioning
properly. I think queue_manager is needed in this cross-host
communication, but there doesn't seem to be any doco stating how/if to
run it.

Any suggestions would be greatly appreciated. 

-Caroline

ps running queued -D won't help me because I can't access my
mail (sexchange eats it, and no one here can fix it).