I have had an application which uses backgroundrb with several
workers. Bgrb has been working well for me. Recently I changed one of
the workers to use several threads. Initially I used sleeps w/in the
threads, but noticed that the workers were hanging up for periods of time
that seemed to be related to the sleeps. I have seen comments in other
postings about using next_turn() {} although I am not sure what to put in
the block. I don't have reloadable workers, as mentioned in one posting
(at least I don't think I do) and I can't tell if next_turn() has any
effect. Does anyone have any advice for a sleep-like operation so the
threads will yield control? When they are waiting for something to do,
they are not doing any i/o, just checking a fifo in memory. I am using
packet 1.15 and bgrb 1.03. I didn't upgrade past that because we develop
on Macs and deploy on Linux and it seemed different versions of packet
were needed.
Thanks in advance, alan hayes
_______________________________________________
Backgroundrb-devel mailing list
[email protected]
http://rubyforge.org/mailman/listinfo/backgroundrb-devel
