From: Stefan Fritsch <sf@apache.org>
Date: Sat, 5 May 2012 08:30:39 +0000 (+0000)
Subject: Use short lived pool to avoid memory leaks
X-Git-Tag: 2.5.0-alpha~6895
X-Git-Url: http://git.ipfire.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=f23a12fb761043e69ca318e80e27d55e6b7ba8a1;p=thirdparty%2Fapache%2Fhttpd.git

Use short lived pool to avoid memory leaks

Remove the comment about binding upstream and downstream connections. It
seems to be obsolete since r104604, r104605, r105108.

Also avoid allocating memory if we are not handling the connection.

PR: 52275
Submitted by: Naohiro Ooiwa <naohiro ooiwa miraclelinux com>, Stefan Fritsch


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@1334343 13f79535-47bb-0310-9956-ffa450edef68
---

diff --git a/CHANGES b/CHANGES
index ea45df391fe..0a7839d9b6b 100644
--- a/CHANGES
+++ b/CHANGES
@@ -1,6 +1,9 @@
                                                          -*- coding: utf-8 -*-
 Changes with Apache 2.5.0
 
+  *) mod_proxy_ajp: Reduce memory usage in case of many keep-alive requests on
+     one connection. PR 52275. [Naohiro Ooiwa <naohiro ooiwa miraclelinux com>]
+
   *) mod_proxy: Use the the same hostname for SNI as for the HTTP request when
      forwarding to SSL backends. PR 53134.
      [Michael Weiser <michael weiser.dinsnail.net>, Ruediger Pluem]
diff --git a/modules/proxy/mod_proxy_ajp.c b/modules/proxy/mod_proxy_ajp.c
index 93cb1530988..c9f58171f52 100644
--- a/modules/proxy/mod_proxy_ajp.c
+++ b/modules/proxy/mod_proxy_ajp.c
@@ -701,29 +701,15 @@ static int proxy_ajp_handler(request_rec *r, proxy_worker *worker,
     int retry;
     proxy_dir_conf *dconf = ap_get_module_config(r->per_dir_config,
                                                  &proxy_module);
-
-    /*
-     * Note: Memory pool allocation.
-     * A downstream keepalive connection is always connected to the existence
-     * (or not) of an upstream keepalive connection. If this is not done then
-     * load balancing against multiple backend servers breaks (one backend
-     * server ends up taking 100% of the load), and the risk is run of
-     * downstream keepalive connections being kept open unnecessarily. This
-     * keeps webservers busy and ties up resources.
-     *
-     * As a result, we allocate all sockets out of the upstream connection
-     * pool, and when we want to reuse a socket, we check first whether the
-     * connection ID of the current upstream connection is the same as that
-     * of the connection when the socket was opened.
-     */
-    apr_pool_t *p = r->connection->pool;
-    apr_uri_t *uri = apr_palloc(r->connection->pool, sizeof(*uri));
-
+    apr_pool_t *p = r->pool;
+    apr_uri_t *uri;
 
     if (strncasecmp(url, "ajp:", 4) != 0) {
         ap_log_rerror(APLOG_MARK, APLOG_DEBUG, 0, r, APLOGNO(00894) "declining URL %s", url);
         return DECLINED;
     }
+
+    uri = apr_palloc(p, sizeof(*uri));
     ap_log_rerror(APLOG_MARK, APLOG_DEBUG, 0, r, APLOGNO(00895) "serving URL %s", url);
 
     /* create space for state information */