Hi. "rev" seems to have some issues in Solaris. Whenever I attach "Rev::TCPServer" to an event loop and make connection, I get the following error. /usr/local/lib/ruby/gems/1.9.0/gems/rev-0.2.2/lib/rev/listener.rb: 35:in `accept_nonblock'': Resource temporarily unavailable - accept(2) (Errno::EAGAIN) from /usr/local/lib/ruby/gems/1.9.0/gems/rev-0.2.2/lib/rev/ listener.rb:35:in `on_readable'' from /usr/local/lib/ruby/gems/1.9.0/gems/rev-0.2.2/lib/rev/ loop.rb:98:in `run_once'' from /usr/local/lib/ruby/gems/1.9.0/gems/rev-0.2.2/lib/rev/ loop.rb:98:in `run'' from test.rb:67:in `<main>'' I am running OpenSolaris snv_93 and using gcc and ruby 1.9.0. Installation(via gem) went without a problem though. daesan
I take it that normal TCPSocket connections work? On Sat, Aug 9, 2008 at 7:40 AM, Dae San Hwang <lists at daesan.com> wrote:> Hi. > > "rev" seems to have some issues in Solaris. Whenever I attach > "Rev::TCPServer" to an event loop and make connection, I get the following > error. > > /usr/local/lib/ruby/gems/1.9.0/gems/rev-0.2.2/lib/rev/listener.rb:35:in > `accept_nonblock'': Resource temporarily unavailable - accept(2) > (Errno::EAGAIN) > from > /usr/local/lib/ruby/gems/1.9.0/gems/rev-0.2.2/lib/rev/listener.rb:35:in > `on_readable'' > from > /usr/local/lib/ruby/gems/1.9.0/gems/rev-0.2.2/lib/rev/loop.rb:98:in > `run_once'' > from > /usr/local/lib/ruby/gems/1.9.0/gems/rev-0.2.2/lib/rev/loop.rb:98:in `run'' > from test.rb:67:in `<main>'' > > I am running OpenSolaris snv_93 and using gcc and ruby 1.9.0. > Installation(via gem) went without a problem though. > > daesan > _______________________________________________ > Rev-talk mailing list > Rev-talk at rubyforge.org > http://rubyforge.org/mailman/listinfo/rev-talk >
On Sat, Aug 9, 2008 at 7:40 AM, Dae San Hwang <lists at daesan.com> wrote:> Hi. > > "rev" seems to have some issues in Solaris. Whenever I attach > "Rev::TCPServer" to an event loop and make connection, I get the following > error. > > /usr/local/lib/ruby/gems/1.9.0/gems/rev-0.2.2/lib/rev/listener.rb:35:in > `accept_nonblock'': Resource temporarily unavailable - accept(2) > (Errno::EAGAIN) >Interesting... it appears Solaris is flagging the listener socket as readable even when it isn''t. Not having access to a Solaris system anymore this is a somewhat hard problem for me to debug. However, the current HEAD revision of the Subversion repository is now configured to swallow Errno::EAGAIN when the Socket calls #accept_nonblock, but emits a warning. Can you give this a try and see if it works around the problem? If you see a bunch of: warning: listener socket spuriously readable ...messages being emitted in a loop then the problem can''t be solved simply by swallowing Errno::EAGAIN and I''ll have to look for a better solution. -- Tony Arcieri medioh.com -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://rubyforge.org/pipermail/rev-talk/attachments/20080811/34321b10/attachment.html>