Henri thinks we should discuss that here, so I'm forwarding the original mail.
Basically I don't like the time. The alternative is to use something like GetTickCount(). The patch uses the semi-dynamic endpoint cache. One should set the high enough worker.cachesize, and the number of opened channels will drop after worker.cache_timeout. Since IMO there was the general +1 for Jan's patch, here is the original with added endpoint timeout. > -----Original Message----- > From: Mladen Turk [mailto:[EMAIL PROTECTED]] > Sent: 22. lipanj 2002 18:06 > To: GOMEZ Henri > Subject: mod_jk cache and socket timeout > > > Hi Henri, > > Here is the Jan Singer's socket timeout patch with added > cache timeout (since its not in the tree already). Still > don't like the time() difftime, but it'll do. > > Added workers.cache_timeout (defaults to 15 secs). > > With high cachesize, Apache 2.0.40-dev/TC 4.1.5 XP/WIN2K > behaves perfectly. Done the test on 4x1.6GHz PIII Xeon and it flies... > > > Greetings, > MT. >
jk_ajp13.h.patch
Description: Binary data
jk_connect.c.patch
Description: Binary data
jk_connect.h.patch
Description: Binary data
jk_util.c.patch
Description: Binary data
jk_util.h.patch
Description: Binary data
jk_ajp12_worker.c.patch
Description: Binary data
jk_ajp_common.c.patch
Description: Binary data
-- To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]> For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>