Re: Distributed bulk building for slow machines

[Aleksey Cheusov <cheusov%tut.by@localhost>]

 >> Instead, keep one machine
 >> building a given package until it's finished or not heard from
 >> in a while.

> You'd need something that would report at fixed intervals "still alive
> and still working"
Personally I rely solely on non-blocking select(2)/read(2) and their
return value. If transport program unexpectedly exits (remote hosts
shuts down or reboot), wip/paexec (tool for distributing tasks over
network/CPUs) behaves normally. I haven't seen any problem for years
with ssh.  If transport program hangs up, you need to kill it manually.

-- 
Best regards, Aleksey Cheusov.