Unidentified subject!

From: T. Esting <t_esting@dont-contact.us>
Date: Tue, 13 Apr 1999 18:22:43 PDT

  Dear Squid community -

  On an Ultra5/267MHz/256MB machine, we have been able to wheeze by serving
several thousand users (20-40K requests / hr) using squid 1.1.22. However,
spend the majority of our time in disk wait mode. I recently upgraded to
2.1PATCH2 and asynchronous I/O seemed to take care of that bottleneck
nicely. However, while the change improved I/O, it seemed to dramatically
affect CPU usage: our run queue sizes went from 1 < q < 2 to 2 < q < 10
for a single-CPU machine). %runocc suffered similarly, and the system
became less usable than the version 1 system. Have others experienced the
issue? Are there workarounds in later versions of squid 2.x? Are there
parameters (4 dnsservers were kept constantly busy by our user load, as well
8 redirectors) that ought to be considered and/or reconsidered?



Get your free, private email at http://mail.excite.com/
Received on Tue Apr 13 1999 - 19:22:17 MDT

This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:45:50 MST