Skip to content
  1. Feb 08, 2018
    • Yann Ylavic's avatar
      mod_slotmem_shm: follow up tp r1822509. · 49a8c7bc
      Yann Ylavic authored
      Fishy "unlink semantic" (description) does not apply anymore.
      
      
      
      git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@1823572 13f79535-47bb-0310-9956-ffa450edef68
      49a8c7bc
    • Yann Ylavic's avatar
      mod_proxy_balancer: follow up tp r1822509. · 87cd549e
      Yann Ylavic authored
      Rework server_rec ID so that it doesn't change on restart (or stop/start)
      unless it's Host(s)/IP(s):port(s), ServerName and/or ServerAlias(es) changed.
      
      The goal being to reuse SHMs (and persisted files) names as much as possible,
      with minimal bindings to configuration changes (as far as mod_proxy_balancer
      is concerned). So if the ServerName and first Host/IP:port are unique we use
      that first, otherwise the ServerAlias(es) and other Host(s)/IP(s):port(s) are
      also taken into account, and finally if that's still not enough the server
      index is also used (pathological case handled for correctness with regard to
      the underlying mod_slotmem_shm's reuse code).
      
      
      
      git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@1823564 13f79535-47bb-0310-9956-ffa450edef68
      87cd549e
  2. Feb 07, 2018
  3. Feb 06, 2018
  4. Feb 05, 2018
  5. Feb 03, 2018
    • Yann Ylavic's avatar
      mpm_event: move lingering close "sucker" from the listener to worker(s). · 1bc8f261
      Yann Ylavic authored
      This was the last non-constant time action performed by the listener thread.
      
      It's now handled by the worker thread directly after entering lingering close,
      which should directly address the cases when the socket is already closed
      remotely at that time, hence avoid more scheduling (it may be the common case
      for some scenarios).
      
      And it's only if the above would need blocking (i.e. more data to suck) that
      the socket is added to the pollset for the listener to re-schedule a worker
      later when ready. If no worker is available at that time then the socket is
      forcibly closed (similarly to what's done for keepalive connections in this
      case).
      
      Also, since process_lingering_close() is now called by a worker thread and
      with almost no depth in the call stack, we can grow the size of the "suck"
      buffer from 2K to 32K to potentially call recv() up to sixteen times less.
      
      
      
      git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@1823047 13f79535-47bb-0310-9956-ffa450edef68
      1bc8f261
  6. Feb 02, 2018
  7. Feb 01, 2018
  8. Jan 31, 2018
  9. Jan 30, 2018
  10. Jan 29, 2018
  11. Jan 27, 2018
  12. Jan 26, 2018
  13. Jan 24, 2018
  14. Jan 21, 2018