Same port worked for just 3 servers before rehanging. New port worked like a fresh instance.
Yes. That mirrors the behaviour I'm seeing. If you retry the original port some time later--I haven't tied down how much later yet--it "recovers", which suggests this is a timeout issue, but I am at a loss to see where else to apply timeouts?
I;ve no reports of the same behaviour under *nix (yet), so this may be a windows-only problem. I guess I'll have to write the equivalent code in C to isolate whether it is windows itself, or Perl on windows. Which is a pain in the butt.
In reply to Re^4: Socket hang. (Windows or Perl? Solutions?)
by BrowserUk
in thread Socket hang. (Windows or Perl? Solutions?) (Updated)
by BrowserUk
| For: | Use: | ||
| & | & | ||
| < | < | ||
| > | > | ||
| [ | [ | ||
| ] | ] |