httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Glenn <>
Subject Re: mod_mem_cache bad for large/busy files (Was: [PATCH] removesome mutex locks in the worker MPM)
Date Fri, 03 Jan 2003 06:48:47 GMT
On Thu, Jan 02, 2003 at 09:54:58PM -0800, David Burry wrote:
> interesting... so then why did using mod_file_cache to specify caching a
> couple dozen known-most-often-accessed files decrease disk io significantly?
> I'll try the test you mention next time I get a chance.

Out of curiosity, what are your mount options?
Are atime updates enabled?  Try mounting noatime.
  (man mount_ufs)
Are disk quotas enabled?  Try disabling them if they are not needed.
  (man quotaoff)

If you have memory to burn, you can try creating a huge tmpfs partition
  in memory and copying the entire document repository there.
  (man tmpfs; man mount_tmpfs)
  (man mount_cachefs; man cfsadmin)

What sort of resource limits are on the httpd or apache user (under which
the webserver runs)?  (man ulimit)  Think you might be bumping into any
arbitrary limits?  What about on the system level (man sysdef)
  Run 'sysdef' and look for the "Tunable Parameters" section.

Are you using sac (man sac) and what environment is it setting?

(I'm by no means an expert on Solaris, but I hope this helps)

Aha!  Try this link:
Fun with noatime, logging, and forcedirectio options.


View raw message