Re: cvs commit: httpd-2.0/modules/experimental cache_storage.c cache_util.c mod_cache.c mod_cache.h mod_disk_cache.c mod_mem_cache.c

[Bill Stoddard]

Justin Erenkrantz wrote:
--On Thursday, September 23, 2004 2:55 PM -0400 Bill Stoddard 
<[EMAIL PROTECTED]> wrote:

This little gem causes a regression. Because cache->handle is left NULL,
we never cleanup stale entries in the cache (in cache_save_filter). Once
CacheMaxExpires hits, the stale entry will never be garbage collected
from the cache and replaced by a new entry.  Two ways to fix this come to
mind right off hand (neither are optimal i expect):

Well, I don't know about mod_mem_cache, but I don't believe this affects 
mod_disk_cache.  mod_disk_cache overwrites the old entry if it is told 
to do so.  It's possible that mod_mem_cache should do the same.
Will investigate.
FWIW, your 2 patches are identical.  Not sure if that's what you 
intended or not.  ;-)  -- justin

Ooops, the other patch was to set cache->handle to h.
Bill